Urban Pulse by Gemeente Amsterdam, Economie

URBAN PULSE Understanding resource flows and dynamics in Amsterdam

URBAN PULSE - Final report AMS Kick-Start project

URBAN PULSE AMS Kick-Start Project Final Report Version 160129

Table of Contents 1 2 3 4 5

Project Title Authors report Project partners Management summary Extended abstract 5.1. 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9

Introduction Urban Material Flow Analysis Amsterdam Stakeholdersâ&#x20AC;&#x2122; perspective on data needs Spatio-temporal dynamics bottled water Amsterdam Main conclusions Impact and benefits Metropolitan Region Amsterdam Spin-off and valorisation Ideas for follow-up projects from project partners List of references

Publications realized 6.1 6.2 6.3 6.4 6.5

7 8

2! 2 2 3 5 5 7 9 11 14 15 16 17 17

Scientific publications Professional publications Presentations in AMS format Online or social media publications Other dissemination activities

Key data-sets realized by project Annex

19 19 19 20 20

21 22

URBAN PULSE - Final report AMS Kick-Start project

1. Project Title Urban Pulse: Understanding resource flows and dynamics in Amsterdam

2. Authors report Sven Stremke, Marc Spiller, Ilse Voskamp and Corné Vreugdenhil (all WUR) With text contributions by Job Spierings (Waag Society), Titus Tielens (Port of Amsterdam), Jan-Peter van der Hoek (Waternet), Sietse Agema (AEB) and Eveline Jonkhoff (City of Amsterdam)

3. Project partners Wageningen University and Research (WUR, lead) Arnold Bregt, Huub Rijnaarts, Jan Willem van der Schans, Marc Spiller, Sven Stremke, Niels Tomson, Ilse Voskamp, Corné Vreugdenhil

Delft University of Technology (TUD) Jan-Peter van der Hoek, Gijsbert Korevaar, Ben Zhu

City of Amsterdam Eveline Jonkhoff, Bob Mantel

Waag Society (Waag) Tom Demeyer, Ivonne Jansen-Dings, Frank Kresin, Martin Risseeuw

Waternet Jan-Peter van der Hoek, André Struker

AEB Amsterdam (AEB) Sietse Agema, Evert Lichtenbelt

Port of Amsterdam (PoA) Jan Egbertsen, Youri Hildebrand, Titus Tielens

Alliander Marcel van Hest, Christian Klep

URBAN PULSE - Final report AMS Kick-Start project

4. Management summary Globally, urbanizations accommodate more than 50% of the world’s population and are estimated to be responsible for 70% of pollution and resource depletion. The process of resource consumption and pollution can be conceptualized as an urban metabolism. Urban metabolism refers to “the sum total of the technical and socio-economic processes that occur in cities, resulting in growth, production of energy, and elimination of waste”. Until now, the majority of existing studies on urban metabolism investigated urban metabolism at the scale of the city or region using aggregated data to determine overall balances. However, to support policy makers, urban planners and designers in their effort to implement a transition towards a more resources efficient urban metabolism, higher temporal and spatial resolution analyses are required. The aim of the Urban Pulse project was to assist in the transition towards a circular urban metabolism by advancing the understanding of spatial and temporal dynamics of resources flows through the city of Amsterdam. In other words, the aim was to understand the Urban Pulse of Amsterdam. The two main objectives were: I. II.

Gather data and identify data needs, gaps and availability First spatio-temporal evaluation of selected resource flows in Amsterdam

Methodology The Urban Pulse project adopted a transdisciplinary research approach, integrating, among others, the fields of engineering, urban planning, landscape architecture, big data, geo information as well as academia and practice. The analysis of Amsterdam’s metabolism - at the scale of the municipality - was structured around four flows: energy, water, food and materials. A variety of scientifically sound methods was used and adapted in this project, including: material flow analysis (Urban-EUROSTAT), temporal drinking water demand modelling (SIMDEUM: SIMulation of water Demand, and End-Use Model), spatial analysis (GIS: Geographical information systems), data demand and availability mapping. A large number of data sets for flows in the Amsterdam municipal region were gathered and processed. A metadata standard was developed for these data sets. Access to the available data will be made possible via the AMS data platform. Results Project results show that Amsterdam has an urban metabolism that is characteristic for port cities, where throughput flows make up a large fraction of overall flows. Specific to Amsterdam is the fact that 77% of the fuels imported are passing through the city, indicating its position as major fuel harbour in the world. Furthermore, the project gives detailed accounts of water, biomass and energy flows that sustain the city, how they enter and leave the city. It shows that water dominates all flows when measured in mass. The study further reveals that renewable energy generated locally accounts for about 7% of the electricity and gas consumed in Amsterdam. Food and feed related phosphorus flows in Amsterdam account for about 750 tons/year, of which the majority ends up in the wastewater. There are current initiatives to recover phosphorus from 3

URBAN PULSE - Final report AMS Kick-Start project

wastewater (Fosvaatje) and the company ICL Fertilizers, located in the harbour of Amsterdam, is aiming to replace 25.000 ton of its mined phosphorus input with alternative phosphorus sources by 2025. Good sources to recover phosphorus are meat and fish residuals as well as coffee ground. In this project the information needs and data availability for the implementation of circular urban metabolism interventions was assessed. Results indicate that there is limited high-temporal resolution data available on urban water and energy flows. Yet the views of key stakeholders in Amsterdam indicate that information of high and very high temporal (and spatial) resolution is needed to inform the planning and design of resource-conscious interventions. However, it was also emphasised that â&#x20AC;&#x153;a lack of dataâ&#x20AC;? is by no means the only barrier that needs to be overcome to realize a circular urban metabolism. Furthermore the sales of water bottles were analysed for Amsterdam as an example to showcase temporal and spatial dynamics of a resource flow (i.e. bottled drinking water). To do this, a detailed data set on water bottle sales per week and post code area was obtained. The analysis revealed that a total of 21.7 million litres of bottled water was sold in Amsterdam in 2014. A clear seasonal pattern is found for the amount of bottled water sold; more water is sold in the summer period and less in the winter. The spatial distribution of the amount of bottled water sold showed that two areas have relatively high values: the northern part of Amsterdam-Zuidoost (postcodes 1102-1104) and the historical centre of the city (postcodes 1011-1012). In the city centre, comparatively small bottles are sold. This shows that many people come to the centre of the city, to work and recreate there and buy their water for the day. Conclusion and implication for management In conclusion, the material flow analysis of Amsterdam municipality conducted in the Urban Pulse project presents a good baseline for the City of Amsterdamâ&#x20AC;&#x2122;s department of Urban Planning and Sustainability and others interested to understand the metabolism of Amsterdam as a whole. It is recommended that the study is updated in regular intervals to evaluate progress against the sustainability ambitions of the City. However, it can also be concluded that there is a real need for data of higher temporal and spatial resolution. The project shows that some of this data (e.g. from Waternet) can be made available when the right partners are involved in a project that capitalises on joint interests and builds trust. The project also revealed that alternative data sources with high spatio-temporal resolution data, such as commercial data are available too (e.g. IRI data). It is therefore recommended to the AMS institute to continue and expand the research cooperation between academia, industry and government. This can ensure access to unique data, creation of new information and meaningful insights for the transition to a circular urban metabolism.

URBAN PULSE - Final report AMS Kick-Start project

5. Extended abstract 5.1 Introduction Globally, urbanized areas accommodate more than 50% of the world’s population and are estimated to be responsible for 70% of pollution and resource depletion (Rees and Wackernagel 2008). The conversion of resources can be conceptualized as urban metabolism. Urban metabolism refers to “the sum total of the technical and socio-economic processes that occur in cities, resulting in growth, production of energy, and elimination of waste” (Kennedy et al. 2007). Currently, the urban metabolism of Amsterdam and other cities is linear. In this linear metabolism, resources are used once and then discharged to the environment (Figure 1). This linear consumption pattern is at the heart of the present resources exploitation, saturation of ‘waste’ sinks and consequently the imbalance of (global) ecosystems. A transition towards a circular urban metabolism will: • • • •

Increase resource efficiency, Eliminate waste by using it as a resource for generation of new valuable products, Support resilience of urban systems, and Be crucial for achieving environmental sustainability (Girardet 1999).

Figure 1: Conceptualisation of linear and circular urban metabolism (Lehman, 2011).

Up until now, the majority of existing studies on urban metabolism investigate urban metabolism at the scale of the city or region. Aggregated data are used to quantify resource flows and to disclose how a city functions at a specific moment in time (Kennedy, 2007; Ngo and Pataki 2008). Although such urban metabolism knowledge can be useful for urban planning, design and decisionmaking, it has hardly been applied in urban planning and design, as pointed out by Kennedy et al. (2011). One reason for this lack of application is that metabolic studies at the regional scale do not match the level at which practical urban

URBAN PULSE - Final report AMS Kick-Start project

planners and designers operates; namely the district scale, neighbourhood or building block (Spiller & Agudelo 2011). However, such fine scale accounting analyses on city and more detailed scale levels- is said to be hampered by a lack of data (Pincetl et al. 2012; Shahrokni et al. 2015). The question is whether this is also the case in Amsterdam. Is the data of urban resource flows such as water, food, energy and waste that is needed for fine scale urban metabolism analyses available and accessible to relevant stakeholders in Amsterdam? Can these data provide insight in the dynamics of the city’s resource flows and accordingly be of support for Amsterdam’s transition towards a circular metabolism – often referred to as circular economy? The aim of the Urban Pulse project is to assist in the transition towards a circular urban metabolism by advancing the understanding of spatial and temporal dynamics of resources flows in Amsterdam. In other words, the aim is to understand the Urban Pulse of Amsterdam. The two main objectives are: I. II.

Gather data and identify data needs, gaps and availability First spatio-temporal evaluation selected resource flows in Amsterdam

It is important to stress that the Urban Pulse kick-start project –reported in this document - focused on the first objective and contributed to the second objective. The second objective (and other emerging knowledge gaps) will be addressed more thoroughly in future Urban Pulse projects for which new proposal are being formulated. The above-mentioned objectives were pursued for a selection of urban flows, namely energy, water, food and organic/inorganic materials in the municipality of Amsterdam. Representatives from local utilities, businesses and the City of Amsterdam identified those flows as the most important ones during a sequence of workshops that took place prior to the start of the Urban Pulse project. The actual time line of the project is shown in figure 2 below.

Figure 2: Urban Pulse project timeline with key milestones.

URBAN PULSE - Final report AMS Kick-Start project

In the following sections, selected findings are presented: Material flow analysis of Amsterdam (section 5.2), stakeholders’ perspective on data needs (section 5.3), and spatio-temporal dynamics of bottled water (section 5.4). Information about additional results and data can be found in the appendix of this report. 5.2 Material Flow Analysis for Comprehensive Assessment of the Urban Metabolism: A Case Study of Amsterdam1 Keywords: urban metabolism, material flow analysis (MFA), resource management, urban planning, circular economy Introduction Ongoing urbanization, resource depletion and climate change emphasize the need to design and plan cities that foster sustainable urban resource management (Agudelo-Vera et al. 2011). In order to plan cities that generate less environmental pressure it is essential to understand how urban systems function with respect to resource flows (Decker et al. 2000). Urban metabolism (UM) studies have become a key approach to researching the material flows and stocks of urban systems (Kennedy et al. 2011; Zhang 2013; Castán Broto et al. 2012). One established method for analysing the UM is the Eurostat material flow analysis (MFA). The majority of comprehensive European urban MFA studies are based on this Eurostat method (Barles 2009; Browne et al. 2011; Hammer & Giljum 2006; Niza et al. 2009; Rosado et al. 2014). However, for a comprehensive assessment of the UM this method has its limitations. It does not account for all relevant resource flows, such as locally sourced resources and it does not differentiate between flows that are associated to the city’s resource consumption and resources that only pass through the city. This research sought to gain insights into the UM of Amsterdam by performing an MFA employing the Eurostat method (Eurostat 2001; Hammer et al. 2003). Materials and methodology Modifications to the Eurostat method were made to enhance its performance for comprehensive UM analyses. To identify potentially useful adjustments and additions to the original Eurostat method, we first reviewed the five exclusively Eurostat-based European studies (Barles 2009; Browne et al. 2011; Hammer & Giljum 2006; Niza et al. 2009; Rosado et al. 2014), which examined six different cities. We then formulated a resource classification for the modified Eurostat method by modifying the original Eurostat resource classification in the light of insights obtained from the literature review. To assess the proposed modifications we organized stakeholder workshops with selected academic, societal, and industry partners. The stakeholders’ commitment and workshops played a crucial role in identifying and obtaining the necessary datasets for this MFA. Almost all the resource flows could be quantified using reliable datasets that contained local data on Amsterdam for 2012. The results of the Eurostat and the modified Eurostat method were compared to evaluate the additional information value of the modifications. To facilitate interpretation of the 1

Full paper accepted and under final review at the Journal of Industrial Ecology 7

URBAN PULSE - Final report AMS Kick-Start project

Amsterdam findings, the results of the Amsterdam MFA were benchmarked with Hamburg and Vienna (Hammer & Giljum 2006) and results were compared to a previous study of Amsterdam’s UM (Gorree et al. 2000). Main Findings The results show that Amsterdam’s UM is dominated by water flows and by portrelated throughput of fossil fuels. Water imports amount to a total of 81,354 kt (see figure 3), the same order of magnitude as all other imports combined. This confirms the finding from the 1998 Amsterdam MFA (Gorree et al. 2000) that water flows make up a significant share of Amsterdam’s UM. Considering fossil fuels, 77% (37 million tonnes) of the 48 million tonnes of fossil fuels imported can be considered flows passing through the city. Furthermore, the study shows that in 2012 the equivalent of 24 kt biomass, 83 kt (3,418 TJ) renewable energy, and 617 kt secondary resources, including 214 kt (8,980 TJ) energy, were sourced locally. These are relatively small flows, because the large volumes of (portrelated) imports and exports dominate Amsterdam’s overall metabolism. The comparison of Amsterdam’s resource flows with those of Hamburg and Vienna shows that the dominance of fossil fuel throughput is specific to the UM of Amsterdam. The benchmark study also revealed that in comparison with Vienna and Hamburg, the total amount of renewable energy generated in Amsterdam is relatively low. However, normalized per inhabitant Amsterdam generates an equal amount as Vienna and two-thirds of the renewable energy/capita generated in Hamburg.

Figure 3: Overview of water flows that are part of the Amsterdam Urban Metabolism.

Discussion Although the city’s environmental pressure has decreased since 1998 in terms of per capita waste generation and drinking water consumption, the majority of the city’s municipal waste is still not recycled, but incinerated. The modified Eurostat method provides a more detailed understanding of the UM than the urban Eurostat MFA. This is because the modified resource typology achieves a more detailed resource classification and it has a higher level of “completeness” of the flows accounted for. Moreover, accounting for and detailing throughput flows yields improved insights into the nature of a city’s imports, exports, and stock. 8

URBAN PULSE - Final report AMS Kick-Start project

5.3 Stakeholdersâ&#x20AC;&#x2122; perspective on data needed for resource-conscious planning & design interventions2 Keywords: urban metabolism, urban planning, urban design, spatio-temporal resolution, scale Introduction Urban metabolism (UM) has become a key approach to analyse resource flows in cities and to develop sustainable urban resource management concepts (Kennedy et al. 2011; CastĂĄn Broto et al. 2012; Zhang 2013). Although it is argued that UM analyses can be useful for urban planning, design and decision-making (CastĂĄn Broto et al. 2012; Chrysoulakis et al. 2013; Kennedy et al. 2011), application of this knowledge is limited to this moment (Kennedy et al. 2011) with very few exceptions (Chrysoulakis et al. 2013 ). Therefore, the aim of this study is to research what information on urban water and energy flows is needed for planning and design decision-making regarding resource-conscious interventions and on which level of detail in space and time. Because fine scale accounting -UM analyses on city and more detailed scale levels- is said to be hampered by a lack of data (Pincetl et al. 2012; Shahrokni et al. 2015; Codoban & Kennedy 2008), the second aim of this paper is to assess the gap between what data is known to be available and what data is needed by decision makers. Materials and methodology A case study was performed to find out which information on resource flows is needed and available for informing resource-conscious interventions. In workshops and face-to-face interviews, stakeholders were invited to express the information need for resource-conscious planning and design interventions in terms of its spatio-temporal resolution. Stakeholders involved were researchers, environmental managers from public utilities, landscape architects as well as urban planning and design practitioners involved in the management and planning of resource flows in the city of Amsterdam. The method by Vervoort et al. (2014) was adapted for the workshops and interviews. In phase 1, participants were provided with an A4 sheet of paper on which they could describe a resourceconscious intervention that they envision implementing in Amsterdam. Stakeholders were also asked to specify in a graph at which spatial level this intervention takes places and which time frame they envision for implementation. The time-scales of the graph ranged from seconds to decades and spatial scale levels ranged from individual building to The Netherlands. In phase 2, the participants were given a new sheet of A3 paper and were asked to specify the information needed for (a selection of) the interventions mentioned in phase 1, in terms of its spatial and temporal resolution. A desktop study was performed to compose an overview of the data available on energy and water flows in Amsterdam, including the spatial-temporal resolution of this data. Subsequently this overview was used to assess whether the information need expressed in the workshops can be met by the data available (see figure 4).

Full paper on the results of this part of the project is in preparation 9

URBAN PULSE - Final report AMS Kick-Start project

Information need 1: regional electricity demand &supply; 2: local electricity demand; 3: electricity supply per supplier Data availability 4: residential electricity demand; 5: electricity supply of one potential supplier

Figure 4: Information need and data availability for intervention “Super Smart Grid”.

