GIS Portfolio by Anish Pendharkar

GIS Portfolio Anish Pendharkar

Analyzing inequities in distribution of WiFi Hotspots in Manhattan

Mapping density of Residential Units in Queens from 2003-13

15.4 % Manhattan’s individual poverty rate

Density of Residential Development from 2003-13

Type Free

38.78 %

Limited free

HHs live in census tracts with higher invidual poverty rate than this

Partner Sites

Median HH Income $ 11.6 k - $ 52.3 k

45.62 %

$ 52.3 k - $ 89.3 k

of Manhattan’s

$ 89.3 k - $ 127 k

fall in these tracts

$ 127 k - $ 175.25 k $ 175.25 k - $ 250 k 0

0.75

1.5

3 Miles

US Geological Survey (2016), Data Sources: NYC Dept of City Planning (2016) NYC Dept of Information Technology & Telecommunications(2016), US Census Bureau. 2015 American Community Survey, 5-Year Estimates. Table DP03: Selected Economic Characteristics by census tract, New York County, New York State. [dataset]. Accessed via American FactFinder. 2016

The New Housing Marketplace program was one of the flagship initiatives of Bloomberg administration. To study the residential development in Queens during this period, I mapped the number of new residential units constructed under the program, from 2003-13. Comparing it with the overall residential tax lots developed during this period, I found that the program had followed the characteristic type of exisitng housing fro each neighborhood.

20 ft

tandem.

4 44 3 33

2Greenery 22

We with mapped from a residential lot Location sorting was applied a trees around the study area, and compared them withFig loca2 Trees per 5.2-5.6 Census Tract < 5.2 revealed the same results. tions of residential lots in the neighbourhoods.

away from the Parks: tree density analysis 1Street 11 McCarren McCarren McCarren Park Park Park

Location sorting was applied with a Above Above Above graph graph graph shows shows shows thethe the street street street tree tree tree density density density values values values at at increasing atincreasing increasing distance distance distance Fig 4: Density of (distance-bands) (distance-bands) (distance-bands) from from from mcCarren mcCarren mcCarren Park Park Park centroid. centroid. centroid. revealed the same results. Residential lots The The The line line line corresponds corresponds corresponds to to density todensity density measured measured measured along along along street street street network network network http://www1.nyc.gov/site/planning/data-maps/open-data/dwn-pluto-mappluto.page. (Fig (Fig (Fig 1) 1) and 1)and and line line line corresponds corresponds corresponds to to density todensity density measured measured measured along along along straight straight straight line line line Then, street trees points distances distances distances from from from thethe the centroid centroid centroid (Fig (Fig (Fig 2) 2)2) were mapped alonside the residential tax lots.of Figof 2 McCarren on the park FigFig Fig 0 shows 00shows shows thethe the location location location McCarren of McCarren park park and and and surrounding surrounding surrounding street street street trees. trees. trees. right shows proximity of tree points to Both Both Both thethe the methods methods methods have have have different different different buffer buffer buffer rings rings rings shapes shapes shapes and and and also also also represent represent represent residences. Fig 2 and 3 show density of residential different different different areas areas areas onon the onthe the map, map, map, as as seen asseen seen from from from FigFig Fig 3, resulting 3,3,resulting resulting in in different indifferent different density density density tax lots and the total number of trees per census values. values. values. tract respectively. The intensity of colors in both Fig 3 Trees per figures move in tandem. FigFig Fig 1 and 11and and 2 show 22show show thethe the location location location of of buffer ofbuffer buffer rings rings rings with with with respect respect respect to to McCarren toMcCarren McCarren park park park Tract Census area. area. area. WeWe We can can can also also also observe observe observe that that that rings rings rings containing containing containing greater greater greater parts parts parts of of Park ofPark Park area area area have have have lower lower lower density density density values values values than than than adjacent adjacent adjacent rings. rings. rings.

