Vegetation condition, Credo Station, Great Western Woodlands, WA by Richard Thackway

Tracking change and trend in vegetation condition at selected sites on Credo station, Great Western Woodlands

Richard Thackway Report prepared for the Western Australian Department of Environment and Conservation, Kalgoorlie and the CSIRO Division of Ecosystem Sciences, Perth.

VAST Transformations Revised 23 September 2013

ISBN 9781876141660

Tracking change and trend in vegetation condition at selected sites on Credo station, Great Western Woodlands by

Richard Thackway

Postal address: VAST Transformations 5 Spowers Circuit Holder ACT 2611 Copies available from: VAST Transformations 5 Spowers Circuit Holder ACT 2611 Email: rthackway@netspeed.com.au Web: http://www.vasttransformations.com/ Preferred way to cite this publication: Thackway, R., (2013). Tracking change and trend in vegetation condition at selected sites on Credo Station, Great Western Woodlands. Report prepared for the Western Australian Department of Environment and Conservation, Kalgoorlie and the CSIRO Division of Ecosystem Sciences, Perth. Revised 23 September 2013. VAST Transformations, Canberra: Westerlund Eco Services, Rockingham, Western Australia, p36.

Cover aerial photo of Credo Station homestead, Kalgoorlie, WA. by Leslie Westerlund

ISBN # 9781876141660

Publisher: Leslie C. Westerlund: â&#x20AC;&#x153;Westerlund Eco Servicesâ&#x20AC;?: Rockingham, Western Australia. Email: leslie_westerlund@yahoo.com Revised 23 September 2013

Acknowledgements Numerous individuals contributed land management, biophysical and ecological information for documenting the historical record for Credo Station. Their interest in and support of the project is gratefully acknowledged; Jim Addison (DAF WA Kalgoorlie), Christopher Auricht (Auricht Projects Adelaide), Graeme Behn (WA DEC Perth), Robin Bowden (EGHS Kalgoorlie), Greg Brennan (DAF WA Geraldton), Shane Cridland, Neil Gibson (WA DEC Perth), David Gobbett (CSIRO Perth), Ron Hacker (DPI NSW Sydney), Jeff Halford (Glen Iris Farms Esperance), Judith Harvey (WA DEC Perth), Angas Hopkins (SEWPac Canberra ), Ian Kealley (WA DEC Kalgoorlie), Paul Novelly (DAF WA Perth), Hugh Pringle (ecosystem Management Understanding Alice Springs), Suzanne Prober (CSIRO Perth), Harry Recher (Sydney), (Philip Thomas DAF WA Perth), Stephen Van Leeuwen (WA DEC Perth), Ian Watson (CSIRO Townsville), Nigel Wessels (WA DEC Kalgoorlie), Leslie Westerlund (Credo Station), Colin Yates (WA DEC Perth), Katherine Zdunic (WA DEC Perth). The support of the Kalgoorlie office of the Western Australian Department of Environment and Conservation is also acknowledged. Funding for this project was supported from two sources: the Australian Supersite Network, part of the Australian Governmentâ&#x20AC;&#x2122;s Terrestrial Ecosystems Research Network (www.tern.gov.au), a research infrastructure facility established under the National Collaborative Research Infrastructure Strategy and Education Infrastructure Fund - Super Science Initiative - through the Department of Industry, Innovation, Science, Research and Tertiary Education; and by the Western Australian Department of Environment and Conservation. Nigel Wessels and Suzanne Prober provided helpful comments on an early draft of the report.