Main Findings A total of 52 different interventions were identified in the workshops and interviews, including 27 spatial or technological interventions. For 16 of these interventions information needs were elicited. For 50% of the interventions (8 out of 16), stakeholders expressed the need for information that has both a very high spatial resolution (building – street level) and a very high temporal resolution (seconds – 12 hours). Information with a (very) high spatial resolution appears to be somewhat more important than information with a (very) high temporal resolution. For 15 interventions the need for information on building – district level (high - very high spatial resolution) is expressed, whereas the need for information on the seconds to weekly level (high - very high temporal resolution) is expressed for 12 interventions. Yet, out of the 13 databases accessible in Amsterdam, only one provides data with a very high temporal resolution (the water flow rate per minute at drinking water pumping stations) and two have data with a medium resolution (energy generated at the city’s waste-to-energy plant per month). The other 10 databases have a low temporal resolution (between 1 and 5 years). Regarding the spatial resolution of these datasets, 11 datasets have a very high or high spatial resolution and just two dataset provide data only as detailed as the municipal scale. Discussion There appears to be a gap between the data that is available on urban water and energy flows in Amsterdam and the information needed by decision makers to inform the planning and design of resource-conscious interventions they envision implementing in the city. In particular, there is a data gap regarding 10

URBAN PULSE - Final report AMS Kick-Start project

data that has a more detailed temporal resolution than yearly. Nevertheless, stakeholders indicated that they perceive other barriers then data availability as equally or more important to realize implementation of resource-conscious interventions, like costs and external collaboration constraints. 5.4 Exploring the spatio-temporal dynamics of bottled water purchases in Amsterdam (IRI data) Keywords: Bottled water, Amsterdam, spatial and temporal dynamics Introduction The Urban Pulse project is aiming to get more insights into the dynamics of urban resource flows in the city of Amsterdam, by contributing to the spatiotemporal evaluation of selected resource flows in Amsterdam. Therefore, a study of bottled water purchases in Amsterdam was performed. With this study the usefulness, possibilities and limitations of a high spatio-temporal resolution resource flow analysis were explored, using a commercial dataset on bottled mineral water sales in Amsterdam. Materials Urban Pulse obtained a dataset from IRI, a company that collects information on items that are sold by supermarkets and retailers3. The obtained dataset from IRI describes all bottled mineral water that is sold by retailers and supermarkets (except Lidl and Aldi) in the municipality of Amsterdam for the year 2014. The dataset describes sales on a weekly basis for the year 2014 and per 4-digit postcode group. The IRI data included three variables: the sales in volumes (litres), units (-) and euros (â&#x201A;Ź). Due to privacy rules, some 4-digit postcodes were aggregated to form a postcode group. Methodology Iterative exploratory analysis The exploration of the IRI data was done iteratively. One project member started with performing a first analysis on the dataset and reported this to the other project members. Project members gave feedback and made suggestions for the next round of exploratory analysis. In the end, a total of four rounds of analysis were performed this way. Spatial and temporal dimension The dataset was explored by taking different combinations of dimensions into account. Analyses were done with and without spatial and/or temporal variation. In such a way, analysis were performed on (1) year total data for Amsterdam, (2) weekly data for Amsterdam, (3) year data for Amsterdam per postcode and (4) weekly data for Amsterdam per postcode. Variables taken into account From the variables sales in volumes (litres), units (-) and euros (â&#x201A;Ź), three extra variables were derived: the average volume of a sale unit (litres/unit), the average 3

https://www.iriworldwide.com/nl-NL 11

URBAN PULSE - Final report AMS Kick-Start project

price per sale unit (â&#x201A;Ź/unit) and the average price per volume of bottled water (â&#x201A;Ź/litre). Main Findings Year total data In 2014, a total of 21.7 million litres of bottled water was sold in Amsterdam, which resulted in a total 1.9 million euro of tax revenue. The average volume and price of a sale unit was respectively 1.45 litres and 0.88 euro, giving an average price per volume of 0.61 â&#x201A;Ź/litre. Weekly data for Amsterdam A clear seasonal pattern is found for the amount of bottled water sold: more water is sold in the summer period and less in the winter. The average price per unit and volume, as well as the average volume per unit seem to be constant over the year. This indicates that only the amount of water sold is changing over the year and not the price of water (see figure 5). The bottled water consumption in Amsterdam is in winter period 39, in summer 55 and during heat waves 78 truckloads/week, assuming truckloads of 9 tonnes. Time-trend analysis with meteorological variables showed that the average temperature is highly correlated with the drinking water sales (R2-values > 0.80). It appeared that the sale of non-carbonated water is more related to the temperature than carbonated water. Carbonated water showed a less strong seasonal pattern than the non-carbonated water. So, the average temperature is an influencing variable and can be used to predict the bottled water sales with a linear or exponential function (R2 values of respectively 0.82 and 0.87). No clear relation has been found between bottled water sales and events in the city centre. Sales could also not directly be related to the number of tourists that stay overnight in Amsterdam. Year data for Amsterdam per postcode The spatial distribution of the amount of bottled water sold, showed relatively high values in two areas: (1) the centre of the city (postcodes 1011-1012) and (2) the northern part of Amsterdam-Zuidoost (postcodes 1102-1104). For the centre of the city this could be explained by the fact that this area includes the central train station and many touristic places. As a result, tourist who come to Amsterdam are mainly concentrated in this area. Also, many people come here on their way to/from work and possibly buy their water for the day in this area. The spatial distribution shows that the further away from the centre of the city, the larger the sale units were. Linked to this, the centre of the city has the highest price per volume of water (more than 1 euro/litre). The further away from the centre, the lower the price per volume (less than 50 cents/litre) (see figure 5).

URBAN PULSE - Final report AMS Kick-Start project

Figure 5: Average price per volume (â&#x201A;Ź/litre) of bottled water, in time per week and in space per district.

Weekly data for Amsterdam per postcode It is clear that all districts in Amsterdam show roughly the same seasonal pattern for the amount of bottled water sold and almost all districts show peaks at the same moment in time. It is striking that all seem to have a peak around week 29. This indicates that there was something during that week that was causing high water consumption in the whole of the city. The extreme weather conditions that week, a heat wave, are a plausible explanation. Weather conditions like temperature are roughly equal within the city, which means that it influences the drinking water sales in the entire city. In general it can be stated that there are no big changes in time per district for the three derived variables (like for example the average price per volume). In the centre of the city the sale units or bottles are the smallest and sold for the highest price compared to the rest of the city. Discussion The IRI data did not include the Lidl and Aldi supermarkets, accounting for a total of 23 supermarkets in Amsterdam (in 2014). Since these supermarkets are relatively inexpensive compared to the other included chains, this incompleteness in the dataset results most probably in an overestimation of the average price per volume. This incompleteness also means that findings in the spatial distribution of the bottled water consumption could be wrong. However, the findings based on temporal dynamics are not directly affected by the incompleteness of the dataset. Although the findings of this study are not complete, this study provides a good first insight in the dynamics of bottled water purchase in Amsterdam and can serve as example for the use of big data in the light of urban metabolism research. 13

URBAN PULSE - Final report AMS Kick-Start project

5.5 Main conclusions In conclusion, the material flow analysis (MFA) of Amsterdam municipality carried out in the Urban Pulse project, is a good baseline for the City of Amsterdam’s department of Urban Planning and Sustainability and others interested to understand the metabolism of Amsterdam as a whole. Results show that Amsterdam has the typical metabolic profile of a harbour city. Its unique feature is a dominant throughput flow of fossil fuels. The modified Eurostat method used to assess the city’s metabolism provides a richer understanding of the urban metabolism, compared with the already existing urban Eurostat MFA. This is due to: • • •

A higher level of “completeness” of the MFA; A more detailed resource classification; Accounting and detailing of throughput flows.

The Urban Pulse project resulted in a new data gap analysis tool for the study of resource-conscious interventions. Using this tool, it was shown that there is a real need for data of higher temporal and spatial resolution than currently available. The tool in itself is an excellent means for actors concerned with urban metabolism to study their data needs versus data availability. It was found that some of the needed data simply not exist yet. In addition, much of the existing data is not accessible or only with restricted access. However, the Urban Pulse project showed that some of this data can be made available when the right partners are involved in a project that capitalises on joint interests and builds trust. Example data collected in this project were individual household water demand data (Waternet) and detailed data on the material flows into the city (Port of Amsterdam). The project also showed that commercial data sets are an option to gather valuable information (e.g. IRI data discussed in section 5.4 above). It is therefore recommended to the AMS institute to continue and expand the research cooperation between academia, industry and government. This can ensure access to unique data, creation of new information and meaningful insights for the transition to a circular urban metabolism. It is further recommended to repeat urban metabolism studies in regular intervals using the advanced methodological framework for MFA developed in the Urban Pulse project. This will enable the municipality to monitor progress regarding its sustainability ambitions. Finally, follow-up projects should build on Urban Pulse to generate more insight into the spatial and temporal dynamics of resource flows in Amsterdam and elsewhere.

URBAN PULSE - Final report AMS Kick-Start project

5.6 Impact and benefits for the Metropolitan Region Amsterdam For this section as well as section 5.7 and 5.8, the societal and industry partners of the Urban Pulse project took the lead and provided input, along with reflections upon the process and findings of the project. - Urban Pulse has generated a better insight into the 'urban pulse,' i.e., the rhythm and flow of materials, people and energy within the metropolitan region of Amsterdam. This insight can ideally (theoretically) be used to identify inefficiencies - i.e., incidences where streams are burdening the local area, are needlessly complicated, or where waste is generated and not used appropriately. Understanding the inefficiencies can help to define opportunities for improvement, i.e. creating material flow shortcuts, or identifying waste streams that can be turned to a better use. The current data set has helped to identify inefficiencies/opportunities. (Port of Amsterdam) - The outcome of Urban Pulse helps to find out impact and opportunities for future technologies to develop in the Amsterdam region. Therefore, it gives an insight in the transition path necessary and economic impact of these transitions can be assumed and calculated. (AEB) - The plastic bottle case is a good example of resource inefficiencies and shows that this is not solved with one technology only but the independencies between several stakeholders in this case: water (and investments in refill infrastructure), prevention of waste, sales volume of retailers and public perception. So it shows the necessity of an integrated approach. (AEB) - It's good that we can benchmark it [Amsterdams metabolism] on European level by using the Eurostat method and data, this makes it [the MFA] better used for impact calculations. This was a good start, we need to make sure that we keep using the same method and model when going into more detail per specific stream. (AEB) - The greatest impact for Waternet had the material flow analysis in which drinking water and wastewater were incorporated. This resulted in a detailed overview of materials in the water flows in Amsterdam, which will help Waternet to take decisions in which direction to go with respect to resource recovery initiatives. (Waternet) - Comprehensive Knowledge Archive Network (CKAN) and open data platforms in general are still mainly used for sharing open data from government(al) sources. The possibilities for sharing research/scientific data via CKAN platforms was discussed and researched in conjunction with the CitySDK Linked Data platform. This provided important input for the AMS data platform, which would be the first data platform combining civic, governmental & research data. (Waag)

- The city of Amsterdam has strong ambitions regarding the transition towards a circular economy; Amsterdam wants to be a frontrunner. In this context, it is important to get an in depth insight into the circular economy: quantitatively on different levels, impact on economy, ecology and society. 15

URBAN PULSE - Final report AMS Kick-Start project

The Urban Pulse kick-start project provided insights into resource flows through the city and spatial their impact. (City of Amsterdam) - The Urban Pulse collection of data on resource flows was used by the "Amsterdam Circular" initiative. (City of Amsterdam) - Cooperation between researchers and sharing of data helped to realize a unique report "Amsterdam Circular: Vision and roadmap for the city and region". (City of Amsterdam) 5.7 Spin-off and valorisation4 - Together with AEB, Waternet is involved in the Power to Protein project: production of single cell protein from wastewater. The Urban Pulse project was a driver for Waternet to join this project. (Waternet) - A PhD grant for Urban Metabolism & design research has been awarded to Roberta Pistoni (Versailles/France) who will be co-supervised by UP researcher Sven Stremke, Wageningen University. - The AMS data platform will build upon the results of Urban Pulse. (Waag)

- On the basis of Urban Pulse data, a new cooperation between AMS and Circle Economy was established, about standardization of data and indicators. (City of Amsterdam) - A combined initiative started in order to develop a so-called "circular dashboard" on the level of the city (City of Amsterdam, AMS, TNO, Circle Economy, and Metabolic). It focuses on the development of indicators and preferably real-time monitoring of circularity, on the level of the city to illustrate the transition towards a circular economy. (City of Amsterdam) - UP stimulated and informed Dr. Daniela Perrotti/INRA Paris to create a proposal for a MARIE CURIE proposal (H2020-MSCA-IF-2015) together with UP researcher Sven Stremke. The goal of the action entitled ADAGIUM is to develop advanced knowledge for climate change mitigation and adaptation in temperate-climate cities by means of green infrastructure. The decision on the proposal is due to in February 2016. - A member of the consortium (Marc Spiller) contributed to the successful bid for VANG-lokaal (initiated by Ministerie van IenM, Rijkswaterstaat, de VNG en de NVRD) together with Royal Haskoning DHV, Cirkellab, Cirkelstad and Wageningen UR, with the title: Circleregio â&#x20AC;&#x201C; Serious gaming in de transitie naar en Regionale Circulire Economie. The focus of this project is reducing solid waste production and closing resource loops from household waste. A tool will be produced that has a gaming character; government organisations are target groups for the game.

For this section, the societal and industry partners of the Urban Pulse project provided input. 16

URBAN PULSE - Final report AMS Kick-Start project

5.8 Ideas for follow-up projects from the project partners5 - Next steps could include a follow-up project zooming in specific inefficiencies/ opportunities to further explore solutions, map out stakeholders, [identify] levers for change, and any action program to realize the improvement potential. (PoA) - Make a connection between the data sets that have been produced and concrete applications/business ideas. (PoA) - Perhaps making the data available to the wider public can spark ideas or interest by entrepreneurs, with new ideas for business opportunities in the circular economy. (PoA) - Identification per stream what aggregate level data need to be collected was part of the program; this has to find its way in follow up research programs. (AEB) - Expand to a broader region like the metropolitan area of Amsterdam. (AEB) - Identify research programs in order to collect data on specific streams at the right aggregate level. And make economic impact visible on City national and European level. And keep this programs connected via Urban Pulse. (AEB) - We need to understand how to measure circularity. What are valuable indicators? What are the total, systemic effects of certain interventions? (City of Amsterdam)

5.9 List of references Agudelo-Vera, C.M. et al., 2011. Resource management as a key factor for sustainable urban planning. Journal of environmental management, 92(10), pp.2295–303. Available at: http://www.ncbi.nlm.nih.gov/pubmed/21641714 Barles, S., 2009. Urban Metabolism of Paris and Its Region. Journal of Industrial Ecology, 13(6), pp.898–913. Browne, D., O&apos;Regan, B. & Moles, R., 2011. Material flow accounting in an Irish city-region 1992-2002. Journal of Cleaner Production, 19(9-10), pp.967–976. Castán Broto, V., Allen, A. & Rapoport, E., 2012. Interdisciplinary Perspectives on Urban Metabolism. Journal of Industrial Ecology, 16(6), pp.851–861. Chrysoulakis, N. et al., 2013. Sustainable urban metabolism as a link between bio-physical sciences and urban planning: The BRIDGE project. Landscape and Urban Planning, 112, pp.100–117. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0169204612003295 Codoban, N. & Kennedy, C. a., 2008. Metabolism of Neighborhoods. Journal of Urban Planning and Development, 134(1), pp.21–31. Decker, E.H. et al., 2000. Energy and Material Flow Through the Urban Ecosystem. Annual Review of Energy and the Environment, 25, pp.685– 740. Eurostat, 2001. Economy-wide material flow accounts and derived indicators. A Methodological guide, Luxembourg. 5

For this section, the societal and industry partners of the Urban Pulse project provided input. 17

URBAN PULSE - Final report AMS Kick-Start project

Gorree, M., Kleijn, R. & Voet, E. Van Der, 2000. Materiaalstromen door Amsterdam [Material flows through Amsterdam], Amsterdam: Milieudienst Amsterdam, Centrum Milieukunde Leiden. Hammer, M. et al., 2003. Material Flow Analysis on the Regional Level: Questions, Problems, Solutions., NEDS Working Papers #2 (04/2003). Hamburg. Hammer, M. & Giljum, S., 2006. Materialflussanalysen der Regionen Hamburg, Wien und Leipzig [Material Flow Analysis of the regions of Hamburg, Vienna and Leipzig], NEDS Working Papers #6 (08/2006). Hamburg. Kennedy, C., Pincetl, S. & Bunje, P., 2011. The study of urban metabolism and its applications to urban planning and design. Environmental pollution, 159(89), pp.1965–73. Niza, S., Rosado, L. & Ferrão, P., 2009. Urban Metabolism. Journal of Industrial Ecology, 13(3), pp.384–405. Pincetl, S., Bunje, P. & Holmes, T., 2012. An expanded urban metabolism method: Toward a systems approach for assessing urban energy processes and causes. Landscape and Urban Planning, 107(3), pp.193–202. Available at: http://dx.doi.org/10.1016/j.landurbplan.2012.06.006. Rosado, L., Niza, S. & Ferrão, P., 2014. A Material Flow Accounting Case Study of the Lisbon Metropolitan Area using the Urban Metabolism Analyst Model. Journal of Industrial Ecology, 18(1), pp.84–101. Shahrokni, H., Lazarevic, D. & Brandt, N., 2015. Smart Urban Metabolism: Towards a Real-Time Understanding of the Energy and Material Flows of a City and Its Citizens. Journal of Urban Technology, 22(1), pp.65–86. Available at: http://www.tandfonline.com/doi/full/10.1080/10630732.2014.954899. Vervoort, J.M. et al., 2014. Visualizing Stakeholder Perspectives for Reflection and Dialogue on Scale Dynamics in Social–Ecological Systems. Human Ecology Review, 20(2), pp.302–348. Zhang, Y., 2013. Urban metabolism: a review of research methodologies. Environmental pollution, 178, pp.463–73.