Fig1 : 1/4th mile distance bands

ark?

s constantly away from

ses 30% at e from the

the study with locathe neigh-

FigFig Fig 0. 0. McCarren 0.McCarren McCarren Park Park Park && Street &Street Street tree tree tree Location Location Location 5.2-5.6

< 5.2

5.6-6.0

6.0-6.8

Half Mile Distance

of street1/8th trees in the half mile radius, fall mile 1/8th 1/8th mile mile within 1/4th mile 1/4th 1/4th mile mile

20 ft

3/8th mile 3/8th 3/8th mile mile from a residential lot 1/21/2 mile 1/2mile mile

ied with a

1/8th 1/8th 1/8th mile mile mile 1/4th 1/4th 1/4th mile mile mile 3/8th 3/8th 3/8th mile mile mile 1/21/2 mile 1/2mile mile

Fig 2: Measured along straight Fig Fig Fig 2. 2. Measured 2.Measured Measured along along along line distance Straight Straight Straight line line line distance distance distance

Fig 1: Measured along street Fig1. Fig1. Fig1. Measured Measured Measured along along along networks

Street Street Street Network Network Network

Fig 3: Comparison of the area Fig Fig Fig 3. 3. Area 3.Area Area covered covered covered byby by covered by rings, by both methods

rings rings rings ofof both ofboth both figures figures figures

buffer rings were created considering center point to beThe at centroid the Park. Distance were * Second, The **The The nodes nodes nodes correspond correspond correspond to to actual to actual actual values values values measured measured measured byby buffer bybuffer buffer rings. rings. rings. The The lines lines lines areare only areof only only visual visual visual guides, guides, guides, notnot not tobands to be tobe interpreted beinterpreted interpreted as as values asvalues values created, first measuring distances along the streets, second, along straight line distances. As seen from Fig 7, for for â&#x20AC;&#x2DC;in-betweenâ&#x20AC;&#x2122; forâ&#x20AC;&#x2DC;in-betweenâ&#x20AC;&#x2122; â&#x20AC;&#x2DC;in-betweenâ&#x20AC;&#x2122; distances. distances. distances.

and the graph below, the transition is random. Also rings containing greater area of thehttps://www1.nyc.gov/site park show lesser Data Data Data :from 1):Street 1) : 1)Street Street Shapefiles.New Shapefiles.New Shapefiles.New York York York City City City Department Department Department of City ofofCity City Planning. Planning. Planning. 2017. 2017. 2017. LION LION LION v17A. v17A. v17A. [ESRI [ESRI [ESRI FileFile Geodatabase]. File Geodatabase]. Geodatabase]. https://www1.nyc.gov/site https://www1.nyc.gov/site /planning/data-maps/open-data/dwn-lion.page. /planning/data-maps/open-data/dwn-lion.page. /planning/data-maps/open-data/dwn-lion.page. density. /data-maps/open-data/dwn-pluto-mappluto.page. 2) Street 2)2)Street Street Tree Tree Tree points. points. points. .Department .Department .Department of Parks ofofParks Parks andand and Recreation Recreation Recreation (DPR) (DPR) (DPR) . 2018. . 2018. . 2018. 2015 2015 2015 Street Street Street Tree Tree Tree Census Census Census - Tree - Tree - Tree Data.https://data.cityofnewyork.us Data.https://data.cityofnewyork.us Data.https://data.cityofnewyork.us /Environment/2015-Street-Tree-Census-Tree-Data/pi5s-9p35. /Environment/2015-Street-Tree-Census-Tree-Data/pi5s-9p35. /Environment/2015-Street-Tree-Census-Tree-Data/pi5s-9p35.

Street Tree Density Analysis

7 6

Street Tree* Density

Adjacent graph shows the street tree density values at increasing distance (distance bands) from Mccarren Park Centroid.

2 1 0

McCarren Park Centroid 0

Distance in miles 1/8

20 ft

from a residential lot

http://www1.nyc.gov/site/planning/data-maps/open-data/dwn-pluto-mappluto.page. http://www1.nyc.gov/site/planning/data-maps/open-data/dwn-pluto-mappluto.page.

3476 / 4630

he second density of per census ntensity of vary in

Half Mile Distance

of street trees in the half mile radius, fall within

ensity Analysis

decreases as one moves away from the park. und McCarren Park, Brooklyn

6.0-6.8

3476 / 4630

Distance Distance Distance

tree points to residences.The second in miles inmiles miles Centroid Centroid and third graphic in show density of 0 00 Centroid residential lots and trees per census How do the tree cover changes when we walk away from McCarren 0 00 1/8 1/8 1/8 1/4 1/4 1/4 toto 3/8 3/8 3/8 tracts 1/2 1/2 1/2 The intensity of toto to toto to to toto to respectively. Park? colours in both visuals vary in To observe this, first I created quarter mile rings, starting from the tandem. Along Along Along Street Street Street Network Network Network Along Along Along Straight Straight Straight line line line distance distance distance boundaries of McCarren Park. Fig 1 shows the the density constantly

5.6-6.0

1/4

Along Street Network

3/8

1/2

Along Straight line distance

Above graph shows the street tree density values at increasing distance (distance-bands) from mcCarren Park centroid.