Foreword Almost all of Australia’s vegetated landscapes have been affected by changes in land-use and landmanagement practices. The continent is now a diverse mosaic of fragmented and modified native vegetation, and converted and replaced vegetation cover types. Compiling a record of the responses of plant communities to land-use changes can assist land managers make improvements in their practices to meet wider social, economic and environmental goals. However, documenting the historical and contemporary use and management of a site and assessing the effects on vegetation condition can be a complex task. The absence of a consistent approach to assist with reporting transformations of plant communities over space and time remains a source of contention – and even conflict – between those involved in conservation and protection, and those responsible for sustainable land use and management. Land managers work with, or change, the ecological function of areas of land using land management practices with the aim of deriving multiple ecosystem services e.g. clean water from catchments, high quality habitat for selected species, and/or increased timber production from forests. The process of managing or changing a vegetation condition state from one state to another over time is widely practiced across several land use types including grazing in the rangelands, harvesting native timber, converting native vegetation to crops and/or pasture. Safford et al. (2012, p320) note that Over the past few decades, the importance of the natural variability of environments and the biological systems they contain has been more formally recognized, particularly in the concept of the historical range variation (HRV) of ecological systems. The concept acknowledges that systems vary over time. By considering the history of these systems, one can determine a range of conditions that encompass this past variation. This range can then define an appropriate “comfort zone” for land managers. The assumptions are that (1) this historical variation is inherent to the natural dynamics of the system and its components; (2) consequently, the biota are evolutionarily adapted to environmental variation within this range; and (3) the term average condition that is relatively stable, at least over the period of history that is relevant to conservation and management. The concept has been widely applied, especially to the management of forest ecosystems …. Until recently in Australia, there has been no standardized national system for ecologically accounting for the effect of anthropogenic practices on vegetation condition over time. VAST-2 was developed to track relationships between land management interventions and their effects at the site level over time. The VAST-2 system is designed to assist researchers and land managers record, track and report trends and changes in the condition of native plant communities. The VAST-2 system combines a structured spatiotemporal literature review and stakeholder interviews with 22 vegetation transformation indicators to assess and report the outcomes/effects of land management interventions on native plant communities. Change is assessed relative to an unmodified reference state. The report documents the application of the VAST-2 system to four sites in Credo station, Great Western Woodlands, Western Australia.

Executive Summary This report documents the transformation of two plant communities between the early 1800s to 2012; Salmon Gum Eucalyptus woodland (VEG_ASSOC 468) and Salmon Gum Eucalyptus open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506) found at four sites on Credo Station. Use of vegetated landscapes changes them. Changes, at both the site and landscape scale, can include degradation, modification, conversion, fragmentation, restoration, regeneration and increased connectivity. Until recently there has been no standardised national system to account for human-induced changes of plant communities at the site level over time. VAST-2 system provides land managers and planners a nationally consistent approach for assessing and reporting vegetation condition at sites over time. This report applies the VAST-2 system for collecting, collating and analysing historical records of land management for their effects on rangeland condition found at Credo Station. Condition is scored using 22 indicators that cover three components: site regenerative capacity, vegetation structure and species composition. Critical to the approach are the responses of each plant community relative to a reference state and the interactions between long-term rainfall and land management. Results are presented graphically. The four sites reported below are somewhat representative of the history of use and management of Credo Stationâ&#x20AC;&#x2122;s broader patterns of vegetation. All of Credo Station has been subject to around 100 years of grazing flocks of sheep and herds of cattle. A small area around 20 ha was used for cropping between 1906 and 1965. While this report has examined only four sites of limited spatial extent it is reasonable to conclude that the majority of Credo Station has retained its native vegetation in a modified state (VAST class II i.e. 60-80% relative to a fully natural reference state) e.g. sites B (Chadwin Paddock) and C (Carnage Paddock). Areas which have been subject to less intensive timber harvesting and grazing pressure e.g. Gallah paddock and the TERN Supersite, the condition of the vegetation is classified as unmodified (VAST class I i.e. 80-100% relative to a fully natural reference state). It seems reasonable to conclude that greenstone areas which have not been subject to mining or intensive mineral exploration would also be classified as VAST I because these areas have shallow and skeletal soils which have not been intensively used by grazing animals, except in some cases as sheep camps. Areas that have historically been subject to intensive use and management e.g. cropping and have subsequently been allowed to naturally regenerate, e.g. site D; such areas are classified as transformed (VAST III i.e. 40-60% relative to a fully natural reference state). Such areas on Credo Station are very restricted in area. Small areas of Credo Station have been and are subject to total removal of native vegetation e.g. farm infrastructure (buildings, dams and roads) and mining. If these sites were assessed using the VAST-2 system they would probably have been classified as Removed replaced (VAST VI i.e. <1% relative to a fully natural reference state). Evidence from an evaluation of WARMS site from the perspective of pasture management and production indicates that the pastoral condition of these sites is improving over time. It seems reasonable to conclude that the majority of Credo Station like much of the rangelands in the Great Western Woodlands is improving at the site level in terms of vegetation condition relative to the condition experienced in the 1930s and 1960s. Such a trend is shown in four Credo Station vegetation transformation sites.