URBAN PULSE - Final report AMS Kick-Start project

6. Publications realized by project members 6.1 Scientific publications Van der Hoek, Struker & de Danschutter (2015) Amsterdam as a sustainable European metropolis: Integration of water, energy and material flows, Urban Water Journal, DOI: 10.1080/1573062X.2015.1076858 Voskamp, Stremke, Spiller, Perrotti, Van der Hoek and Rijnaarts; Material Flow Analysis for a comprehensive assessment of the Urban Metabolism of Amsterdam. Journal of Industrial Ecology (accepted for publication) Perrotti and Stremke; Can Urban Metabolism Knowledge leverage Green Infrastructure Design in the context of Climate Change? Landscape and Urban Planning (in review) Voskamp, Spiller, Stremke, Vreugdenhil, Bregt and Rijnaarts; A stakeholders perspective on the information needed for resource-conscious planning and design interventions. Journal of Resource Conservation and Recycling (in preparation) Vreugdenhil, Voskamp, Stremke, Spiller and Bregt; Dealing with privacy sensitive data within circular economy analysis - Guidelines for data providers and researchers (working title). International Journal of Digital Earth / International Journal of Applied Earth Observation and Geoinformation / Journal of Spatial Information Science (paper in preparation, journal to be selected)

6.2 Professional publications Hempen, F. & S. Stremke. 2015. Hartslagmeting Amsterdam. Generation E: AEB journal, December 2015, p.17-19. Spiller, 2015. The new urban metabolism (Le Nouveau Metabolisme Urbain) Le1 hebdo No 54 Stremke, S. 2015. Nieuwe IdeeĂŤn voor de Stofwisseling van de Stad. In: ESG Jaarboek 2014, 52â&#x20AC;&#x201C;55. 76: WUR ESG. http://edepot.wur.nl/335286. Zoelen, B. V., Holtslag, B. & S. Stremke. 2014. De stad as levend laboratorium. Het Parool, p.5. Voskamp, I.M. and S. Stremke, 2014. The Pulse of the City: Exploring Urban Metabolism in Amsterdam http://www.toposonline.nl/2014/the-pulse-of-the-cityexploring-urban-metabolism-in-amsterdam/

6.3 Presentations in AMS format Voskamp, I.M., S. Stremke, M. Spiller, H.H.M. Rijnaarts, A. van den Brink (2015), The Urban Metabolism of Amsterdam: A comprehensive Material Flow Analysis, SASBE conference Pretoria, 14 th of December 2015, South Africa. van der Schans J.W. (2015), Urban Food Systems, Keynote at Paris FoodFutures, AMS workshop at COP21 Paris, 5 th of December 2015

URBAN PULSE - Final report AMS Kick-Start project

Korevaar, G. (2015), Urban Metabolism, Multi-actor decision-making in urban-rural interaction – Plenary at Rohevik Green Growth Forum – Tartu (Estonia) – 24th of September 2015 Stremke, S., D. Oudes, I.M. Voskamp (2015), Carbon heroes: Exploring higher-density energy landscapes, International Conference on Energy Landscapes, Dresden, 16th of September 2015 van der Schans J.W. (2015), Sustainable Urban Food Supply, Let’s talk solutions, AMS lecture series, 11th of June 2015 Voskamp, I.M. (2015), Unraveling the dynamics of Amsterdam’s metabolism, Urban Metabolism mini-symposium, 20th of May 2015 at AMS Spiller, M. (2014), Understanding dynamics of urban resource flows, Invited speech at the North Holland water innovation conference.

6.4 Online or social media publications Urban Pulse video created by WAAG Society: https://youtu.be/s2fbTFic1no Spiller M. and Brecht A. (2015) WUR TV - Tegenlicht meet up – “ Urban Challenges” on the 26th of March https://wurtv.wur.nl/p2gplayer/Player.aspx?id=d8prwY Spiller, M. & S. Stremke, 2014, URBAN PULSE abstract for AMS newsletter and AMS website, September 2014, http://www.ams-institute.org/solution/urban-pulse/

6.5 Other dissemination activities Quick Circle Scan Amsterdam Metropolitan Region 2015, Sven Stremke and Marc Spiller were members of the advisory expert group to the project consortium Interview of Ilse Voskamp on the 27th of November 2015, regarding the data use and collection in Urban Pulse - for a publication on research initiatives in municipalities on big and open data, by KING (Kwaliteits Instituut Nederlandse Gemeenten). Organization of Urban Metabolism mini-symposium on the 20th of May 2015 at AMS with about 50 delegates Participation Sven Stremke in AMS workshop ‘energy in the city’ on the 12th of February 2015 and various other AMS workshops between Nov. 2014 and Dec. 2015 Participation Ilse Voskamp in UvA Masterstudio Urban Planning ‘The Metabolic City’ 12-17th of January 2015 Presentation of Urban Pulse students Tom van Heeswijk and Changsoon Choi to U.S. ambassador visiting WUR on the 8th of December 2014 (Master thesis project) Presentation Urban Pulse proposal at the AMS launch conference on the 20th of June 2014 in Amsterdam

URBAN PULSE - Final report AMS Kick-Start project

7. Key data-sets realized by project Table 1 provides an overview of all relevant datasets that were collected for the Urban Pulse project. The table describes for each dataset to which ‘UP-flow’ it is related, what it is about and whether there are restrictions on the usage of it. Some datasets are not related to one of the UP-flows, but were used to make other non-spatial datasets spatial and for general use. These datasets are indicated in the table as ‘general’. Vreugdenhil and Bregt are both part of the AMS Data Platform and will take care that these datasets will be included in the AMS Data Platform. Table 1: Overview of data sets collected by the Urban Pulse project Codename of dataset

UP Flow General

Main variables/ parameters Elevation

Access Restrictions? no

AHN2_5m_surfacelevel CBS_BBG

General

Landuse

CBS_Stats_100m

General

Demographic Statistics

CBS_Stats_500m

General

Demographic Statistics

CBS_Stats_Buurt

General

Demographic Statistics

CBS_Stats_Gemeente

General

Demographic Statistics

CBS_Stats_Wijk

General

Demographic Statistics

RWS_KM

Combination

Climate Monitor

Waternet_WaterUsageLongterm_Restrict ed Waternet_WaterUsageLongterm_Open

Water

Drinkwater Usage

yes

Water

Drinkwater Usage

Waternet_Waterpumps

Water

Drinkwater Usage

Waternet_WaterUsage2014

Water

Drinkwater Usage

yes

Liander_EnergyUsage_Yearly

Energy

Energy Usage

Liander_EnergyUsage_DayProfiles

Energy

Energy Usage

MunAms_Geusage

Energy

Energy Usage

MunAms_MapsAmsterdam

Combination

Several

OrigineBottledWater_2014

Food

Bottled water sources

SupermarketsAmsterdam_2014

Food

Supermarket location

IRI_BottledWater_2014

Combination

Household_Waste_2012

Materials

Bottled water consumption Household waste

AEB_greengas

Energy

Green gas

yes

AEB_MSW_NL_origin

Materials

Waste

yes

Wind_stats

Energy

Wind energy

Zon_Ams

Energy

Solar energy

PORT_stat2012

Materials

imports & exports

yes

AEB_stats

Energy

waste to energy

yes

Food_Consumption_Model

Food

Food consumption

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

URBAN PULSE AMS Kick-Start Project Annex Final Report Version 160129

! ! !

Table of Contents I. II. III. IV. V. VI. VII. VIII. IX. X. XI.

Synthesis of existing studies for Amsterdam resource flows List of Urban Metabolism reference studies from other cities/regions Identified data needs, gaps, availability and data management First evaluation Urban Pulse Amsterdam (addition report section 5) Methodology for Removing the Privacy Sensitivity of a dataset (RPS) Exploring the consumption of bottled water in Amsterdam (IRI data) Metadata Standard for AMS-Urban Pulse Selected publications about the Urban Pulse project Link to higher education at WUR and TUD References Overview data sets

! ! !

URBAN PULSE – Annex final report AMS Kick-Start project

! !

I. Synthesis of existing studies for Amsterdam resource flows !

Existing comprehensive material flow analysis A study of the material flows through Amsterdam is available from 1998 (Figure 1 and 2) (Gorree et al. 2000). Results of this study show that water and fossil energy sources make up 95% of the total flow through Amsterdam (by weight). The remaining 5% are comprised of a diverse number of different materials. The reports points out that the environmental impact of these 5% may be comparatively high. Key recommendations of this report were: • Conduct a separate MFA for fuels and water. • Industry partners should be involved in MFA studies to gain a better insight into industrial flows. • There is much data available for construction material and it is recommended to explore this in greater detail than done in this study. • Complement comprehensive MFAs with more specific MFAs if there is an interest in details. • To overcome data limitations, three methods (or a combination thereof) to gather data are recommended: sampling of data streams, estimations of values possibly from gab samples, use of national statistics in combination with city level correction factors. Energy: Existing studies The new Energy Atlas for Amsterdam presents data about energy consumption as well as potentials both for energy savings and renewable energy provision in the city. Data are provided at the scale of district, neighbourhood and building block. It is meant to be a decision support tool both for inhabitants and businesses. Finally, it can inform the discussion on whether Amsterdam should/can be self-sufficient on the basis of local renewables or if some share should be imported from the greater region. Water: Existing studies A key source for understanding water flows in Amsterdam is the study of (de Fooij 2015). This study presents a MFA of water, phosphorus and organic matter flows through Amsterdam for 2013. This analysis uses the Amsterdam municipal boarders as systems boundaries and does not account for industrial water flows. Results show that Waternet produced 57.2 million m3 of drinking-water for distribution in Amsterdam and that about 75 million m3 of wastewater are discharged by Waternet annually. Furthermore, the total amount of organic matter in Amsterdam wastewater system was about 41kton (COD) originating from faeces, urine, toilet paper and grey water. This organic matter contains between 580 to 700 ton of phosphorus.

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

Figure 1: Material balance Amsterdam 1998 with water (Gorree et al. 2000).

Figure 2: Material balance Amsterdam 1998 without water (Gorree et al. 2000). 24

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

Food: Existing studies There are no comprehensive empirical studies about food flows in Amsterdam. In 2012 Rabobank Amsterdam published a study to calculate whether Amsterdam could be fed from the land in the province North Holland (Rabobank 2010, Hongerig Amsterdam, kan de stad door haar eigen regio worden gevoed?). This is a study based on statistics not on actual flows. The study shows that in theory the Province North Holland has more than enough land to provide Amsterdam for potatoes, vegetables and milk, just enough land to provide for eggs and fruits, but far too little land to produce the meat consumed in Amsterdam (p. 26, table 3.1). In 2011 Buck Consultants performed a study commissioned by the relevant food companies to picture the incoming and outgoing flows of food on the Food Centre Amsterdam (FCA) wholesale market: (FCA 2011). This study is based on interviews with some companies and practically informed estimates. This study shows that actually only about 3-4% of the food handled at the FCA originates from the region (province NH, see slide 13). Hence there is an imbalance between what the city consumes and what the province can produce. This imbalance is addressed in several documents such as the Rabobank Amsterdam Food vision (Rabobank 2012, Voedsel verbindt Amsterdam) and the Municipal Food vision (Gemeente Amsterdam (2014) Voedsel en Amsterdam, Een voedselvisie en agenda voor de stad). Although increasing the share of regionally produced food in the cityâ&#x20AC;&#x2122;s food consumption was an explicit goal of the Amsterdam Food vision (Amsterdam 2014, p. 24), the new Municipal Government did not take over these ambitions and restricted its role to facilitate rather small scale urban agriculture initiatives (Agenda Groen 2015-2018, p. 26-27). Materials See study of Gorree et al. (2000) described above.

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

II. List of Urban Metabolism reference studies from other cities/regions Table 1: Summary of published MFA studies Study! Gorree!et! al.!(2000)!

System!boundary! Amsterdam;! boundaries!based!on! municipality!

Base!year! 1998!

Hendriks! et!al.! (2000)!

Vienna!city;! boundaries!not! further!specified!

1998!

Chambers! et!al.! (2002)!

Greater!London;! boundaries!based!on! boroughs!(34! included)!

2000!

Hammer! et!al.! (2003)!

Hamburg,!Vienna,! Leipzig;!City!and! Metropolitan!region;! boundaries!based!on! administrative! entities!(Eurostat! NUTS!system)! (Greater)!Budapest;! boundaries!not! further!specified!

Hamburg:! 1992V2001! Leipzig:! 1992V2001! Vienna! 1995V2003!

Paris!Municipality,! Metropolitan!area! and!Administrative! region;!boundaries! based!on! administrative!areas! Lisbon!city;! boundaries!based!on! municipality!

2003!

PomĂĄzi! and!SzabĂł! (2008)! Barles! (2009)!

Niza!et!al.! (2009)!

2005!

2004!

Browne!et! Limerick!City!Region;! al.!(2011)! boundaries!based!on! electoral!districts!

1992V2002!

Rosado!et! al.!(2014)!

2003V2009!

Lisbon!Metropolitan! region;!boundaries! based!on! municipalities!

Scope!of!investigation! Focus!on!big!streams:!water,!fossil!fuels,!building! materials,!minerals!and!agricultural!products;!CO2,! food,!textiles;!Some!of!these!categories!are! itemized!in!great!detail;!Mixed!level!of!detail;! Water,!air,!energy!sources,!produce!and!consumer! goods,!construction!materials,!waste!and!waste! water;!Six!indicator!substances:!carbon,!nitrogen,! aluminum,!iron,!lead!and!zinc;! Materials!not!specified!in!detail;! Flows,!production,!import!and!export,!and!waste!of! construction,!crude!materials,!wood,!metals,! chemicals!and!fertilizers,!misc.!manufactures,!misc.! articles,!waste,!food,!water,!which!here!are!listed! as!categories!and!are!then!itemized!broadly;!! Great!level!of!detail;!! Fossil!fuels!(coal,!crude!oil,!natural!gas),!minerals! (metal!ores,!industrial!and!construction!minerals),! biomass!(agriculture,!forestry,!grazing,!fish),! biomass!products,!chemical!products,!other! industry!products,!imports!and!exports;! Materials!not!specified!in!!detail!and!allocated! according!to!methodology!groups,!e.g.!DMC! Water,!(natural)!gas,!electricity,!heat,!food,!waste! water,!municipal!solid!waste,!air!pollutants!(CO2,! SO2,!NOx,!CO!and!particulate!matter);! Materials!not!specified!in!detail;! Fossil!fuels,!minerals,!Biomass!(crops,!wood),! municipal!waste,!C&D!wastes,!fertilizers,!road!salt,! sewage!sludge;! Materials!not!specified!in!detail!and!allocated! according!to!methodology!groups,!e.g.!DMC! Industrial!and!municipal!waste,!C&D!waste,! wastewater!solid!fraction,!air!emissions,! nonrenewable!resources!(nonmetallic!minerals,! fossil!fuels,!metals),!biomass!(agriculture,!forestry,! fishery),!nonspecified!materials,!water! Domestic!extraction,!imports!and!exports!of! agricultural!and!marine!biomass;!forestry! products;!fossil!fuels;!metalliferous!ores,!waste! and!scrap!metal;!and!construction!materials!and! other!crude!minerals;!Household,!commercial!and! industrial!waste,!greenhouse!gas!emissions!and! acidifying!gases.! Materials!not!specified!in!detail!and!allocated! according!to!methodology!groups,!e.g.!DMC! 28!material!types!in!the!five!Eurostat!categories:! fossil!fuels,!metals,!nonmetallic!minerals,!biomass! (forestry,!crops!and!animal!products),!chemicals! and!fertilizers,!others!

! 26

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

III. Identified data needs, gaps, availability and data management Data gaps and availability In general a large amount of data is available about resources in Amsterdam, but their accessibility and level of detail is variable (see appendix XI and main report section 5.3). There are publicly available statistics concerning the socio-economic situation, waste, water and energy available from CBS, the municipality and Onderzoek, Informatie en Statistiek Amsterdam. However, the data that is publicly available is limited to yearly totals. To complete the comprehensive MFA, it was necessary to obtain restricted data sets from the project partners. These data sets were more comprehensive and contained more detailed information on the type of flows and their routes into the city. In particular, the data sets from the Port of Amsterdam (materials, biomass, food), AEB (biomass, waste), Waternet (water, biomass) and the municipality (waste) were crucial for conducting a MFA of Amsterdam. Furthermore, a dataset of LEI about arable areas and yields in Amsterdam (based on the yearly Landbouw telling) was useful to understand role of (urban) agriculture regarding resources needs within the boundaries of the city. Dynamic (temporal) and high-resolution (spatial) data for Amsterdam are scarce. Key datasets obtained included annual water meter readings per household (Waternet), flow meter drinking water readings per minute for Diemen Noord (outside municipal boarders) and 3 minute interval readings rioolgemaal at Prinseneiland. Data for gas consumption per building block level and year were obtained from Alliander. In addition, a dataset about the sales of bottled water in Amsterdam per post code and week was purchased from IRI (for more details see main report section 5.4). The water data have been used to validate a drinking water model, which is capable to model drinking water demand and wastewater discharge at household level as well as neighbourhood level. Currently, a follow up research is conducted that applies this validation as an input for modelling the drinking water demand patterns across the whole of Amsterdam (see section IV). The IRI data have been purchased to illustrate the possibilities for dynamic data on the case of bottled drinking water. By pursuing the IRI data, the project has highlighted opportunities for new data sources. The long-term ambition should be to reduce the price tag that these data have and make them available for the public. AMS may be able to take a first step towards this by seeking to partner up with IRI. Another alternative data source is a mobile phone application on food consumption that has been developed by LEI. In this APP people upload their food consumption data. We propose this is a route to acquire food consumption data, independent of retail outlet and directly from the end 27

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

consumer. However, currently the data uploaded is limited and not Amsterdam focussed. In conclusion, for urban metabolism studies at city scale there is a solid database available, but data access is often restricted. Furthermore, data are distributed across many parties, which results in a large data gathering effort. Data with high information value (spatial and temporal resolution) are very limited and often restricted, but UP project shows that there are opportunities to obtain this type of data. Finally, modelling could be a method to generate high-resolution data, if this is desired. Data management and AMS Data Platform Urban Pulse contributed to the setting up of the new AMS Data Platform. This was the role of work package 6 (Data Management). Specifically, their contribution was to standardise data collection and storage within the urban pulse project. The results produced are: Firstly, a metadata standard for the Urban Pulse project that is compatible with leading standards. This standard will form the basis for a metadata standard for AMS and is currently in use by other AMS projects, thereby helping in the data management of the newly created institute (annex VII). Secondly, a database of all relevant datasets, which are all described according to the metadata standard. Therefore, we make sure that data is not â&#x20AC;&#x153;lostâ&#x20AC;? but available to future projects and researchers (overview in annex XI). Thirdly, a methodology for removing the privacy sensitivity of a dataset, formulated in close collaboration with Statistics Netherlands (CBS). This is a first approach to deal with privacy sensitive data in societally acceptable way. This methodology should be further developed to enable researchers to work with privacy sensitive data in a clear transparent way (see annex V).