Site Recommendations

suitable locations or newFinding Public Library in Queens Public Library in Queens

for new

Ancillary Tool: Weighted Decision Maps Essential requirements characteristics

N Recommended Sites

Desirable characteristics Prioirty to sites at greater

• Priority to sites far from • According to Building existing branches libraries Objective was to locate vacant landRegulations, parcels in Queens, New York City, • Higher density overnew lowerpublic libraries. should a minimum which canarea house Somehave essential characterisitcs density areas builtin–up area the of 2000 were identified first, which could help bridging existing access • High percentages of the adult sq.ft inequities in public libraries. Using this essential characterisitcs, I population we want developed ranked maps. The •Also, maps were thenpossible added together to • High first percentages of the siteswhich to rank high on develop a finalage weighted decision map, helped in identifying the population of school scores derived. suitable parcels fulfulling all requirements. • High most percentages of the population born outside of US • High percentages of the population aged five years or •Method = Sites which olderEssential that speak a language provided scope for less other than English at home characteristics than 2000 sq.ft built up and speak English “less than area were dropped from very well.

Ancillary

the selection. •Formula = FacilityFAR* Lot Area <= 2000 •Also Score>=35, where the maximum possible ranking was 37.

• Method = Kernel density maps withrequirements Weighted scores were used to adjudge the preliminary set of such sites. According to building code in NYC, libraries should have a

minimum built- up of 2000 sq.ft. Sites which provided scope for less than this amount were filtered out of the selection, using the formula = FacilitiyFAR*Lot Area <= 2000. Also, only site with scores in the top quantile were considered.

distances from exisitng •More accessible branches •Abutting areas with high footfall •Located on major streets, not in ineer pockets. •Would not come under Flood risk areas according to 2007 and 2015 index.

• •

•Lots having a commercial Areas with higher overlay were identified. residential density •These lots are classified for retail use in residential districts, therefore are located on major arterial roads. •Lots under flood risk, mostly from long island city were filtered.

• •

Atlantic avenue 143-18, Pinegrove street Hillside avenue Out of these 9 sites, some were traffic islands, some too narrow or irregular shaped to be of use. 3 sites were thus selected.

Higher percentage of school age population

121 sites

3 sites were selected after verifying feasibility of construction and nature of ownership (city-owned) through Google maps and NYC Tax lot data. The sites are accessible by transit, have major cultural institutions in vicinity, as well as located on prominent thoroughfares.

Higher percentage of foreign born population

18 sites

Site 2:

Site on Atlantic avenue

Site 3:

143-18 Pinegrove street

High: 39

Decision Map

Site 1: Vacant lot on Atlantic Av

Sites with more accessibilty by transit. Located on major streets, not in inner pockets. Only sites which would not come under Flood risk areas according to 2007 and 2015 vulnerability indexes were considered.

Recommended 8104 sites Sites

148-18 Hillside avenue

Main weighted decision map. The reds represent highest scores, and the blues the lowest. This map was further multiplied with a layer of vacant lots, to identify the sites.

Desirable Characterisitcs

Site 1:

9 sites

Site 2: 143-18 Pinegrove Street, Liberty Avenue

Site 1: Basic Characteristics Address

Vacant lot on HillsideBasic Characteristics Low : 6 Avenue ATLANTIC AVENUE Address

The Atlantic Avenue runs through the breadth of Queens, on the south side. The Block no. of Jamaica, which has Block no. Hillside avenue cuts through the heart 9313 selected site satisfies all our criterias as can be seen from the table, and being a site seen a lot of affordable housing projects in the specified Theon Liberty avenue, also is a major artery crossing through Zone District time for commercial overlay has advantage of additional FAR as well. Being a major R6Ain Queens, is known As can be seen from the table and the pictures, theZone Districtperiod. The area, as many neighborhoods street, and also from C2-3 category, describes its good accessibility. Queens. Also, it enjoys for its diverse ethnographic populations. TheOverlay site provides a has stands a regular square shape and huge ground area, with relaxed Parking norms being a commercial overlays in R6-10 areas,site which Overlay C2-3 huge area, especially important because thereOwner lies no libraries advantage of a corner location as well. Many eateries, a for high density residential neighbourhoods. or similar institutions around the 1580 area (sqasftseen from maps, and church, supermarkets and schools surround the area, and isLot Area Lot Area street views) of also the most The photos show the retail and cultural buildings surrounding theone site, its densely developed regions from 2003-13 as Lot Dimensions 20 x 80 ft Lot Dimensions seen from the first slide. location near a corner fuel station. The first picture down here, shows the vicinity of site with The were Facility FAR selected 3 verified Facility multi-storied parcels old buildings, andfinally thus reinforce ourFAR decision based on GIS analysis. using satellite images from Google. Score 37 Score