iii

Contrasted against this site level assessment of vegetation condition is the increasing concern over progressive fragmentation of the native vegetation at the landscape and regional levels associated with mining sector.

Table of contents Acknowledgements ................................................................................................................ i Forward ................................................................................................................................ ii Executive Summary .............................................................................................................. iii Table of contents ................................................................................................................... v INTRODUCTION ..................................................................................................................... 1 METHOD ............................................................................................................................... 2 Selection of sites ............................................................................................................................................ 2 Data curation and publishing......................................................................................................................... 5 Analytical method – VAST-2 .......................................................................................................................... 6 Rainfall information ....................................................................................................................................... 9 RESULTS .............................................................................................................................. 13 Transformation of salmon gum woodland (VEG_ASSOC 468) (Site A) ......................................................... 14 Transformation of salmon gum woodland (VEG_ASSOC 468) (Site B) ......................................................... 16 Transformation of almost treeless bluebush /saltbush shrubland a variant of Salmon Gum Woodland (VEG_ASSOC 468) (Site C) .......................................................................................................... 19 Transformation of Eucalyptus open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506) (site D) ................................................................................................................................................. 21 DISCUSSION......................................................................................................................... 23 General synopsis - sites and landscape........................................................................................................ 23 Pre-pastoral activities .............................................................................................................................. 23 Grazing native pastures -Sites A, B and C ................................................................................................ 23 Recovery from cropping – Site D ............................................................................................................. 25 Post- pastoral activities ........................................................................................................................... 25 Interpreting change and effect ...................................................................................................................... 25 Resilience of rangeland ecosystems ............................................................................................................... 26 Benefits for land managers .......................................................................................................................... 27 Scaling up to the landscape level ................................................................................................................. 27 Site to regional levels .................................................................................................................................... 28 CONCLUSIONS ..................................................................................................................... 28 References .......................................................................................................................... 29 Appendix A.......................................................................................................................... 32 Appendix B .......................................................................................................................... 34