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

IV. First evaluation Urban Pulse Amsterdam Note: Content presented here is in addition to the results presented in section 5 of the report. In the following, the analyses of several specific flows of Amsterdamâ&#x20AC;&#x2122;s Urban Pulse are presented. Temporal dynamics of drinking water use MSc thesis Anna Goede, supervised by Marc Spiller (WUR, 2015) To understand the dynamics of water use we worked closely with Waternet and the Kringloop Water Research institute (KWR). Waternet provided minute data from flow meter reading, while we used models from KWR to replicate the measured data. Figure 1 shows a common pattern for drinking water use of the area of Diemen Noord. Though Diemen Noord is another municipality it is expected that drinking water dynamics can be also applied to Amsterdam when socio-economic data are used as correction factors. 140" 120"

m3/h%

100" 80" 60" 40"

00:00:00" 00:45:00" 01:30:00" 02:15:00" 03:00:00" 03:45:00" 04:30:00" 05:15:00" 06:00:00" 06:45:00" 07:30:00" 08:15:00" 09:00:00" 09:45:00" 10:30:00" 11:15:00" 12:00:00" 12:45:00" 13:30:00" 14:15:00" 15:00:00" 15:45:00" 16:30:00" 17:15:00" 18:00:00" 18:45:00" 19:30:00" 20:15:00" 21:00:00" 21:45:00" 22:30:00" 23:15:00"

20"

Figure 1: Simulated and measured weekday water use patterns in Diemen Noord (Goede and Spiller 2015).

URBAN PULSE – Annex final report AMS Kick-Start project

Food and feed related Phosphorus flows in Amsterdam MSc thesis Bouke Bakker, supervised by Marc Spiller (WUR, 2016) Phosphorus (P) is an essential macronutrient and a non-renewable resource. Depletion of Phosphorus reserves is expected within one or two centuries (Herring, 1993; Steen 1998; Cordell, 2009; Van Vuuren et al., 2010; Cooper et al., 2011). Therefore the phosphorus cycle has to be improved to reduce Amsterdam’s dependency on a depleting resource and increase sustainability. In total 746.9 ton of P is consumed through human food and feed for pets and livestock in Amsterdam (figure 1). For human consumption milk and dairy products contribute the largest share, with 34%. The average Amsterdam citizen consumed a high amount of P per day (1658 mgP/day; The median habitual intake for 25 -75% of the Dutch population is between 1,135 – 1,803 mg/day for men and 1,136 – 1,381 mg/day for women). The reason for this is that Amsterdam citizens consume 41 kg/year more milk and dairy products than the average Dutch citizen.

Phosphorus%consumed%through%food%&% feed% 57"[t],"8%"

Human"

212"[t],"28%"

Pet" 478"[t],"64%"

Livestock"

Figure 1: Phosphorus consumed through food and feed in Amsterdam.

In Amsterdam P is discharged in three ways, with the wastewater (WWTP), the solid waste (MSW) and by defecation by pets and manure of livestock (figure 2). It can be seen that the discharge via the wastewater and the solid waste make up more than 2/3 of all P flows, but also defecation by pets and livestock contribute a significant share.

URBAN PULSE – Annex final report AMS Kick-Start project

Discharge%of%phosphorus%in%waste% 141"[t],"19%" Wastewater" Municipal"Solid"Waste" 150"[t],"21%"

437,"60%"

Open"defecaJon"[pets]" and"manure"

Figure 2: Amount and distribution of phosphorus disposal in Amsterdam.

Of the food waste (only inedible parts of food; wasted food etc. not included) slaughterhouse waste, bones and fish bone make up the largest share of P flows (>50%). But also coffee ground contributes about 1/3 to the P balance in residual food waste (figure 3). These two flows could therefore be a good first target for separate collection and recovery of P. Composi9on%91,715%kg%P%in%inevitable%food%waste% 5,689" 2,434" 5,979"

24" Meat"and"ﬁsh" CoﬀeeOground" Peels"and"stumps"

27,256"

50,371"

Eggshells" Thee"stains" Cheese"wax"Crust"

Figure 3: Composition of food waste in Amsterdam (figure only shows food residuals and not wasted food etc.)

Current P recovery projects are the Fosvaatje of Waternet, where struvite is recovered form wastewater sludge. Another target sector is the fertiliser producer ICL located in the Amsterdam Harbour. ICL Fertilisers strives to replace 25.000 ton of phosphorus (by 2025) from ores with ash, bone meal and struvite. Similarly the company Orgaworld (Greenmills) is an important player in Amsterdam. Since 2010 they process 120000 ton of supermarket waste and other organic waste to energy and fertiliser. 31

URBAN PULSE – Annex final report AMS Kick-Start project

The Urban Energy Metabolism in Amsterdam MSc thesis Changsoon Choi and Tom van Heeswijk, supervised by Sven Stremke and Ilse Voskamp (WUR, 2014) Abstract: Since the industrial revolution, fossil fuels became a fundamental resource to sustain human societies. Considering energy, many cities today represent a linear urban metabolism whereas fossil fuel resources are for the most part imported, and used inefficiently. Large outputs of waste are disposed. Greenhouse gas emissions as a consequence stimulate disastrous effects of global warming. Additionally the depletion of fossil fuels forces us to seek alternatives. Therefore, a more effective urban energy metabolism driven by renewable energy assimilation is needed. This Master thesis offers energy-conscious strategies for urban planning that are able to improve an urban energy metabolism. Amsterdam is used as a case study whereby it’s energy metabolism is analyzed in order to illustrate how energy-conscious strategies could improve a city’s energy metabolism in practice. Therefore an example of the often-missing link between urban metabolism and energy-conscious spatial planning is made. The study consists of two parts: a literature study and a case study. A literature study was conducted on two subjects: urban metabolism and energy-conscious concepts and strategies. The energy-conscious concepts and strategies are identified and framed by urban metabolism concepts. Subsequently, a case study on Amsterdam is conducted based on the Urban Harvest Approach (Leduc and Van Kann, 2013). This method exists of five steps: (1) land-use inventory, (2) energy demand inventory, (3) local energy potential analysis, (4) energetic linkages analysis, and (5) exploration of network patterns. The method is applied on metropolitan scale: the municipality of Amsterdam (see figure 1, 2 and 3 below). Consequently, an energy vision for Amsterdam is developed, based on energy-conscious strategies framed by urban metabolism and the case study (see figure 4). Various energy-conscious strategies gained a ranked relevance for improving three metabolic components: input, output and internal processes in a city’s energy metabolism, reaching an optimized linear metabolism. Having Amsterdam used as a test case, illustrated paths of achieving a sustainable urban energy metabolism can contribute in increasing sustainability of cities. !

URBAN PULSE – Annex final report AMS Kick-Start project

Figure 1: Representation of the ‘energy ecosystem’ (or urban energy landscape) of Amsterdam with key stakeholders (Stremke, 2014).

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Energy balance Amsterdam 05-02-2015 Tom van Heeswijk Changsoon Choi

WASTE ENERGY 52.410

HEAT 2.001 WOOD KETTLES 72 SOLAR THERMAL 29 AEB 3.438

SOLAR PV 23 WIND ONSHORE 476

HOUSEHOLD 18.010 ELECTRICITY GENERATION

19.450 UTILITY 20.671

IMPORT ELECTRICITY

INDUSTRY 6.646

223 IMPORT NATURAL GAS

32.914

USED ENERGY 36.275

BIOGAS 157

GAS SUPPLY 27.538

AGRICULTURE

EXPORT ELECTRICITY 3.646

IMPORT HEAT 676

IMPORT COAL 522.126

IMPORT OIL AND REFINED PRODUCTS 848.859

TRANSSHIPMENT OIL AND REFINED PRODUCTS

Figure 2: SANKEY diagram of the Amsterdam energy system in 2012 (Van Heeswijk and Choi, 2014).

EXPORT OIL AND REFINED PRODUCTS 946.865

EXPORT COAL 475.246

TRANSPORT 12.003

ENERGY IN AMSTERDAM

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

Figure 5.3 Energy infrastructure of Am (2014).

residua process heating buildin this he The He networ district The transhi 8 comb in prox subseq plant. Two kV whi electric 20.000 lines in transfo amoun AEB Figure 5.3 Energyinfrastructure infrastructure of of Amsterdam. Amsterdam. Edited Edited from from ArcGIS TenneT (2011) and Energy Atlas waste a Figure 3: Energy TenneT (2014), (2011) and Energy Atlas (2014) (2014). (Van Heeswijk and Choi, 2014).

residual heat is utilized. Residual heat from energy generation processes is directly extracted and added to the district heating network. Buildings (e.g. homes, offices and industry buildings) connected to the heating network are able to utilize this heat for example for space heating and water heating. The Hemweg powerplants are not connected to the heating network, although they have a potential as large supplier of district heating. The shipment intensity indicates where goods are transhipped in the port of Amsterdam. Power plant Hemweg 8 combusts coal each year, therefore Hemweg 8 is located in proximity to transhipment locations for coal. Coal is subsequently being transferred on conveyor belts to the power plant. Two voltages of power lines are present: 380 kV and 150 kV which are owned and maintained by TenneT. TenneT is an electricity transporter company in Europe with approximately 20.000 km of high voltage power lines (TenneT, 2011). Power lines in Amsterdam are directly connected to power plants via Thesedistricts substations regulate the Figure 4: Overall proposal fortransforming energy exchangesubstations. between the different of Amsterdam. Some of right which are energy sinks (consumption exceed potential areas) andgrid. some are sources amount of power ofsupply/red electricity to the (potential supply exceeds consumption/green areas) (Van Heeswijk and Choi, 2014). AEB is a company located in the port and converts domestic waste and biomass into energy in the form of electricity and 35

URBAN PULSE – Annex final report AMS Kick-Start project

V. Methodology for Removing the Privacy Sensitivity of a dataset (RPS) Author: Corné Vreugdenhil, Wageningen University – 17 December 2015 It is quite common for researchers to work with privacy sensitive data. It is a great opportunity for a researcher to analyse a privacy sensitive dataset that is not open for public. However, the researcher must be careful with publishing results or conclusions from such a privacy sensitive dataset in order not to break legal privacy rules. This document describes a methodology of how to deal with privacy sensitive data if one wants to publish (part of) it. This methodology is developed for the Urban Pulse project (UP), a project of the Amsterdam Institute for Advanced Metropolitan Solutions (AMS). Urban Pulse is about getting more insight in the dynamics of the city of Amsterdam, where the focus is on flows of water, energy, materials and food. Researchers of Urban Pulse currently work with several datasets of which some are containing privacy sensitive data. This methodology is developed with the help of Statistics Netherlands (CBS, Eric Schulte Nordholt and Peter-Paul de Wolf). They shared their expertise on this subject and gave useful advice. Their contribution to this methodology development can mainly be found in the first section of this chapter. The privacy rules Dutch legal privacy rules In the Netherlands, the ‘Wet Bescherming Persoonsgegevens’ (WBP, law on protecting personal data), determines the legal privacy rules. An easy to read (Dutch!) summary and explanation on this law can be found here: http://www.justitia.nl/privacy/. After visiting CBS we summarized the privacy rules as the following: A published dataset may not provide privacy information of the individual. The privacy information may not be traceable to the individual person, individual company or individual group. Subjectivity of implementing privacy rules The legal privacy rules seem logical and clear, however the implementation of these rules is quite subjective. The reason for this is that there is no universal definition whether a specific information variable is privacy sensitive or not. The WBP states that something is privacy when it can be used to identify an individual or can value somebody. Variables to identify somebody are for example a phone number or (IP-)address. Variables to value somebody are for example their IQ, income or healthiness. But is it for example privacy sensitive information how much water somebody consumes per day? Or how much 36

URBAN PULSE – Annex final report AMS Kick-Start project

rental you pay per month for your house? Or how much hours you work at your work? CBS shared us the experience that privacy sensitivity is varying in space and time. In the Netherlands we consider some information privacy sensitive, while other countries (even within the EU) do not consider them privacy sensitive and vice versa. In the past we considered something privacy sensitive, that we do not see as privacy sensitive anymore these days and how will it be in the future? CBS gave an example of Scandinavian countries where they decided to provide insight in the yearly income of each household and their financial capacity. The idea was to update this information every year, however they stopped updating this dataset. The issue was that kidnappers used this information on financial capacity to select victims and to claim an as high as possible amount of money. Whether information is privacy sensitive depends on whether you look at the individual person, company or group. Some information is privacy sensitive for the individual person (age or IQ), however maybe not sensitive for the individual company (age of company or electricity consumption) and maybe also not for the individual ethnic group (average age, average income, average electricity consumption). Publishing of spatial data & privacy sensitivity CBS also pointed at two possibilities on publishing a spatial dataset in spatial sense. A map can be made ‘clean’ by (1) blurring or smoothing the final image to publish, or by (2) first ‘clean’ the original data (table) and map the ‘clean’ data. The result of the first one is that the whole picture stays complete, however, the image isn’t sharp any more. The result of the second version is that (‘clean’) details are still there, however the image also contains holes or areas with ‘NoData’. The decision which methodology is best depends on the story ones want to tell with the image. Rules of thumb Realizing that implementing the privacy rules is not so straightforward, one needs at least guidelines. CBS provided us some useful rules of thumbs and advisements: ! No dominance o There may not be a dominant player in the group that is represented by one individual number. Don’t show for example the average electricity consumption in a certain street with 3 households and one large company. The large company is dominant here and the average electricity consumption is a very 37

URBAN PULSE – Annex final report AMS Kick-Start project

good estimate of the specific electricity consumption of that one large company. ! No ones and twos o An individual number may not represent only one or two individuals. If one individual, you publish directly the information of an individual. If two individuals, an individual that is one of the two can find information about the other one, since he knows his own value (for example electricity consumption). ! No group disclosures o An individual number may not represent a group of individuals that belong to the same ethnic or group. A so-called ‘disclosure’ should be avoided. o This is quite a subjective one. In the past it was for example not done to publish statistics like average income or criminal activity per ethnic group. Nowadays, CBS publishes such statistics for some specific ethnic groups that are quite large in the Netherlands. They do not show statistics of the smaller ethnic groups since these are too small. ! Use the minimum level of detail that is really necessary. o What level of detail is really needed to tell what you want to tell with it? Use this minimum resolution needed and not more, since this is not necessary to tell your message and will only cause more risk to break the privacy rules. Stepwise implementation of the privacy rules (see figure below) Step 0: Check for privacy sensitivity Is there anyway privacy sensitive information in the dataset you want to publish with respect to persons, companies or groups? If not sure, just follow the following steps. If you think there is no privacy sensitivity at all, jump to step 7 and at least check this with the data provider. Step 1: Define aim of the publication It is important to decide what the aim is of the publication. What do you want to tell the public with this (visualization of the) data? Step 2: Decide the visualization method How do you want to tell what you want to tell? Will you spatially map the data? Or present results in a table (or graph, but this can be handled equal as table)? Step 3: Decide the way of ‘cleaning’ – Spatial data In case of visualizing a map and depending on what you want to tell with the spatial visualization of the data, decide whether you want to blurring/smoothing your final image, or first ‘clean’ the source data. 38

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

Step 4: Decide the necessary level of detail Map cleaning: What spatial resolution does your map at least need to tell what you want to tell with it? Use the minimum resolution needed and not more, since this is not necessary to tell your message and will only cause more risk to break the privacy rules. Table/source cleaning: What level of detail do you need to tell what you want to tell? How small and what is the composite of the group that is represented by one individual number? Use the largest size as possible, since it is not necessary for your message to have smaller sizes and it will otherwise only cause more risk to break the privacy rules. Step 5: Define and implement the cleaning rules Map cleaning: Use blurring or smoothing to clean the image you want to publish. Do not present exact numbers in the image. Table/source cleaning: Think of the rules of thumb given by CBS. Define cleaning rules for the dataset in a way you ensure that no privacy information is traceable to (1) dominant groups, (2) that you do not provide information on one or two individuals together and (3) check, if possible, whether numbers are representing specific ethnic groups. Step 6: Check usefulness off the cleaned table/image After cleaning your map or table, can you still tell what you wanted to tell with your visualization? If not, consider another level of detail or other representation type. If yes: Congratulations, you can publish the visualization in the way you needed for your message and will not harm people with it. Step 7: Check with data provider When your map, graph or table is finished and you think it is clean of privacy sensitivity, please check the result with a colleague, and at least with the data provider (if present). Since privacy sensitivity in datasets and the cleaning of this is very subjective, it is very important to check with others whether your opinion is the right one!

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

URBAN PULSE – Annex final report AMS Kick-Start project

Example of RPS implemented for a privacy sensitive WaternetDrinkwater dataset The Aim Get more insight in the drinking water consumption of Amsterdam, by mapping in several ways the drinking water consumption in Amsterdam without publishing privacy data. Main goal is to see what different mapping of the same dataset means for interpretation of that dataset. Mapping will be done using the following spatial features: ! Postcode areas: 4-, 5- and 6-digits level; ! Administrative areas: districts (wijken) and neighbourhoods (buurten); ! Gridded: 100- and 500-meter resolution. And each spatial feature providing the following information: 1. Consumption of all connections; 2. Consumption of only household connections; 3. Consumption of household and small business connections. Consumption is given as a Total Yearly Usage, abbreviated to ‘TYU’. The Visualization Method Visualization method is mapping The Cleaning Method Cleaning of the privacy sensitivity will be done by cleaning the source data before mapping. The Level of Detail Needed The purpose of this publication is getting more insight in the drinking water consumption and comparing several levels of detail. For that reason the source data will be kept with a detail level as high as possible. The Rules To deal with privacy sensitivity of the dataset, the following rules were applied: 1. If within one spatial feature only 1 or 2 connections are present, all information is set to ‘Not Applicable’ (NA). a. If zero connections, there is no privacy information, so the values stay zero and will not be converted to NA. 2. If within one spatial feature only 1 or 2 connections of the connection type ‘Household’ or ‘Small Business’ are present, the TYU of these connection type will be set to NA in order to safely visualize the drinking 41

URBAN PULSE – Annex final report AMS Kick-Start project

water consumption for only the households or small business connections. a. If zero connections, there is no privacy information, so the values stay zero and will not be converted to NA. 3. If within one spatial feature only 1 or 2 connections are present within a certain connection type, their total yearly usage within that connection type may not form a larger part than 50% of the summed total yearly usage (TYU) within the whole spatial feature. If the part of one connection type within such a spatial feature is larger than 50%, the summed TYU within the whole spatial feature will be set to zero in order to safely visualize the drinking water consumption for all connections together. a. Example 1: An spatial feature contains 2 Large Business connections with a summed TYU of 200.000 m3/year and the summed TYU over all connections within this spatial feature is 300.000 m3/year. The part of the two Large Business connections is 66%, so this may only be visualized and published with the summed TYU of this spatial feature set to NA. b. Example 2: An spatial feature contains 1 Large Business connection with a TYU of 100.000 m3/year and the summed TYU over all connections within this spatial feature is 250.000 m3/year. The part of the one Large Business connection is 33%, so this may just be visualized and published. c. This rule will not be applied on connection type ‘fire extinguisher system’ (Brandblusinstallatie), since this connection type is not thought to be privacy sensitive. The Check of Usefulness The purpose of this visualization is getting more insight in what removing privacy sensitivity means with different visualizations. So actually all resulting maps after cleaning are by definition useful. The table below shows the percentage of features that are, for a certain variable (Total consumption, household part or small business part) set to NA.