Higher percentage of population with less fluency in English Source : Google Maps

Source : Google Maps

oduction:

mine the street network, we make use of open data sources that

uch data sources provide excellent resolution of urban mobility and

Research Questions:

Methodology and Data:

for resilience and emergency evacuation purposes.We are seeking for the answer to the following questions: What is the role of Times Square in the street network? How is the spatial distribution of routes connecting Times Square and other places? Which part of the street network is highly likely to be traveled? What is the impact on the street network at Times Square in New York City after infrastructure failure?

Group project Tools: Network Analysis, Getis Ord GI Test, Betweeness and Centrality measures

eline Analysis:

!! ! ! ! ! ! !! !!! ! !!! !!

! (!

We conducted a baseline analysis to identify the structure of the street network including integration, and network kernel density. This phase helped us acquaint a knowledge of the street network and examine the role of Times Square in the network.

! ! !

! ! !! !! ! ! ! ! !! ! !

! ! !

! (! (! (

! (

! ( ! (! (

! (

! (! ( ! (! ( ! ( ! (

! (! (

! (

Second, we blocked the roads from Columbus circle to 34th Street and 6th Avenue. Hotspots are located right next to the road blockage i.e the impact is concentrated only on the area just beside Times Square.

Baseline Analysis

Extracting the street network from LION Single Line Street Base Map, we established the citywide street model at first. hering network Routes:

! ( ! (

! (

! ( ! !

! (

! ( ! ( ! (

! !!

! (

Based on the street network, the closest facility analysis layer solved the optimal routes through minimal time cost and shortest trip length criteria.

! ! ! ! !

! !

! !!

! !

! ! !!

! !

! ! !

! !

! ! !! !

! !

! ! !! !! ! !

! !

! ! !

! !! !

! !

! ! !

! !

For a hypothetical scenario, we assumed a lateral road blockage, from 11th Avenue to 2nd Ave, 47th to 49th streets. The impact of a lateral street closure extends uptown till the end of Manhattan. The map shows results of a Hotspot Analysis by Getis-Ord GI ! test on the difference. !

! !

! The Aggregation of Minimum Time Cost Routes

! !

!! ! ! ! ! !

! ! !

! ! !! !! ! ! ! !! !! ! ! ! !! ! !!

!! !

! ! ! !! !

! !

!! !

! (

! !

! ! ! ! ! !! ! ! ! !! ! ! ! !!! !

! (

! ! !

! ! ! !

! !! ! ! !! !

! !!

! !

! (

!! (!! ! ( ! ( (! ! ! ! (!! !

! !!!! !! !

! (

! !

! ! !

! !

!! ! ! !! ! !! !! ! !!!! !! ! ! ! ! ! !! ! ! ! !!! ! !

! ! ! !

!! !

! !

! ! !! !

! !! ! ! !! !

! !

! ! !

!! ! ! !

! !! ! ! ! !!

! !

!! ! ! !!!!! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! !

! (

(!! ! (!! ( !!! ! (!( ! ! ! !

! ! !!

!!!

!!!!!!

! ! !! !! ! ! ! ! !

! !

! ! !

! !!

! ( !

! !

! (

! !!

!! !! !! !!

! ! !! !! ! !! ! ! ! ! ! ! !! !!! ! !! ! ! ! !! !!! ! ! !! ! ! ! !! ! !! ! ! !!! !!!! ! ! ! !!! ! ! ! !! ! ! ! !! ! !!!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!!! !! !! !! ! ! !!!! ! ! ! ! !! !! ! !! ! ! !! ! !! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! !!! ! ! !! !! ! ! ! !! ! ! ! ! !! ! !!!! !! ! ! !! ! ! ! !! ! ! !!!!! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! !! ! ! ! ! !!! !! !!!! ! ! ! ! ! !!! ! ! ! !!! ! !! ! ! ! ! !! !! ! !! ! ! !! ! !! !!