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Percentage)of)features)set)to)NA)for)privacy)reasons Gridded,)100)m)resolution Gridded,)500)m)resolution Administrative)areas,)Municipality Administrative)areas,)Districts)(wijk) Administrative)areas,)Neighbourhoods)(buurt) Postcode)areas)(building)(BAG))representation),)4Qdigits)level Postcode)areas)(building)(BAG))representation),)5Qdigits)level Postcode)areas)(building)(BAG))representation),)6Qdigits)level Postcode)areas)(area)(ESRI))representation),)4Qdigits)level Postcode)areas)(area)(ESRI))representation),)5Qdigits)level Postcode)areas)(area)(ESRI))representation),)6Qdigits)level

Total)TYU Household)TYU Small)business)TYU 10% 11% 35% 12% 17% 19% 0% 0% 0% 0% 0% 11% 1% 0% 1% 2% 1% 0% 2% 2% 14% 3% 4% 25% 2% 1% 0% 2% 2% 14% 4% 5% 25%

The Check with Data Provider Waternet gave permission by email on 6 November 2015.

RE Waterdata.msg

URBAN PULSE – Annex final report AMS Kick-Start project

VI. Exploring the consumption of bottled water in Amsterdam (IRI data) Author: Corné Vreugdenhil, Wageningen University – 13 January 2016 1 Introduction In the context of the AMS Urban Pulse project, an exploratory study was done on a dataset of bottled water consumption in Amsterdam. This dataset was obtained from IRI1 a company that collects information on items that are sold by supermarkets and retailers. The obtained dataset from IRI (for now ‘IRI data’) that is available for the Urban Pulse project is on all bottled mineral water, sold by retailers and supermarkets (except Lidl and Aldi) in the municipality of Amsterdam for the year 2014. The total volume (litres), total price (€) and total sale units (-) are provided on a weekly basis for the year 2014 and per 4-digit postcode group (due to privacy rules, some 4-digit postcodes were aggregated to form a postcode group). The exploration of the IRI data was done iteratively. One project member started with performing some first analysis on the dataset and reported this to the other project members. Project members gave feedback and made suggestions for the next round of exploring analysis. In the end, a total of four rounds were performed this way. This report is a summary report of all findings. 2 Methods 2.1 The Spatial and Temporal Dimension Involved An important aspect of this exploration of the IRI data was to take into account the temporal and spatial variability of the dataset. In order to do so, the dataset was explored with different combinations of dimension taken into account. Table 1 shows how the variation in temporal and spatial dimension was taken into account and which section of this report belongs to it. Table 1: Combinations of dimensions taken into account

Temporal)Variation) None! Weekly! None! Weekly!!

Spatial)variation) None! None! Per!IRI!area! Per!IRI!area!

Report)Section) 3.1.!Year!total!data! 3.2.!Weekly!data!for!Amsterdam! 3.3.!Year!data!for!Amsterdam!per!postcode! 3.4.!Weekly!data!for!Amsterdam!per! postcode!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 1!https://www.iriworldwide.com/nlBNL!!

URBAN PULSE – Annex final report AMS Kick-Start project

2.2 Volume, Units and Euros: Their derivatives The IRI data includes three variables: the sales in volumes (litres), units (-) and euros (€). From these variables, three other variables were derived that tell other information on bottled water sales. The three derived variables were: The average volume of a sale unit (litres/unit), the average price per sale unit (€/unit) and the average price per volume of bottled water (€/litre). 2.3 Abbreviations and Definitions In the IRI dataset some abbreviations and definitions are used that have to be clarified (see table 3 below). Table 2: Abbreviations and definitions

Unit) Volume))

Defined!by!IRI!as!one!object!or!unit!that!can!be!sold,!also!called!a!'sale!unit'.!! Can!be!one!bottle!of!water,!but!also!one!package!of!6!bottles.! The!volume!of!the!bottled!water!in!litres!

Euro)

The!price!of!the!bottled!water!in!euros!

Base)

This!'base'!is!defined!by!IRI!and!is!a!modelled!variable,!based!on!the!measured! data,!! that!indicates!how!much!water!is!sold!under!normal!price!conditions,!meaning! that!no!special!offers!are!included.! An!abbreviation!for!'Koolzuur!Houdend'!and!means!that!only!carbonated!water! is!included! An!abbreviation!for!'Koolzuur!Vrij'!and!means!that!only!nonBcarbonated!water!is! included!

KH) KV)

3 Findings 3.1 Year Total Data

summarizes the water consumption for the whole of Amsterdam in 2014, so no temporal or spatial variation. The following new variables were derived from the available ones: the price per unit, price per volume and volume per unit. Table 3

From some first facts have to be highlighted. In 2014, a total of 21.7 million litre of bottled water was sold in Amsterdam. In 2014, Amsterdam had 765715 inhabitants, which result in an average consumption of 28.4 litre/inhabitant in 2014. In reality, inhabitants of Amsterdam consume in total less bottled water, since (day) tourists in Amsterdam are not taken into account now, but they consume of course part of this 21.7 million litre. Since 2016, the excise duty on Table 3

URBAN PULSE – Annex final report AMS Kick-Start project

mineral water in the Netherlands is 8.83 €/hectolitre2. The Amsterdam consumption of bottled water (same as mineral water) in 2014 resulted in a total € 1.917.975 of taxes. The average volume per unit was 1.45 litres and the average price per unit was 0.88 euro. This is quite a large volume and this immediately shows that we are looking to sale units and not to bottle units. IRI defined (see Table 2) ‘unit’ as a sale unit, which can be one bottle or one package of several bottles. Figure 3 shows an example of a sale unit that consists of 6 bottles of 0.5 litres in one package. The average price per volume was 0.61 €/litre. Figure 3 shows the price per unit and per litre of bottled water at the Lidl supermarket. It strikes that the price for bottled water at the Lidl supermarket (0.24 €/litre) is 2.5 times cheaper than the average price per litre found in Amsterdam. One has to realize that the Lidl and Aldi supermarkets are generally cheap and that these supermarkets are not included in the IRI dataset. It can therefore be concluded that the impression given by the IRI dataset of the average price per litre is an overestimation of the reality, since the Lidl and Aldi supermarkets are not included.

! Figure 3: Example of bottled water price, their 'sale unit' and price per litre.

From it can be stated that the average price per volume doesn’t differ that much between carbonated and non-carbonated water. This is actually striking, since carbonating water sounds more costly than keeping water nonTable 3

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2!http://download.belastingdienst.nl/douane/docs/tarievenlijst_accijns_acc0552z70fol.pdf!

URBAN PULSE – Annex final report AMS Kick-Start project

carbonated. However, a look at the websites of supermarkets, like for example www.AH.nl, showed that this is really truth! Even more striking is the fact that the average price per volume is equal for total volumes and Base volumes. However, this could be due to the fact that only 3.5% is sold under a special offer condition. A last clear conclusion is that most sold bottled water is noncarbonated (65,2%).

Table 3: Year totals for Amsterdam Units Units_KV Units_KH Units_Base Units_Base_KV Units_Base_KH 14943136.5 9794217.0 5148919.5 14300643.0 9325501.9 4975141.1

Year7of7data:72014

Volume Volume_KV Volume_KH Volume_Base Volume_Base_KV Volume_Base_KH 21721121.1 14162657.1 7558464.1 20954783.1 13622949.0 7331834.1 Euro Euro_KV Euro_KH Euro_Base Euro_Base_KV Euro_Base_KH 13184583.3 9059638.6 4124944.7 12718672.6 8675434.5 4043238.1 Total

Average7volume(l)7per7unit Average7price7(€)7per7unit Average7price7(€)7per7volume7(l)

1.45 0.88 0.61

Total_KV Total_KH Base 1.45 1.47 0.92 0.80 0.64 0.55

Base_KV 1.47 0.89 0.61

Base_KH 1.46 0.93 0.64

1.47 0.81 0.55 !

3.2 Weekly Data for Amsterdam This section is on the weekly water consumption in the whole of Amsterdam, so spatial variation isn’t involved while time variation is involved. First the time-trend is studied as it is in the first section, the other section relate this time-trend with weather circumstances and tourists visiting the city of Amsterdam. 3.2.1 The Time-Trend !

shows all the variables that were originally given by IRI per week. All these variables show the same yearly trend: more water is sold in the summer period and less in the winter. A clear distinction can be made between the base and total volumes sold. The base-trend is a stable trend without remarking peaks (except that there is a beautiful yearly variation). The total-trend shows peaks during the year with an extreme one in the summer period. It looks striking that sometimes the total-trend is lower than the base-trend, however this is due to the fact that the base-trend is not really measured, but a modelled variable based on the total-trend (see Table 2). For carbonated and noncarbonated water the same trends can be seen as for the total trends, which means that the part of carbonated water over the total amount of water doesn’t change much during the year. Figure 4

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

shows the three variables given by IRI per week as the three derived variables. It appears that the amount of volumes, units or euros has a seasonal trend (as already shown in ! Figure 5

Figure 4);

while the three derived variables do not show this. The average prices per unit and volume, as the average volume per unit seem to be constant over the year, indicating that only the amount of water sold is changing over the year and not that much the prices of water. During the summer people in Amsterdam consume around 500.000 litres of bottled water per week, while in the winter this is around 350.000 litres per week. This is a difference of 150.000 litres per week. In terms of transportation this extra amount of bottled water, this difference of 150 m3 of water is equivalent to a weight of 150 tonnes kg/week and 21 tonnes kg/day. And the special week 29, during the heat wave, a total of 700.000 litre water was sold, which is 200.000 litres more than average during summer period (in this case also same difference with the week before.). This means that in one week, 200 tonnes kg had to be transported extra compared to normal. The bottled water consumption could be expressed in truckloads, where the maximum load of an average truck in the Netherlands is 9 tonnes kg3: The bottled water consumption in Amsterdam is in winter period 39, in summer 55 and during heat waves 78 truckloads/week.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 3!http://www.swov.nl/rapport/rB2003B35.pdf!!

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! !

! Figure 4: Weekly data for Amsterdam (2014).!

! Figure 5: Weekly bottled water consumption expressed in variables that show a clear seasonal pattern (units, volumes and euros) and variables that do not show a seasonal pattern (average volume per unit, average price per unit and average price per volume).

3.2.2 Meteorology The weather condition could be influencing the amount of water sales. In order to test this idea, meteorological data was compared with the water sales. Daily 49

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

measurement data from 2014 is downloaded from the closest meteorological station, station number 240 of KNMI at Schiphol. The daily measurements were aggregated per week to compare it with the weekly data of IRI. The following meteorological variables were studied: ! ! ! ! ! !

T_mean T_max SunHours_mean GlobalRadiation_mean Humidity_mean Evaporation_mean (mm/day)

The average temperature during a week (degrees Celsius) The maximum temperature of the week (degrees Celsius) Average time per day of sunshine during the week (h/day) Average global radiation per day during the week (J/cm2) The average relative humidity during the week (%) The average potential evaporation per day during the week

Values for the linear correlations are given in Table 4 expressed in the squared version of Pearsonâ&#x20AC;&#x2122;s product, the R-squared or shortly the R2. Correlating the volumes of bottled water sold per week with the weekly meteorological variables, it turns out that the average temperature is most correlated with the drink water sales (expressed in volumes). Table 4: Linear correlation values between IRI data and meteorological variables R2 T_mean T_max SunHours_mean GlobalRadiation_mean Humidity_mean Evaporation_mean Volume 0.82 0.80 0.44 0.58 0.24 0.68 Volume_KV 0.83 0.82 0.47 0.62 0.25 0.72 Volume_KH 0.78 0.72 0.36 0.49 0.19 0.59 Volume_Base 0.81 0.72 0.41 0.64 0.21 0.72 !

The graph in Figure 6 shows the volume sales against the average temperature. It turns out that the relation is best described by an exponential function, however, the R2 for linear functions is not that much lower than the exponential ones. The highest correlation (R2 values: 0.88 exponential, 0.83 linear) is found for the volume sales of non-carbonated water. It appears that non-carbonated water sale is more related to the temperature than carbonated water. This viewed in Figure 7 where the volume-percentage of carbonated water is shown over the year 2014. Here a yearly fluctuation is found where relatively less carbonated water is sold in the summer (the more warm) period. This could explain why temperature slightly better describes the variation in noncarbonated water. It appears that week 29 has the highest temperatures and even included a heatwave at Schiphol (5 days temperature above 25 degrees)! It was in the end not an official national heatwave, since the temperature was a bit lower in the Bilt, the official head meteorological station of KNMI. However, following the requirements on a heatwave, there was a heatwave according to measurements from the meteorological station at Schiphol.

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

The graph in Figure 8 shows that the average temperature is an influencing variable and can be used to predict the bottled non-carbonated water sales with a linear or exponential function. Still the peak in week 29 seems an exceptional extreme peak and it could be that in this week something else happened that cannot only be explained by the average temperature, or other meteorological variables. Figure 8 only shows the predictive value of the average temperature and the non-carbonated water, however the average temperature can also be used to predict the total volumes of bottled water sold. It can be stated that the average temperature is an influencing variable and can be used to predict the bottled water sales with a linear or exponential function (R2 values of respectively 0.82 and 0.87).

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 6: Average temperature against the bottled drink water sales. Including trend lines. 52

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 7: Time-trend of volume-percentage of carbonated water against total volume sold for the year 2014. Also included is the average volume per unit through the year. 53

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 8: A linear and exponential function modelling the actual water sales of non-carbonated water for the whole of Amsterdam, based on the average temperature. 54

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

3.2.3 Tourists The peaks of bottled water sales in weeks 14/15 and 37 (see ! Figure 4),

could be explained by a certain event or happening taking place in a specific part of the city. Figure 9 shows the number of visitors of events in the city centre of Amsterdam (postcode areas 1011-1012), next to bottled water sales (thanks to Niels Tomson). According to this figure, there is no clear relation between bottled water sales and events in the city centre, while one could actually expect this. However, visitors of an event could also buy their bottles of water at temporary shops that are not included in the IRI dataset.

! Figure 9: Bottled water sales in the city centre of Amsterdam, daily mean temperature and the amount of visitors of specific events in the city centre (Niels Tomson).

55! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

shows the total number of guests and nights in hotels in Amsterdam per month for the year 2014. What can conclude from this is that (1) there are many tourists in Amsterdam and (2) that there is a seasonal pattern where more visitors are in Amsterdam during summer than winter period. While there is a seasonal pattern equal to the bottled water sales, there is not a high correlation. Week 29 popped out in the IRI data, which is in July, here July doesnâ&#x20AC;&#x2122;t show an extreme, it is even smaller than August. Good to realize that on yearly basis almost 12,5 million nights and 6,7 guests were visiting Amsterdam and stay at least one night. The IRI data shows a total amount of almost 15 million units of water. Figure 10

! Figure 10: Bottled water sales (volume) in Amsterdam and the number of visitors expressed in guests and nights.

56! !

URBAN PULSE – Annex final report AMS Kick-Start project

3.3 Year Data for Amsterdam per postcode (2014) Using the ESRI version of postcode areas4 in Amsterdam, the IRI dataset could be visualized in the spatial dimension. Due to privacy rules/agreements with the retailers, IRI had to aggregate some of the 4-digit postcodes with each other to one area. Figure 11 shows the ESRI version of 4-digit postcode areas in Amsterdam and the areas used by the IRI data. First thing that has to be concluded is that there is no data (and so no retailers) in the Northwest and Northeast regions. Next to this, incidentally some 4-digit postcode areas also do not have retailers, according to the IRI dataset. Figure 12 shows the IRI areas with the contours of Amsterdam in the background in order to have an idea where the areas actually are. The IRI data is summarized per IRI area for the whole year 2014, in order to have insight in the spatial variation in water sold. Figure 13 shows the total volume (litres) of water sold in 2014 per postcode area as grouped by IRI. Two areas peak out: (1) the centre of the city (postcodes 1011-1012). This could be due to the fact that many people come here, work here and buy their water for the day. And also tourist who came to Amsterdam mainly are concentrated in this area (this area includes the central train station and many touristic places). (2) The northern part of ‘Amsterdam-Zuidoost’ (postcodes 1102-1104). So far no clear clue why in that area so much water is sold. Could be that there is a large shopping area (Amsterdam Arena) which functions the same way as the centre of the city. In Figure 15 the year average volume per sale unit is mapped per district. This figure shows that the further away from the centre of the city, the larger the sale units were. Figure 16, that visualizes the year average price per sale units per district, shows that clearly some areas have high prices per sale units (more than 1.00 euro) and others are relatively much cheaper (less than 60 cents). However, Figure 16 doesn’t show a clear spatial pattern and that could be explained by the fact that here the variable is a combination of sale unit size and prices. Figure 17, that visualizes the year average price per volume, shows a more clearly spatial pattern. The centre of the city has the highest price per volume (more than 1 euro/litre) and the further away from the centre, the lower the price per volume (less than 50 cents/litre). Figure 18, showing the year average volume of bottled water consumed per inhabitant, shows a relatively high consumption in the centre of the city. This can be explained by the presence of many (day) tourists in this part of the city.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 4

http://www.esri.nl/arcgis-content-datasets

57! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

It is important to realize here that the IRI does not include data from the Aldi and Lidl supermarkets. Figure 14 shows the locations of these supermarkets, in total a 23 supermarkets (Lidl = 13, Aldi = 10) that are not included in the IRI data. From Figure 14 it can be concluded that missing these supermarkets is influencing the view on the spatial distribution of water sales, since these supermarkets provide normally their items for low prices. The spatial distribution of these 23 supermarkets shows that the centre of the city has no missing supermarkets, but the surrounding areas do. In other words: There is a change that the extreme high peak of water sales in the centre isnâ&#x20AC;&#x2122;t that extreme since surrounding areas are lacking data. !

58! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 11: The 4-digit postcode areas (ESRI) and the aggregated IRI postcode areas in Amsterdam.

59! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 12: The IRI areas (aggregated 4-digit postcodes) in Amsterdam.

60! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 13: Total volume of bottled water sold in 2014 per IRI area.

61! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 14: Total volume of bottled water sold in 2014 and the locations of Aldi and Lidl supermarkets in Amsterdam (Litre/year).

62! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 15: The year average volume per sale unit per district (litre/unit).

63! !

URBAN PULSE – Annex final report AMS Kick-Start project

! Figure 16: The year average price per sale unit per district (€/unit).

64! !

URBAN PULSE – Annex final report AMS Kick-Start project

! Figure 17: The year average price per volume per district (€/litre).

65! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 18: The year average volume of bottled water consumed per inhabitant (litre/inhabitant/week).

66! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

3.4 Weekly Data for Amsterdam per postcode (2014) This section looks to the IRI dataset in the spatial and temporal dimension, by looking to the bottled water consumption per week and per district (aggregated postcode group). Figure 19 shows per week the total volume of water sales and per district. Figure 20 shows the same, except that here the values are normalized per district, so the trends of the districts can be compared with each other. It strikes that they all seem to have a peak around week 29. This indicates that there was something during that week that was causing high water consumption over the whole of the city. The extreme weather conditions that week, a heat wave, could explain this (see section 3.2.2). Weather conditions like temperature are normally equal within the city level, which means that it influence the drink water sales over the whole city. What strikes is that some districts have their maximum sale in week 14 and 15 (1065, 1079, 1053 and 1057-1059), and two others in week 37 (1078 and 1079).

Figure 19: The weekly bottled water consumption per inhabitant and per district (litre/inhabitant/week).

67! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 20: The weekly bottled water consumption per inhabitant, per district - normalized per district (-).

! Figure 21: The weekly average volume per sale unit (litre/unit) per district.

68! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

Figure 25 and Figure 26 are showing the derived variables in time per district. In general it can be stated that there are no big changes in time per district for these three variables. Mind the y-axes, variations are limited. What strikes in these figures are some particular lines that represent the districts or postcode areas 1011, 1012 and 1017, the centre of the city. In Figure 21 these areas show the smallest volume per sale unit, and the highest price per volume. This means that in the centre of the city the sale units or bottles are the smallest and sold for the highest price compared to the rest of the city.

Figure 22: The weekly average price per sale unit (â&#x201A;Ź/unit) per district.!

69! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

! Figure 23: The weekly average price per volume (â&#x201A;Ź/litre) per district

4 Discussion The IRI data did not include the Lidl and Aldi supermarkets, which are 23 supermarkets in Amsterdam (see Figure 14). Since these supermarkets are generally cheap, this incompleteness of the dataset results in an overestimation of the average price per volume. This incompleteness also means that findings in the spatial distribution of the bottled water consumption could be wrong. However the findings based on temporal dynamics are not directly affected by the incompleteness of the dataset. The findings of this study are not complete and there is probably more to be found, however this study provides already a good first insight in the dynamics of bottled water consumption in Amsterdam. !

70! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

VII. Metadata Standard for AMS-Urban Pulse Author: CornĂŠ Vreugdenhil, Wageningen University â&#x20AC;&#x201C; 28 August 2015 This document describes the metadata standard that is developed and proposed for the Urban Pulse project (UP), a project of the Amsterdam Institute for Advanced Metropolitan Solutions (AMS). The metadata standard is as far as possible based on the relevant ISO-standards and the INSPIRE standard (see figure 27). Section 1 describes the need of this metadata standard. The development of this metadata standard and the requirements for this standard are given in section 2. Section 3 describes the metadata standard that is formulated for UP, focusing on the structure of the metadata and each metadata element in detail. 1. The need for a special metadata standard for AMS - Urban Pulse 1.1 The general need for metadata Urban Pulse and other AMS projects collect data and store these data in databases. Such databases are only useful and therefore valuable when its data is easily accessible and searchable. In order to prevent that a collection of data will be chaotic, any database needs a certain structure and documentation. The creation of metadata can create structure, and therefore provide more insight in a data collection and ensure that the user can find and access the data it is looking for. Metadata is data about data itself and provides information that can help users to understand whether a dataset is useful or not. Metadata normally is a set of elements that all give certain information about one aspect of the dataset. The most straightforward example of a metadata element is the title of a dataset. 1.2 The general need for metadata standards When it comes to describe metadata of a certain dataset, one needs to know what metadata elements are useful to describe, needs to know what the metadata structure is. Without rules on describing the metadata of a certain dataset, metadata formulations would depend on the person who is creating the metadata. This could only be useful for the person itself and not for its colleagues since they can have different fields of interest and other purposes with the same dataset. Rules and requirements on describing metadata are therefore needed, to ensure that the metadata (and the dataset itself) is valuable for everyone within the a project. These rules are generally formulated in a so called metadata standard. A metadata standard describes the metadata elements that are chosen to be required for all datasets within the database and also specifies requirements on describing each metadata element.

71! !

URBAN PULSE – Annex final report AMS Kick-Start project

1.3 International and European metadata standards Metadata standards are not only project-specific. The International Organization for Standardization (ISO) formulates international metadata standards, including standards for spatial data. The ISO-19115 and ISO 19119 for example are the ISO standards on respectively metadata for spatial datasets and spatial data services. Within the European Union, the standards formulated by INSPIRE (Infrastructure for Spatial Information in the European Community) are leading for describing metadata of spatial data. These INSPIRE standards are based on, amongst other sources, on the ISO19115 standard to ensure compatibility of the metadata with other databases. For projects creating their own database, it is the best to follow the INSPIRE or ISO metadata standards to ensure compatibility with other datasets in the world. 1.4 The need for an UP-specific metadata standard There is one major drawback oto the ISO and INSPIRE standards. To gain full functionality from these metadata standards, one needs to describe an extensive set of metadata elements for each dataset. Describing all these metadata elements is time consuming and therefore not always appreciated by the data custodian5. An expected result of this is that data custodians will not describe all metadata elements, resulting in an incomplete metadata description. UP will collect many spatial datasets, which will be entered in the database by all project members. It would not be wise to require from the data custodians to describe all the metadata elements required by the INSPIRE and ISO standards, since the chance is there that data custodians would ignore the metadata requirements due to the extensiveness of the metadata standard. It is therefore better to limit the amount of metadata elements to a mandatory set that are essential for the purposes of UP. 2. The development of the metadata standard for Urban Pulse This section briefly describes the development of the metadata standard that is specifically formulated for the AMS – Urban Pulse project, for now on the ‘Urban Pulse MetaData Standard’ (UP-MDS). The first part summarizes the requirements on the metadata standard. The second part describes how these requirements are met in the UP-MDS that is given in section 3. 2.1 Requirements on UP-MDS UP-MDS should be simple and clear UP-MDS should be easy to use at moment of data entry, data search and data access. UP-MDS should not form an obstacle in any kind to the data custodian, !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 5!The!data!custodian!in!the!context!of!Urban!Pulse!is!the!project!member!that!adds!a!dataset!or!data!

service!to!the!database!and!is!in!this!context!mainly!responsible!for!describing!the!metadata.!

72! !

URBAN PULSE â&#x20AC;&#x201C; Annex final report AMS Kick-Start project

the data searcher and the data retriever. Structure and elements of UP-MDS should therefore be clear to understand and simple to use. UP-MDS should be compatible with leading standards Each metadata element of UP-MDS must as far as possible link to an existing element described in the INSPIRE metadata standard. This must be done to ensure that the UP metadata standard is compatible with the leading standard. The INSPIRE standards are leading in the European Union, where the current study area of AMS is part of. Since the INSPIRE standards are on their turn based on, and therefore compatible with, the ISO standards, compatibility with an international leading standard is also ensured. UP-MDS should be able to manage flow data UP is mainly focused on understanding the dynamics of resource flows. Spatial data normally provides information on the amount of something at a certain place and time. Spatial flow data however provides information on the flux of something at a certain place and time. In case of spatial flow data, UP-MDS should be able to manage the flow data in a way comparable with managing normal spatial data. 2.2. Implementation of the requirements in UP-MDS UP-MDS should be simple and clear To ensure that UP-MDS is simple, the set of mandatory metadata elements will be limited to the ones that are necessary to meet the purposes of UP. Any other additional metadata will be optional. To ensure that UP-MDS is clear, each individual metadata element will be described in detail. The metadata elements will be grouped to their purpose. A schematic overview of the metadata elements will at least show the grouping of the individual elements. To ensure that UP-MDS is also in practice simple and clear, a test will be done where for some first datasets the metadata will be written according the UPMDS. To ensure that UP-MDS stays simple and clear, it will be tracked whether project members have difficulties with using the UP-MDS. Project members will have the possibility to mention any problem in using UP-MDS, suggestions and complaints. UP-MDS should be compatible with leading standards For each metadata element a link with the comparable INSPIRE metadata element will be given. This link will be given in the metadata standard description given in section 3. The naming of the metadata elements will not 73! !

URBAN PULSE – Annex final report AMS Kick-Start project

always be equal to that of the INSPIRE elements, but the comparable element will be provided. In such a way the metadata according to UP-MDS can be linked with metadata formed according international standards. As far as possible the elements of UP-MDS will be linked with the elements described by the leading rules for the European Community, the Commission Regulation No 1205/2008 for implementing the (INSPIRE) directive 2007/2/EC. If no suitable comparable element can be found in this INSPIRE standard, a suitable comparable element from the ISO standard will be searched for. If no suitable comparable element can be found in both the INSPIRE and the ISO standards, a new element will be defined and described in detail. Defining such a new metadata element will be done according to the rules defined by the ISO-19115 standard in Annex C. UP-MDS should be able to manage flow data To manage flow data, UP-MDS will amongst others contain information on the representation type of the data. In case of a dataset with flow data, one has to set the metadata element ‘representation type’ to ‘flow data’ and specify how the data is representing the resource flow.

74! !

URBAN PULSE – Annex final report AMS Kick-Start project

Figure 27: Overview of the UP – MDS.

75! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

3. The Urban Pulse Metadata Standard This chapter provides an overview of how the metadata standard is structured in metadata groups and provides details on all the individual metadata elements per metadata group. An overview of UP-MDS is provided in the below table. Also, two examples can be found how the UP-MDS is used to describe the metadata of a dataset. In the first example, all metadata elements could be described. In the second example, not all metadata elements could be described, due to a lack of information. 3.1 Descriptions of metadata groups and elements Each metadata group and elements is described below. An element can be mandatory, conditional or optional. Conditional elements are mandatory if certain conditions are met. Optional elements can provide interesting information, however not necessary. Project members are free to describe these optional elements. For each element it is also specified what the so called value domain of the element is. The value domain specifies how the element is described. Possible value domains are for example ‘free text’, ‘one character string’, or a value chosen from a predefined list of values. Identification Group of metadata elements that identifies the data. Title The title characterizes in words the data. The value domain is one sentence of free text. Code name The name of the data as a unique code. The code name should be as short as possible, but it should somehow be easy to recognise the dataset. Spaces and special characters are not allowed in the code name, except underscores “_”. Abstract Summarizes the characteristics of the data. This abstract is free text and may include any information, like for example information on: ! ! ! !

The lineage of the data Purpose of the data The content of the data ...

Information url URL or other link to more information on the resource. The value domain is free text, but should only consist of a URL or other reference to an extern information page or system. Keywords & Dates Group of elements that include keywords for data searchers and relevant dates on the dataset. 76! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

Key phrase The key phrase is classifying the data by subject keywords, e.g. “ ’Amsterdam’, ‘households’, ‘drink water’ and ‘consumption’ for a dataset representing the drink water consumptions of households in Amsterdam. The value domain of this element is a list of words separated by a comma [,]. If the data is linked to one of the resource flows of UP, it is mandatory to include the resource flow as keyword. The possible types are the four resource flows studied by UP: Energy, Water, Food and Materials. Date of metadata The last date on which this metadata was formulated or updated. The date should be according to the ISO 8601 standard: ‘jjjj-mm-dd’. Date of publication (optional) The date on which the dataset was published for the first time. The date should be according to the ISO 8601 standard: ‘jjjj-mm-dd’. Date of creation (optional) The date on which the dataset was originally created. The date should be according to the ISO 8601 standard: ‘jjjj-mm-dd’. Date of last revision (optional) The last date on which the dataset is revised. The date should be according to the ISO 8601 standard: ‘jjjj-mm-dd’. Update frequency (optional) The time-interval in with this dataset normally is be updated. The value domain of this element is free text and for example: ‘yearly’, ‘monthly’, ‘every 16 days’. Temporal & Spatial Extent Group of elements describing the time and place to which the data provides information. Geographical extent (optional) The spatial extent of the dataset in terms of geographical coordinates (x,y) describing the minimum and maximum values of x and y in the coordinate system specified by ‘spatial reference system’, i.e. defining (xmin, xmax, ymin, ymax). Temporal Extent The time period that the dataset represents given as begindate and enddate, i.e. defining (startdate, enddate). The dates should both be according to the ISO 8601 standard (‘jjjj-mm-dd’). 77! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

Geographical Description The spatial extent of the dataset in terms of geographical naming, e.g. “The city centre of Amsterdam”. The value domain of this element is free text. Data type, format and representation Resource type The type of resource which is one of: ‘Spatial data set (dataset)’ or ‘Spatial data services (services)’. Distribution format In what file format is the data available for distribution, e.g. for vector data Shapefile, TAB or something else, and for raster data GeoTIFF, ASCII or something else. Data value units The units in which the data values are expressed. The value domain of this element is free text where for each variable within the dataset the units are described. For each field of variable in the dataset, the following information is required: • • • • •

Name Type Unit Language Description

> Just the name of the field or variable > string(text), float or integers > for example ‘m3’, ‘cm’, ‘kWh’, ‘MJ’ > If words are used, what is the language? > a short description what the variable or field is representing

Spatial representation type How does the data represent the information, i.e. by ‘points’, ‘polylines’, ‘polygons’, ‘raster’ or ‘flows’. The type ‘flows’ means that the data contains information on resource flows, no matter how the flows are represented. If type ‘flows’ is chosen, metadata element ‘Flow data type’ is mandatory. Temporal representation type Specifies whether the data represents the spatial data in time. Can be ‘dynamic’ (variation in time) or ‘static’ (no time involved). If type ‘dynamic’ is chosen, metadata elements ‘Temporal reference system’ and ‘Temporal resolution’ are mandatory. Spatial reference system A description of the coordinate reference system used by the dataset, e.g. the coordinate system WGS84 or the projection. The value domain of this element is free text.

78! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

Spatial resolution The spatial resolution or level of detail in the spatial dimension. In case of raster data this must be expressed as the raster cell size. In case of vector data the data importer must specify what the level of detail is. The value domain of this element is free text. Temporal reference system (conditional) Time reference system used by the dataset. Mandatory if the data contains spatial information in time. The value domain of this element is free text. Temporal resolution (conditional) The time interval or level of detail in the time dimension. The value domain of this element is free text. Spatial data service type (conditional) The type of data service. Mandatory if the ‘resource type’ is equal to ‘Spatial data services (services)’. One has to select a type of service from the predefined list of service types, defined by INSPIRE in Appendix D.3. Flow data type (conditional) How does the data represent the information of flows: (List of options must still be defined) Mandatory if the ‘spatial representation type’ is equal to ‘flowdata’. Resource language (conditional) The language that is used in the resource to form textual data. Mandatory if the data consists of textual data. List of possible languages is defined by the ISO standard 639-2. Access & Constraints Group of elements specifying how to access the data and what constraints are present on access and usage of the data. Access information Free text specifying where the data can be viewed and obtained. Access constraints Free text describing the limitations and rules on accessing the data. If there are no limitations or rules on accessing this data, use the term ‘Open data’. If there are constraints, specify here the rules or constraints given by the provider on accessing the data. Use constraints Free text describing the limitations and rules on using the data. If there are no limitations or rules on using this data, use the term ‘Open data’. If there are 79! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

constraints, specify here the rules or constraints given by the provider on using the data. Publish constraints Free text describing the limitations and rules on publishing parts or derivatives of the data. If there are no limitations or rules on publishing this data in any kind, use the term ‘Open data’. If there are constraints, specify here the rules or constraints given by the provider on publishing the data. Parties & Contact Group of elements providing the parties that are somehow responsible for this data and their contact information. Data custodian The party or person, and its contact information, that specified the metadata information of this data. The value domain of this element is free text. Party on data resource The party or person, and its contact information, that was or still is responsible for the creation of the data. The value domain of this element is free text. Party on data provider The party or person, and its contact information, that provides access and usage of the data. The value domain of this element is free text. Other Parties (optional) Free text specifying other partiest that are somehow related to this data. They could for example be the producers or funders for this data. For each party, its role and contact information should be given. The value domain of this element is free text.

80! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

3.2 Link of UP-MDS with leading standards The following table provides information on how the metadata elements of UP-MDS are related to leading standards. If an element is linked with the INSPIRE Directive 2007/2/EC, its ‘INSPIRE equivalent’ element is given. If an element is not linked with the INSPIRE Directive 2007/2/EC, its ‘ISO equivalent’ element is given from the ISO 19115 standard. Some metadata elements are not linked to a metadata element from a leading standard (formatted in italic-style), since these are defined specifically for UP-MDS to fit the purposes of the Urban Pulse project. UP-MDS Title Code name Identification Abstract Information url Key phrase Date of metadata Keywords & Date of publication Dates Date of creation Date of last revision Update frequency Geographical extent Temporal & Temporal extent Spatial Extent Geographical description Resource type Distribution format Data value units Spatial representation type Temporal representation type Data type, Spatial reference system format and Spatial resolution representation Temporal reference system Temporal resolution Spatial data service type Flow data type Resource language Access information Access & Constraints

Parties & Contact

UP-MDS (Dutch) Titel Codenaam Samenvatting Informatielink Kernbegrip Datum van metadata Publicatiedatum Aanmaakdatum Datum van laatste herziening Update frequentie van de dataset Geografische begrenzing Omvang in tijd Geografische omschrijving Brontype Distributieformaat Data waarde eenheden Ruimtelijke representatie vorm Temporele representatie vorm Ruimtelijk referentiesysteem Ruimtelijke resolutie Tijd referentiesysteem Tijdsresolutie Ruimtelijke data service type Stromingdata type Bron taal Toegangsgegevens

1.1 1.5 1.2 1.4 3.1 10.2 5.2 5.4 5.3

INSPIRE equivalent Resource title Unique resource identifier Resource abstract Resource locator Keyword value Metadata date Date of publication Date of creation Date of last revision

4.1 5.1

Geographic bounding box Temporal extent

1.3

Resource type

3.1.1.335 description - exDesc 2.10.1.271 distributionFormat - distFormat

6.2

Spatial data service type

1.7

Resource language

Toegangsrestricties

Use constraints

Gebruik beperkingen

Publish constraints Data custodian

Publicatie beperkingen Verantwoordelijke voor metadata

10.1

Party on data resource

Verantwoordelijke voor data bron

Party on data provider

Verantwoordelijke voor data aanbieder

Other parties

Andere relevante partijen

2.1.1.37

spatialRepresentationType - spatTpType

2.7.2.196

RS_ReferenceSystem - RefSys

2.7.2.196

RS_ReferenceSystem - RefSys

Spatial resolution

2.2

Access constraints

81!

ISO equivalent

Constraint related to access and use Constraint related to access and use

Metadata point of contact Responsible party (of a 9.1&9.2 specific 'Responsible party role') Responsible party (of a 9.1&9.2 specific 'Responsible party role') Responsible party (of any 9.1 other role)

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

Example of a metadata description according to UP-MDS Metadata elementen Title Code name

Gridded statistics of the municipality of Amsterdam CBS_Stats_100m Gridded (100x100 m) statistics of the Netherlands, Identification Abstract clipped for the municipality of Amsterdam. Contains amongst others demographic statistics. http://www.nationaalgeoregister.nl/geonetwork/srv/dut/search Information URL ?#|1462bbce>ccd3>4362>83d8>4535ca8f44eb Key phrase CBS, statistics, demography Date of metadata 2015-08-18 Keywords & Date of publication 2010-12-08 Dates Date of creation 2010-12-08 Date of last revision 2014-11-20 Update Frequency yearly? Geographical extent 110340, 132400, 476800, 493800 Temporal & Temporal extent 2001-01-01 - 2014-01-01 Spatial Extent Geographical description Amsterdam municipality Resource type dataset Distribution format zipped shapefile Data value units See worksheet 'Appendix Data Value Units' Spatial representation type Vector Temporal representation type na Data type, Spatial reference system EPSG:28992 (RD_new) format and Spatial resolution 100 meters representation Temporal reference system na Temporal resolution na Spatial data service type na Flow data type na Resource language Dutch http://geodata.nationaalgeoregister.nl/cbsvierkanten100m/ato Access information m/cbsvierkanten100m.xml Access & Access constraints None Constraints Use constraints None:Ghttp://creativecommons.org/licenses/by/3.0/nl/ Attribution:G'©GCentraalGBureauGvoorGdeGStatistiek'G Publish constraints http://creativecommons.org/licenses/by/3.0/nl/ Corné Vreugdenhil, Junior Researcher GeoData custodian Information and Remote Sensing at WUR, email: Parties & corne.vreugdenhil@wur.nl Contact Party on data resource CBS, infoservice@cbs.nl Party on data provider NGR/PDOK na Other parties

82! !

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT!

VIII. Selected publications about the Urban Pulse project Reference: Hempen, F. & S. Stremke. 2015. Hartslagmeting Amsterdam. Generation E: AEB journal, December 2015, p.17-19.

83! !

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT!

84! !

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT!

85! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT! Reference: Zoelen, Holtslag and Stremke. 2014. De stad as levend laboratorium. Het Parool, p.5.

nieuws

VRIJDAG 20 JUNI 2014

AMS ‘Derde universiteit’ start vandaag officieel

De stad als levend laboratorium De wetenschappers van het vandaag begonnen Amsterdam Institute for Metropolitan Solutions gebruiken de Amsterdammer als proefkonijn. BART VAN ZOELEN

it de eerste meetresultaten blijkt meteen waarom Amsterdammers op warme zomeravonden de slaap maar moeilijk kunnen vatten. De dertig weerstations die dit voorjaar in de stad zijn opgehangen door onderzoekers van de Wageningen Universiteit, laten nu al zien dat het in hartje stad vijf graden warmer kan zijn dan het KNMI had voorspeld. Van buurt tot buurt kan dit bovendien sterk verschillen, bleek uit de eerste resultaten na het warme Pinksterweekend. In de dichte bebouwing van het centrum blijft de hitte langer hangen dan in buurten met meer ruimte en groen. Het vele water in de binnenstad geeft enige verkoeling, zou je zeggen. Maar dat blijkt toch alleen het geval in het voorjaar, als het water nog koud is, legt hoogleraar meteorologie Bert Holtslag uit. Het onderzoeksproject in Amsterdam stamt al van voor de aftrap – vandaag – van het Amsterdam Institute for Advanced Metropolitan Solutions (AMS), de ‘derde universiteit’, een samenwerking van het Amerikaanse MIT, de TU Delft en de Wageningen Universiteit. Maar de weermetingen zijn wel een mooi voorbeeld van de nieuwe kennis over steden die het instituut moet opleveren met Amsterdam als levend lab en de Amsterdammers als proefkonijnen. Het meetnetwerk maakt het mogelijk om een weersverwachting op kleine schaal op te stellen. Het KNMI meet vooral buiten de bebouwde kom, om de metingen zo zuiver mogelijk te houden. Maar het onderzoek werpt ook nieuw licht op het welbevinden van Amsterdammers en de gevolgen van de ‘hitte-eilanden’ in steden. “Wanneer voelen mensen zich comfortabel? We weten uit onderzoek dat een temperatuur boven de 23 graden maakt dat mensen minder goed slapen. In een appartement in de stad kom je daar op zomerse dagen zeker boven. En dan zien we dat mensen weer airconditioning gaan kopen, waardoor hun buren het weer warmer krijgen.” Omdat vooral ouderen gebukt gaan onder de gevolgen van hitte in de stad, levert het onderzoek nuttige informatie op voor de GGD. Ook kunnen stedenbouwkundigen er hun voordeel bij doen bij het ontwerpen van nieuwe straten. Denk daarbij ook aan de afvoer van regenwater waar bestaande buurten niet op gebouwd zijn. “Het lijkt erop dat de toekomst meer kans geeft op pittige buien, zoals op Pinkpop.” Het AMS kan ervoor zorgen dat dit onderzoek nog jaren doorgaat, zegt Holtslag. Nu is het hooguit één tot twee jaar zeker van financiering. Verder zou hij de metingen graag uitbreiden naar de windkracht en straling. Het opvangen van door klimaatverandering vaker voorkomende stort-

buien is maar één van de voorbeelden die het AMS vandaag op de startconferentie in het onderzoekscentrum van Shell in Amsterdam-Noord noemt als typisch stedelijke problemen waar het AMS zich over gaat buigen. Andere onderwerpen: het dichtslibbende transportsysteem, de mogelijkheden van groenteteelt in een stad en hergebruik van grondstoffen. Maar het AMS denkt ook aan afvalstromen, het bouwkundig ontwerp van steden en duurzame energie. “De Amsterdammers gaan het merken,” belooft collegevoorzitter Dirk Jan van den Berg van de TU Delft. “De stad wordt een living lab. Het onderzoek gebeurt in interactie met de Amsterdamse burgerij.” Het Tropeninstituut waar het AMS zijn intrek neemt zal volgens hem dan ook de komende jaren een trefpunt worden voor alle betrokken bedrijven, onderzoekers, de gemeente en de bewoners van Amsterdam.

‘Waar komen straks de zonneparken van de stad?’ Ook Sven Stremke noemt het een groot voordeel dat het AMS voor minstens tien jaar wordt opgezet. De landschapsarchitect uit Wageningen ziet het als een unieke kans om hier de hele overgang naar wind- en zonne-energie mee te maken. “Dit is niet even een project. Amsterdam heeft uitgesproken ambities rond duurzaamheid en de circulaire economie. Voor een onderzoeker is het een enorme kans om mee te kijken en dat misschien zelfs nog te versnellen.”

‘W

at ontbreekt er nog om door te pakken?” is één van de vragen die hij zich zal stellen. Het bundelen van de enorme hoeveelheid data die Waternet, afvalverbrander AEB en stroomnetbedrijf Alliander ieder voor zich verzamelen over het verbruik door Amsterdammers, is een goede eerste stap. Het afval van het ene bedrijf kan een grondstof zijn voor het andere bedrijf. “Voorwaarde is daarbij wel dat alle functies op de juiste plek komen. En waar komen bij wijze van spreken straks de zonneparken van de stad? We mogen in de energietransitie ook weer niet te veel fouten maken. Voor je het weet komt er weer een vervuilende kolencentrale bij. Maar we kunnen in Amsterdam een eind komen.” Stremke merkt op de betrokken universiteiten al veel enthousiasme tonen over de onderzoeksmogelijkheden in Amsterdam. “We hebben nu al zo veel studenten die bij het AMS aan de slag willen. Ze willen dolgraag iets bijdragen aan de stad.” Als Amsterdammer spreekt het hem ook aan. Stremke: “Ik ben proefkonijn en onderzoeker tegelijk.”

Het meetinstrument (aan de lantaarnpaal) van het AMS voor de temperatuur in het centrum (Damstraat).

Een bescheiden, nog niet zo duurzaam pand

niversiteit is een groot woord als het gaat om het Amsterdam Institute for Advanced Metropolitan Solutions (AMS). Het moet vooral een netwerk zijn van aan andere universiteiten en bedrijven verbonden onderzoekers. De aftrap van het AMS was vanmiddag dan ook nog niet in eigen collegezalen, maar in het onderzoekscentrum van Shell in Noord, waar het AMS voorlopig is ingetrokken. Deze zomer krijgt het AMS een eigen plek in het Tropeninstituut. Maar dat is geheel volgens plan. In het bidbook waarmee de universiteiten van Delft en Wageningen met het Amerikaanse MIT vorig jaar de door de gemeente uitgeschreven competiADVERTENTIE

86! !

FOTO MATS VAN SOOLINGEN

tie wonnen voor zo’n urbaan kennisinstituut, werd ook gedacht aan een niet al te groot pand. Er moest ruimte zijn voor een collegezaal met honderd studenten, dertig tot zeventig werkplekken voor onderzoekers en kantoren voor een kleine eigen staf. Het onderzoek zou dan wel in Amsterdam plaatsvinden, maar voor de verwerking van de ingewonnen gegevens konden de wetenschappers gerust terugkeren naar hun campus in Delft, Wageningen of Boston. Alle digitale mogelijkheden om op afstand te werken staan dat ook toe. Ook de betrokken bedrijven – onder meer Accenture, Cisco, IBM, KPN en Shell – werken op die manier mee.

Wat nog niet geheel volgens plan verloopt, is de locatie zelf. Die zou eigenlijk de duurzame en innovatieve bedoelingen van het AMS moeten uitdragen, zo was althans het plan waarmee de competitie werd gewonnen. Met het vroegtwintigste-eeuwse Tropeninstituut is daarvan in elk geval niets terechtgekomen. Het is ook nog niet gelukt ‘een echte smaakmaker’ als directeur aan te trekken, zoals het AMS wilde. Maar die was lastig te vinden, aangezien de universiteit zich eerst nog moet bewijzen. Nu is Renee Hoogendoorn aangesteld als interim-directeur. Zij was eerder projectontwikkelaar en adviseur in ruimtelijke ordening.

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT! Voskamp I. and S. Stremke, 2014. The Pulse of the City: Exploring Urban Metabolism in Amsterdam http://www.toposonline.nl/2014/the-pulse-of-the-city-exploring-urban-metabolism-in-amsterdam/

THE PULSE OF THE CITY: EXPLORING URBAN METABOLISM IN AMSTERDAM 05-11-2014

Considering ongoing, rapid urbanisation and the vast resource consumption of metropolitan areas around the world, it is important to integrate urban resource management with the design of our future cities. But how can resource management become an integral part of planning and designing urban landscapes? Ilse Voskamp Research assistant Landscape Architecture Group

Sven Stremke Assistant Professor Landscape Architecture Group Principal Investigator for Energy at Amsterdam Institute for Advanced Metropolitan Solutions (AMS)

2 |

Amsterdam Institute for Advanced Metropolitan Solutions (AMS) The City of Amsterdam recognized the need to address the challenges cities are facing and so launched a call for the creation of a new research institute on metropolitan solutions in April 2013. The proposal by the consortium consisting of Massachusetts Institute of Technology (MIT), TU Delft, Wageningen UR and several industry and knowledge partners was selected and in June 2014 the official opening of the so-called Amsterdam Institute for Advanced Metropolitan Solutions (AMS) took place. The institute has three pillars: research, education and data platform. Research focusses on how the provision and management of resources and services can contribute to urban sustainability while improving the quality of life. Key themes are water, energy, waste, food, data, and mobility as well as the integration of these themes. The universities have appointed Principle Investigators for each theme and the co-author of this article is responsible for energy. From an interdisciplinary perspective, AMS aims to develop a thorough understanding of the city (i.e. sense the city), to design metropolitan solutions and to integrate these in the city. Understanding urban systems by means of Urban Metabolism Urban Metabolism is a concept that is increasingly used to gain understanding of cities [4]. Using the metaphor of the city as ecosystem or organism, urban metabolism has proven valuable for a range of disciplines -from economics to political scienceâ&#x20AC;&#x201C; to study

The Pulse of the City: Exploring Urban Metabolism in Amsterdam

87! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

Figure 1: Study of the urban metabolism of Brussels by Duvigneaud and Denaeyer-De Smet, 1977 [9].

the resource flows in cities [1]. The majority of urban metabolism research has been performed by industrial ecologists to study the conversion of raw materials, energy, and water into built structures, human biomass, and waste [2]. In this context, urban metabolism is defined as “the sum total of the technical and socio-economic processes that occur in cities, resulting in growth, production of energy, and elimination of waste” [3]. Related urban metabolism studies are primarily focussed on quantifying urban resource flows -like water, energy, material– or fluxes of particular substances within these flows –like phosphorous, nitrogen, metals [4]. The results of one of the earliest studies on urban metabolism are shown in figure 1.

Urban metabolism studies generally used to be limited to quantifying the total inputs and outputs of an urban system [3]. Yet, currently the majority of studies goes beyond this “city as black box” approach and also includes resource flows within cities, from source to sink. The most prominent method used for this is Material Flow Analysis (MFA). Such analyses can provide qualitative and quantitative insight in where resources originate from, how they are converted and when they are disposed. MFAs allow to systematically assess the inputs and outputs of selected resources of a predetermined system as well as the flows and stocks within the boundaries of that system [1]. One of the ways to visualize the outcomes of a MFA is a flow chart that specifies the

The Pulse of the City: Exploring Urban Metabolism in Amsterdam | 3

88! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

resource flows and how they are converted by different processes (figure 2). Another way of visualising MFA outcomes is a Sankey diagram. Sankey diagrams provide information on the sizes of (sub)flows within a system as the widths of the arrows correspond to the quantities of the various flows (figure 3). Urban metabolism has inspired new ways of thinking about urban sustainability. It fuelled, for example, the idea that urban areas should become more self-sufficient and resource demands of cities should not exceed the carrying capacity of their hinterlands. This advocates a shift from the current linear metabolism of cities -using inputs only onceto a circular metabolism that incorporates recycling of resources [1]. Moreover, urban metabolism studies have shown to be useful to define sustainability indicators, for example on energy efficiency [4], and to identify processes that are critical for the (un) sustainability of an urban area [3]. It is also increasingly argued that urban metabolism can contribute to sustainable urban planning and design [4]. Scholarly practitioners and design practice increasingly embrace urban metabolism too. This is illustrated, for example, by the substantial attention the 2014 International Architecture Biennale Rotterdam “Urban by Nature” paid to urban metabolism.

Figure 2: Example of a simple MFA flow chart [10].

Urban Metabolism for planning and designing sustainable cities The envisaged value of urban metabolism for planning and design disciplines has different aspects. The metaphor of the city as ecosystem and urban metabolism in particular can be useful for landscape architects, urban planners and designers to think about a city as system with associated resource flows. When urban planners and designers incorporate metabolic thinking while designing urban form and processes, resource management is integrated in urban designs. Here another useful application of urban metabolism surfaces: when aiming to increase the sustainability of a particular city by reducing resource inputs and/or outputs within a (part of the) system, an understanding of the different flows within that city, districts and neighbourhoods is required. For this, an MFA can be a valuable means. Firstly, an understanding of the status quo of resource flows of a city can provide insight where the design challenges lay to adjust a city’s metabolism. So, when starting a planning or design process on the scale level of the city, a MFA can provide insight in what are key locations to zoom in and propose interventions for. Think for example of locations where large amounts of resources are consumed and/or large quantities of waste are generated. Secondly, urban metabolism analysis can be useful when the location for physical interventions is already known (i.e. working on district and neighbourhood scale). Then the MFA can reveal how resource flows on that particular location connect the site to its surroundings. This is essential knowledge to account for the cross-scale implications of design interventions. In spite of the great potentials, to this moment urban metabolism is hardly used in urban planning and design. One possible reason for this lack of application is that urban metabolism fails to acknowledge the importance of socio-economic indicators

4 |

The Pulse of the City: Exploring Urban Metabolism in Amsterdam

89! !

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT!

(e.g. lifestyle) for achieving sustainability [4]. Yet, designing is a means to identify integrated solutions to ecological, spatial and social challenges. The concept of urban metabolism is (with few exceptions) used in a technological paradigm that fails to acknowledge that the interplay between society and biophysical processes is part of and influences the metabolism of cities [8]. Another reason could be that up till now the link between resource flows and spatial characteristics is not well established. This is clearly exemplified by the majority of MFA representations, which lack a spatial dimension. MFAs do not provide information on the spatial organisation of the flows and processes they describe [5]. Moreover, there seems to be a mismatch between the scale level at which urban metabolism studies are performed (city or regional scale) and the scale level of urban planning and design practice (district, neighbourhood, building block) [7]. Finally,

the concept of time is not properly dealt with in current metabolic studies [6]. To fully understand the metabolism of a city, it is essential to acknowledge the variability of resource provision and consumption through time. Understanding these temporal dynamics is essential when planning and designing sustainable cities. Studying Urban Metabolism in Amsterdam The chair group of Landscape Architecture and the Environmental Technology subdepartment (ETE) have, over the past six months, prepared a comprehensive research proposal to study the urban metabolism in Amsterdam (see figure 4). In October 2014, the so-called Urban Pulse project was awarded as one of the first three AMS research projects. Urban Pulse aims to address the spatial and temporal challenges of urban metabolism studies stated heretofore. The objective of this initial project is to understand and map resource

Figure 3: Sankey diagram of the energy flows in the Netherlands and South Limburg [6].