! ! ( ! ( ( !! ( ! ( (! ! ! (! ( ! ( ! ! ! ( ! !! ! ! (! ( ! ( !! ( ! ( ! !! ! ! ! !! ! ( ! ( ( ( ( ! ! ( ( ! ! ! ( ! ( ! ( ( ! ( ! (! ! ! ! ( !! ( ! ( ! ( ( ! ! ! ! ! ! ( ! ! ! ( ( ! ( !! !! ! !! !! !! ! ! ! !! ! ! ( ! ! ! ( ( ! ( ( ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! !! ! ! !! ! ! !! ! ! ! !! ! !!! ! ! ( ! !! !! !!! ! ! !! ! ! !!!!! ! ! ! ! ! ! ! ! !! ! ! ! !

!! ! !! !!

! ! !!!! !!!!

!! !

! !

! (

! ! !! ! ! ! ! !

! !

! !! ! !! !! ! ! ! ! !

! !

!! !

! (

!!!! !

! !! !

! (

( !! (

! ! !! !! ! !

! !

! !( (

! !

!! !

( ! (! (!

! (

! (! ( ! (! ( ! ! (

! (! (

( ! (! ! (

( ! (!

! (

! ( !

! ( ( ( ! ( ! ! (( ! ! ( !! (

( !! (

! (

( ! ( ! ( ! ( !

! (

! !!

! ( ! (

! (! ( ! (

! !

! !! ! ! ! !

! (

Scenario 2: Lateral Street blockage

! (

on a typical July evening. We extracted taxi data from TLC Trip Record

! (

!! !

! ! ! !

! !!

! (! (

!! ! !! !! ! ! ! ! !

! !

!! ! !! !!

! ( ! (

! !

! ! !

!! !

Isochrones from and to Times Sq

! !

! ! ! !

! (! (

In this phase, we set a specific Percentage scenario when the crowd was of Trips during weekdays and weekends gathering at Times Square on a typical July evening. We extracted taxi data from TLC Trip Record Data to represent travel activities related to Times Square with a simplification of the travel activities as links between hotels and Times Square. Based on the street network, the closest facility analysis layer solved the optimal gregation of Minimum routes through minimal time cost gth Routes and shortest trip length criteria.

!! ! ! ! ! !

! !

Gathering Routes

! (

! ( (! ! (! ! ( ! ( ( ! ( ! ! ! ( ( !( ! ! ( ( ( ! ( ! ( ! (

! ! !

! ! ! !! ! !! ! ! !! !! ! !! ! ! !!!! ! !! !!! ! ! !! ! ! !! !! ! ! !!!! !! Scenario 1: ! ! ! !! !! ! ! !!!! !!! ! ! Blockage ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!!! !! ! !!! !! ! ! ! ! ! ! along ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !!! ! avenues ! !! ! ! !! ! ! ! !!!! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !! !! ! ! ! ! ! ! ! !!! ! ! !! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ( !! !! ! !!!! !!!!! ! !! !!! ! ! ! ! ! ! !!!!! !! ! ! ! ! ! !! !! ! ! !! !! ! ! ! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !! ! ! !! ! ( ! !! !!! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ( !!! ! !! ! ! ! !! ! ! ! ! ! !! ! ! !!!!!! !!!! ! ! ! ! ! ! ! ! ! ! ! ! (! !! ! ! !!!! ( !! ! ! ! ! ! ! ! !!! ! ! ! ( ! ! !! ! ! ! ! ! ! ! ! ( !! ! ! ! ( ( ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! !!!! ! !! ! ! ! ! ! ( ! ( !! ! ! ! ( ! !! ! ! !! !! ! ! !! ! ! ( ( ( !! ! ! !! ! ! ( ! ! !! ! ! ! ! ! ! ! ( ! ! ! ( ! ! ! ! !! ! ! ! ( !!! ! ! ! ! ( ( ! ! ( ( ! ! ! !!! ! ( ! ! !! !! ! ! ! !! ! ! ! ( ! ! ! ! ! ! !! !

! (

We conducted a baseline analysis to identify the structure of el activities as linksstreet between hotels andincluding Times Square. Based on the street the network , the closest facility analysis layer solved the optimal routes through minimal integration, and network st and shortest trip length criteria. All optimalkernel routes will be aggregated density. This phase helped us st likely network and the abovementioned baseline network. acquaint a knowledge of the street network and examine the role of Times Square in the network.

Based on the street network, the closest facility analysis layer solved the optimal routes through minimal time cost and shortest trip length criteria.

! ! ! !

! !! !! blockage of network resiliency. Finally, we found that lateral!!!street is more critical than blockage along the avenues, ! !!! ! ! though the impact of such blockages on travel distances is minimal. While this is a much simplified model of the actual ! !! !! ! scenario, this result is in line with our findings!!!!from baseline analysis. ! ! !