The Pulse of the City: Exploring Urban Metabolism in Amsterdam | 5

90! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

flows in the city of Amsterdam. The flows of water, energy, food and selected materials will be identified in terms of “quantity” (volume/weight) and quality. Their temporal dynamics and location in space will be analysed too. The project aims to provide planners, designers and decision makers with a precise understanding of the dynamic flow patterns in Amsterdam. In addition the researchers will, by closely collaborating with the SENSEable city lab at MIT, explore

how to communicate temporal and spatial variations of urban resource flows by new means of representation. Research will be carried out in collaboration with the city of Amsterdam and a number of knowledge and industry partners. Currently two LAR-students conduct research on the energy metabolism of Amsterdam by means of a minor thesis. Other students that are interested in this topic for their thesis can contact the authors.

Figure 4: Electricity consumption in Amsterdam as part of the energy metabolism [11].

6 |

The Pulse of the City: Exploring Urban Metabolism in Amsterdam

91! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

REFERENCES RESOURCES & ENERGY [1] Broto, V.C., Allen, A. & Rapoport, E., (2012). Interdisciplinary Perspectives on Urban Metabolism. Journal of Industrial Ecology, 16(6), pp.851–861. [2] Decker, E.H. Elliott, S., Smith, F.A., Blake, D.R. & Rowland, F.S. (2000). Energy and Material Flow Through the Urban Ecosystem. Annual Review of Energy and the Environment, 25, pp.685–740. [3] Kennedy, C., Cuddihy, J. & Engel-Yan, J., (2007). The Changing Metabolism of Cities. Journal of Industrial Ecology, 11(2). [4] Kennedy, C., Pincetl, S. & Bunje, P., (2011). The study of urban metabolism and its applications to urban planning and design. Environmental pollution, 159(8-9), pp.1965–73. [5] Moffatt, S. & Kohler, N., (2008). Conceptualizing the built environment as a social–ecological system. Building Research & Information, 36(3), pp.248–268. [6]Stremke, S. & Koh, J. (2011). Integration of ecological and thermodynamic concepts in the design of sustainable energy landscapes. Landscape Journal, 30, 194-213. [7] Spiller, M. & Agudelo-Vera, C.M., (2011). Mapping diversity of urban metabolic functions – a planning approach for more resilient cities. In 5th AESOP Young Academics Network Meeting 2011, 15-18 February 2011, Delft, the Netherlands. pp. 1–14. [8] Wachsmuth, D., (2012). Three Ecologies: Urban Metabolism and the Society-Nature Opposition. The Sociological Quarterly, 53(4), pp.506–523. [9] Mfadiagrams, http://mfadiagrams.blogspot.nl, Last view 5-11-2014 [10] Example of a simple MFA flow chart, http://en.wikipedia.org/wiki/Material_flow_analysis#, Last view 5-11-2014 [11] Boogert, G. den, Hakvoort, L., Heit, R., Le Fèvre, S., Lemmens, B. Mantel, B. Voerman, R. & Vries, B. de., (2014). Energy Atlas Amsterdam Southeast. City of Amsterdam.

References Resources and Energy | 7

92! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT! Reference : Spiller, 2015. The new urban metabolism (Le Nouveau Metabolisme Urbain) Le1 hebdo No 5 Le 1 hebdo - Le nouveau métabolisme urbain

Abonnez-vous à notre newsletter !

OK et recevez le sommaire chaque semaine !

S’identifier

Article précédent

Article suivant

LE NOUVEAU MÉTABOLISME URBAIN

N°54

29 avril 2015

Édito

Marc Spiller, spécialiste en technologie environnementale

Sommaire

Chaque minute, chaque seconde, quelqu’un, quelque part dans la ville, participe au grand cycle ordinaire : on mange, on digère et on se libère. Autrement dit, nous « métabolisons », c’est-à-dire que dans les cellules de chacun de nous se déroulent des processus vitaux qui entraînent la transformation de la nourriture. Et parce que chacun de nous métabolise, les villes aussi métabolisent ! Les repas que nous absorbons ne sont que la partie émergée de l’iceberg de ce métabolisme urbain. Comme les individus, les villes dépendent d’un flux continu de ressources importées, dont la nourriture, l’eau, les matériaux (matériaux de construction, téléphones portables, tissus, etc.) et bien sûr les combustibles, qui répondent à nos besoins en électricité. Comme nous, les villes utilisent ces ressources et les transforment en déchets tels le dioxyde de carbone, la boue de vidange, la pollution des fleuves ou les déchets solides déposés dans d’immenses décharges.

Cet article est réservé aux abonnés. TESTEZ LE 1 GRATUITEMENT PENDANT 15 JOURS EN LIGNE ET RECEVEZ LE PROCHAIN NUMÉRO DANS VOTRE BOÎTE AUX LETTRES!

JE DÉCOUVRE vous avez déjà un compte ? OK

http://le1hebdo.fr/numero/54/le-nouveau-mtabolisme-urbain-909.html[27-1-2016 21:40:18]

93! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

Cite as : Spiller (2015) Le Nouveau Métabolisme Urbain. Le1 hebdo. No.54. http://le1hebdo.fr/numero/54/le-nouveau-mtabolisme-urbain-909.html

The new urban metabolism – from cities as organisms to cities as ecosystems Cities are full of people! Cites are made by people, by the markets they visit, the transactions they make, their social interactions and their resource consumption. Every minute, every second someone, somewhere in the city eats a meal or visits the toilet to dispose an earlier meal. Clearly people metabolise, or in other words, within the cells of every one of us life-sustaining transformations processes take place that convert food to excrement. Because each of us metabolises, so do cities! The meals we eat are only the tip of the iceberg of this city metabolism. As the people, cities depend on a continuous stream of imported resources including food, water, materials (think of construction materials, mobile phones, cloths etc.) and of course fuels to satisfy our hunger for electricity. Like each one of us, cities use these resources and turn them into wastes such as CO2, sewage sludge, polluted rivers or solid waste duped into huge landfill sites. The recognition of this societal metabolism and the use of this metaphor go back to Karl Marx. It has found revived interest since the 1960s (Wolman 1965) when the Metabolism of cities was first described. Since then many studies have investigated the metabolism of cities. These studies viewed cities as organisms that require inputs to sustain their very existence and in this process metabolising inputs to outputs that are commonly labelled as “waste”. Very recently this view of urban metabolism has changed! Rather than viewing cities as organisms they are increasingly likened to ecosystems. In ecology, ecosystems are networks of a diversity of species that interact closely to maximise the efficiency of energy, water and nutrient use. In many cases tightly coupled or symbiotic ecosystem relationships even maintain a continuous recirculation of valuable resources. In the tropical rain forests nutrients that come available from the mineralisation of biomass are directly incorporated into new growth. In this manner no nutrients are lost and resource use efficiency is high. City planners and engineers now seek to learn from the processes that make ecosystems efficient and enable reuse. Currently, modern technologies have brought us closer to the developed what is often referred to as a circular urban metabolism. Take for example the sewage every city produced in large quantities, it is now possible to extract energy from sewage, recover limited nutrients, such as Phosphorous, from it and produce bio-plastics or building materials. However, these technologies are merely an isolated effort. Currently, we are far from understanding the complexities of a cities metabolism and we lack detailed knowledge of the dynamics of this metabolism. In other words, there is no integrated understanding of when and where resources are metabolised, what “waste” resources come free and who could be interested in using them. To obtain the necessary insights into the cities metabolism cities should get smarter. The vision is that real time information becomes freely available so that the “waste” resources a city provides can be exploited, or mined as it is often called, by parties that attach a value to it. Advanced sensing technologies to monitor food, water and energy use and progress in information technology are thought to enable an understanding this pulse of cities metabolism. But no matter how smart the technology gets, cities remain full of people and resources that can be exploited. It is therefore the people, who will make use their ingenuity to find new applications and uses for what today appears to us a non-valuable resource.

Reference Wolman (1965) The metabolism of cities. Scientific American. Vol. 213, issue 3, pages 179-190

94! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

IX. Link to higher education at WUR and TUD Lecture for 35 MSc students Wageningen - Marc Spiller (2014) “Understanding dynamics of urban resource flows” and “Planning for closing resource cycles” Master thesis Changsoon Choi and Tom van Heeswijk (09-2014 –01-2015, finished) – Supervised by Sven Stremke 15 BSc theses projects on sustainable energy landscapes in Amsterdam 2014 – Supervised by Sven Stremke Guest lecture for Arjan van Timmeren at TUD in front of 100 MSc students – Marc Spiller (2014) “Understanding dynamics of urban resource flows” Anna Goede MSc student- Title: Modelling spatial and temporal dynamics of the drinking water flow in Amsterdam (Marc Spiller tutor) Timo Anker MSc student (Marc Spiller tutor with Arnold Bregt) – Title: Mapping potential for the decoupling roof runoff from sewers in Amsterdam Bouke Bakker MSc student (Marc Spiller tutor) – Phosphorus flows in Amsterdam: A substance flow analysis. Carolin Bellsted MSc student Industrial Ecology (Arjan van Timmeren first examiner, Gijsbert Korevaar second examiner) – Title: Material Flow Analysis for a Circular Economy Development Lecture for 20 and 55 TU Delft and Leiden University students (2 editions of the course General Introduction to Industrial Ecology in February and September 2015) by Gijsbert Korevaar: ‘Closing the Loops - on Industrial Symbiosis, Cradle to Cradle and Circular Economy Thomas Dietz 2014, Developing professional urban agriculture in the city of Amsterdam, Research intern at LEI, MSc Environment and Resource Management, Vrije Universiteit, Amsterdam, The Netherlands, supervised by dr. Jan Willem van der Schans; co-supervised by prof. dr. ir. Pier Vellinga, June 2014 Stijn Heemskerk 2015, Opportunities for regional food in the Dutch market for outof-home consumption, Research project of MSc. Environment and Resource Management, Vrije Universiteit Amsterdam, The Netherlands, supervised by prof. dr. ir. Pier Vellinga, co-supervised by dr. Jan Willem van der Schans, Sept. 2015 Lilia Feghiu, MSc thesis student VU Amsterdam, Title: Material Flow Analysis for Amsterdam, Supervised by Pier Vellinga and Sven Stremke, 2015

95! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT!

X. References Annex Barles S (2009) Urban metabolism of Paris and its region. Journal of Industrial Ecology 13 (6):898-913 Browne D, O’Regan B, Moles R (2011) Material flow accounting in an Irish city-region 1992–2002. Journal of Cleaner Production 19 (9):967-976 Chambers N, Heap R, Jenkin N, Lewis K, Simmons C, Tamai B, Vergoulas G, Vernon P (2002) A resource flow and ecological footprint analysis of Greater London. Oxford: Best Foot Forward Ltd Cooper, J., Lombardi, R., Boardman, D., & Carliell-Marquet, C. (2011). The future distribution and production of global phosphate rock reserves. Resources, Conservation and Recycling, 57, 78-86. de Fooij, H (2015) Water as a Resource - Strategies to recover resources from Amsterdam's Wastewater. MSc thesis. Faculty of Engineering Twente. University of Twente FCA (2011) Logistieke inventarisatie Food Center Amsterdam EindrapportageLogistieke inventarisatie Food Center Amsterdam Eindrapportagehttp://www.agrologistiek.eu/downloads/Logistieke%20invent arisatie%20%20Food%20Center%20Amsterdam_eindrapport.pdf (accessed: 26/01/2016) Gorree, M., R. Kleijn, and E. Van Der Voet. 2000. Materiaalstromen door Amsterdam. Hammer M, Giljum S, Bargigli S, Hinterberger F (2003) Material flow analysis on the regional level: Questions, problems. Solutions NEDS Working Paper (2) Hendriks C, Obernosterer R, Müller D, Kytzia S, Baccini P, Brunner PH (2000) Material flow analysis: a tool to support environmental policy decision making. Case-studies on the city of Vienna and the Swiss lowlands. Local Environment 5 (3):311-328 Herring, J. R., & Fantel, R. J. (1993). Phosphate rock demand into the next century: impact on world food supply. Nonrenewable Resources, 2(3), 226-246 Leduc, W. R. W. A., and F. M. G. Van Kann. 2013. Spatial Planning Based on Urban Energy Harvesting toward Productive Urban Regions. Journal of Cleaner Production 39: 180–90. Niza S, Rosado L, Ferrao P (2009) Urban metabolism. Journal of Industrial Ecology 13 (3):384-405 Pomázi I, Szabó E (2008) Urban Resource Efficiency: The Case of Budapest. Hungarian Statistical Review:155-173 Rosado L, Niza S, Ferrão P (2014) A material flow accounting case study of the Lisbon metropolitan area using the urban metabolism analyst model. Journal of Industrial Ecology 18 (1):84-101 Steen, I. (1998). Management of a non-renewable resource. Phosphorus and potassium, (217), 25-31. Van Vuuren, D. P., A. F. Bouwman & A. H. W. Beusen (2010) Phosphorus demand for the 1970–2100 period: A scenario analysis of resource depletion. Global Environmental Change, 20, 428-439.!

96! !

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT!

XI. Overview data sets Table 1: Data sources for energy Source, year and type of source (e.g. report, publication or spreadsheet)

Title

Data (type of flow, temporal resolution, spatial resolution, reference year)

Summary/ comments

Accessibility issues (if applicable)

AmsterdamOpenData, based on data Alliander and DRO, 2013, excel spreadsheet

Verbruiksgegevens Elektriciteit en Gas

Numbers divided by private and business use (not all companies included)

OpenSource

http://www.amsterdamopen data.nl/web/guest/data?state =getPagedCategoryDatasets Alliander, YEAR???, .tab file (a GIS-file)

Electricity and Gas used, CO2emission Per year Postal Code (level four) Reference year 2008 -2011

Yearly Electricity and Gas usage in Amsterdam

Reference year is not clear yet..!

If using this dataset for publishing, credits must be given to Alliander!

AmsterdamOpenData, 2014, SOAP API / wsdl document

Oplaadpunten Elektrisch vervoer

http://www.amsterdamopen data.nl/web/guest/data?data set=oplaadpunten-elektrischvervoer Baseline_questions_AMS,20 13, excel preadsheet

20130225Baseline_questions_AMS

Yearly average electricity and gas usage per building block, for the municipality of Amsterdam. Location and details electric charging stations, including availability Updated every 30 sec. Location (address incl. house nr.) Realtime Wind, Solar, Waste to energy generated Per year Municipal boundary (?) Reference year 2013 (?) -

City of Amsterdam, 2014a, report by DRO

De circulaire stad Amsterdam 2014 / 2018; Part II (p.9 â&#x20AC;&#x201C; 66)

Was not able to access the wsdl documents; instead: http://www.oplaadpalen.nl/z oek/AmsterdamNederland

Data ather unreliable; working document municipality (see next cell)

Received from Bob mantel (Municipality, DRO); source original data not known / accessible

Ambition, strategy, potential

97! !

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT! City of Amsterdam, 2014b, report

Energy Atlas Amsterdam Southeast

http://www.amsterdam.nl/w onen-leefomgeving/klimaatenergie/energieatlas/

City of Amsterdam, 2014c, report by DRO

CO2-uitstootrapportage 2013

City of Amsterdam, 2014d, report by department of Research and Statistics O+S

Jaarboek Amsterdam in cijfers 2014, p.292 - 302

www.os.amsterdam.nl/public aties/amsterdamincijfers

Energy consumption: gas, electricity, cooling demand Per year Per m2, data grouped by function and per 5 connections Reference year 2013 Infrastructure: heating, cooling, sustainable electricity sources (block level) Reference year 2013 Potential: solar, wind, surface water, drinking water, waste water, open and closed-loop TES, geothermal, residual heat, domestic waste, garden waste Per year Per m2 or per km2 (geothermal) or per city district (waste) Reference year 2013 CO2 emissions (in absolute terms and per inhabitant) Per year Amsterdam Municipality Reference year 2013 NO2 concentration in the air (µg/m3) Yearly average of hourly average (source GGD) Point source measurements Reference year: 2009-2013 CO concentration in the air (µg/m3) Yearly average of 24hourly average (source GGD) Point source measurements

Data presented as maps

Emissions related to energy usage (electricity, gas, traffic, heating) 5 measurement points affected by traffic and 5 “background” measurements

3 measurement points

98! !

Also accessible as interactive maps at maps.amsterdam.nl

More data/ original source accessible via: hwww.luchtmetingen.amster dam.nl/informatie.aspx

URBAN&PULSE&PROJECT&–&ANNEX&FINAL&REPORT! Reference year: 2009-2013

www.klimaatmonitor.databa nk.nl online spreadsheets, graphs and short reports, website by Rijkswaterstaat

Klimaatmonitor

O3 concentration in the air (µg/m3) Yearly average of 24hourly average (source GGD) Point source measurements Reference year: 2009-2013 PM10 concentration in the air (µg/m3) Yearly and monthly average of 24hourly average (source GGD) Point source measurements Reference year: 2009-2013 PM2.5 concentration in the air (µg/m3) Yearly average of hourly average (source GGD) Point source measurements Reference year: 2009-2013 Average gas and electricity usage and CO2 emission per dwelling, companies (per sector); CO2 emissions traffic (per road type and vehicle type) Per year Amsterdam Municipality Reference year: 2009-2013

3 measurement points

CO2 emissions and energy usage (gas and electricity) per sector (built environment, mobility, industry and energy, agriculture)and subsector Per year Amsterdam Municipality Reference year: 2008-2012

Option for comparison with other municipalities or province

6 measurement points

5 measurement points

Mostly absolute numbers are given, averages only per dwelling type.

99! !

Original sources of data (e.g. CBS) are indicated

URBAN&PULSE&PROJECT&â&#x20AC;&#x201C;&ANNEX&FINAL&REPORT! www.klimaatmonitor.databa nk.nl online spreadsheets, graphs and short reports, website by Rijkswaterstaat

Klimaatmonitor

EZ/CBS.nl

User profiles for dwellings and businesses

Port of Amsterdam, 2014, excel spreadsheet

Transhipment of product groups

Renewable energy generated (electricity, heat, mobility from incineration, biomass, hydropower, wind, solar, geothermal, biofuels (+some minor categories)) Per year Amsterdam Municipality Reference year: 2008-2012 Yearly average electricity and gas consumption per inhabitant/employee (Dutch averages) Imports of fossil fuels Per year Amsterdam harbours (Amsterdam, Beverwijk, IJmuiden, Zaanstad) Reference year 1990 â&#x20AC;&#x201C; 2013 (2014 incomplete)

Estimations based on Dutch numbers

Yearly totals only (not time and place specifiek)

Accessible

For the local port of Amsterdam the country of origin per product is indicated and both import and export

Received by email from Titus Tielens! Great mismatch between import, local use and export -> Check with Titus!

100! !