Network Resiliency:

Times Square is one of the most famous sites in NYC for tourists, and also one of the most important transport nodes for residents.

Through a we seek to answer: What is the role of Times Square the cityâ&#x20AC;&#x2122;s street network? How ng the streetin network ON Single Line Streetspatial distribution of routes is the ap, we established the connecting Times Square and street network model other places? and what is the impact on the street network at Times Square in New York City after infrastructure failure?

! ! !

Three phases includes Baseline Analysis, Gathering Routes, and Link Criticality. Each phases !! Results ! of closure of any section. However, possibly due to the will be discussed in details. routes and thus easily contains the impact ! ! ! Data Set: NYC Dept. of City Planning. New York City Borough Boundary; NYC Dept. of City !area is not critical in citywide road networks. Further, ! the baseline analysis, we Taxi found that role of Times Square ! ! ! Planning. LION Single Line Street Base Map; NYC Through Taxi & Limousine Commission. Yellow ! ! exploring theVolume routines, we found exist! in for routines with impediments trip length time the samedifferences analysis outer However, theofresearch also and has limitations. Trip Records, Green Taxi Trip Records, FHV Trip Records, High For-Hire Vehiclethat Trip significant !! boroughs. ! !is repeated ! ! ! Records; NYC Dept. of City Planning. Map PLUTO;cost. NYC Taxi & Limousine Commission. NYC ! ! ! ! ! ! !!! ! ! for other modes transportation and explore travel time and distance disruption with agentTaxi Zones. ! ofroute ! ! Also the directions to and from Times Sq based greatly change and volume allocation, which suggests different perspective ! modelling. ! !

Traffic network Analysis at Times Square

hicles daily in CBD. Therefore, taking taxi trip data as a proxy for movements and spatial interaction and Times Square, we aim to the essential elements of the street network in New York City.

Conclusion:

! !! !! !

We created the Isochrone to see how the distribution for travel duration from and to Times Square presents geographically. ! Isochrone illustrates diffusion for other taxi zones to Times Square !! ! !! ! because of the network characteristics during peak hours. ! ! !

!! ! !

Groundwater level change and settlement planning in Delhi Tool: Geographically Weighted Regression

Fig: Planned and Unplanned Settlements in Delhi, with locations of groundwater wells

This is a novel exploratory study with two objectives: first, to develop a quantifiable relationship between socio-economic variables that influence groundwater use and the decadal changes in groundwater levels, and second, to observe if these relationships vary between planned and unplanned settlements. Two kinds of regression analysis are used, one global (OLS) and another local (GWR).

Methodology Geographically Weighted Regression is a local spatial analysis tool, which calculates local coefficients for each unit of analysis. The Y variable, i.e. change in groundwater levels, was correlated with all X variables. The resulting maps are presented here.

Results show that most relationships between these factors and groundwater change are as expected, with some exceptions. Though no significant difference was found in these relationships between planned and unplanned settlements.

First, the groundwater level data for each well was interpolated over all of Delhi, to prepare a raster with groundwater level change for each Assembly Constituency (AC).

House Ownership Coefficients: -9.27 to 8.45 Expect for parts of North and North-West Delhi, which show rise in water levels, other ACs show a fall in water levels with increase in share of HHs living in a house they own. Groundwater Recharge Coefficients: -0.31 to 0.43 A unit increase in groundwater recharge, i.e. an increase of 1 hectare-meter of groundwater recharge, cause a miniscule change of less than 2 cm for all of Delhi

Groundwater level change between 2006-16 Aggregating these values to each AC boundary, mean groundwater levels were obtained, which were used as the dependent variable in GWR model

-4m or less -4m to -2m -2m to 0m 0m to 2m

-4 or less -4 to -2 -2 to 0

Water source within Premises Coefficients: -4.81 to 12.61 ACs of East and South Delhi show a rise in water levels Peripheral ACs of East and North Delhi show a large fall.

0 to 2 2 to 4

Groundwater Wells Unplanned Assembly Constituencies Planned Assembly Constituencies

4 to 6 6 or greater

Use of Tubewells as main source Coefficients: -4.15 to 0.40 Eastern parts of Delhi, show a little rise in water levels i.e. maximum of 2 cm rise with 10% more HHs using piped water.

Car Ownership Coefficients: -5.90 to +8.4 Car ownership is used as a proxy for affluent HHs, which either have piped water connections or can pay for borewell installation.