INTRODUCTION In Australia, a general conclusion of numerous ecologists is that in 200 years we have markedly changed many vegetated landscapes (e.g. Hobbs and Hopkins 1990; Kirkpatrick 1994; McIntyre and Hobbs 1999; Cocks 2000; Lefroy et al. 2000). Land-management decisions over time, reflecting the goals, values and desired outcomes of land managers (Mutendeudzi and Thackway 2010) have resulted in changes to the vegetation structure, species composition and/or regenerative capacity of plant communities. Examples of goal-oriented management include farm forestry for carbon sequestration, manipulating vegetation structure and function to maximise water yield, removing rangeland cover for food or fibre production, enhancing rangeland habitat features for threatened species recovery, and restoring native vegetation cover to ameliorate soil erosion. The continent is now a diverse mosaic of modified native vegetation, ‘replaced’ vegetation cover types, and fragments of vegetation in ‘original’ condition (Thackway and Lesslie 2008). The condition of the vegetation can also be influenced by managing and regulating physical inputs, including water and nutrients, and biological parameters (Trudgill 1977; Maltby et al. 1999), for example, by managing fire or grazing pressure (Noss and Cooperrider 1994), and by planting nonindigenous species. The VAST (Vegetation Assets States and Transitions) framework uses information on the effects of such land management practices to define a continuum of modification ‘states’, with thresholds, to classify changes to Australia’s vegetation, relative to each plant community’s reference state (Thackway and Lesslie 2008). Our capacity, however, to track and monitor these changes in the integrity of ecological systems at sites and/or across landscapes is relatively poor (Trudgill 1977; Daily 1997; Resilience Alliance 2010). This reflects the difficulty of understanding and distinguishing spatiotemporal responses of complex ecological systems to natural processes from responses to human use and management. Development of systems to monitor, evaluate and report the responses of naturally vegetated systems to human use and management has been only piecemeal (Thackway and Lesslie 2008). Documenting historical and contemporary land management and assessing its effects on vegetation can be a complex task. The absence of a consistent national approach to assist with reporting transformations of plant communities over space and time remains a source of contention—and even conflict—between those involved in conservation and protection, and those responsible for sustainable land use and management (Thackway 2012a). Where practical methods have been developed to explain the interactions between ecological systems and human use, they tend to be narrowly implemented in particular socio-ecological settings including for example forestry, dry land agriculture, nature conservation, rangelands or wetlands. Generally these systems operate independently for monitoring, evaluating and reporting. Examples include state government rangeland management agency systems for use before and after logging operations; mining company systems for use before and after strip mining; and fire management organisation systems for use before and after control burns and/or rangeland wildfires. Explanations of these interactions are generally based on rigorous statistical and/or mathematical solutions, but there is no certainty that these proposed solutions provide an adequate description and understanding of socio-ecological patterns and processes, particularly when it involves the wider community (Zellmer et al. 2006). Arguably there is a need to develop and implement a consistent monitoring and reporting system that can be applied across any socioecological sector and scale, can handle qualitative and quantitative information and can explain the socio-ecological complexity (Resilience Alliance 2010). Curtis et al. (2003) note that progress toward sustainable natural resource management is hampered because diverse sources of information on the responses of the environment to management are not integrated. Numerous other authors (e.g. Hobbs and Hopkins 1990; McIntyre and Hobbs 1999; Thackway and Lesslie 2006, 2008; Resilience Alliance 2010) also argue that better integration is needed to adequately address the links between management intervention and ecosystem structure, composition and function.

Several standardised methods for ecological monitoring, accounting for the effect of anthropogenic practices on ecological systems over time have been proposed (e.g. Resilience Alliance 2010). These systems have been developed to compile, integrate and interpret information from a wide range of social and ecological sources. VAST-2 is one such system proposed by Thackway (for application of the system see Thackway 2012a, b; for a detailed scientific explanation see Thackway and Specht in prep.) The VAST-2 system aims to answer three questions at the site level (Thackway 2012b): 1. What is the condition of the native vegetation on my site relative to an accepted natural standard? 2. How can I assess the role of historic land management in changing the condition of the native vegetation on my site? 3. As a land manager, what can I do to change the condition of the native vegetation of my site? This report documents on the application of the VAST-2 system to native Eucalypt woodland/ open woodland and chenopod shrubland communities over time for selected sites on Credo Station. The relevance of this site information to decision makers faced with a range of issues affecting Credoâ&#x20AC;&#x2122;s landscapes including degradation, modification, conversion, fragmentation, restoration and regeneration, is discussed. The findings have the potential to improve our understanding of change and trend, allowing them to be tracked through time. With this better understanding progress towards more sustainable use and better management of our regional landscapes can be made. METHOD Selection of sites The four Credo Station sites were selected because of the availability and willingness of the current and previous land managers to work with the author to apply the VAST-2 handbook (Thackway 2012a). The four sites are described in Table 1. It should be noted the VAST-2 system has been applied to sites in numerous other ecological settings (http://aceas.org.au/portal/). The four sites depict a known gradient of land use and land management intensity. The least disturbed site is site A through to the most disturbed site, site D (Table 1).

Table 1. Description of the four Credo Station sites including plant community, location, current land tenure and overview of the major land cover changes since first contact. NB: Sites A, B and C has a bluebush and/or saltbush open chenopod shrubland Site

Reference plant community

Location defined using Google Earth

File name for accompanying file formats

Major land cover changes and land management practices

Eucalyptus woodland (VEG_ASSOC 468)

30°20'43.38"S; 120°43'30.08"E

Gallahsupersite_credo

Native eucalypt woodland / open chenopod shrubland > cattle grazing native vegetation > sheep grazing native vegetation > cattle grazing native vegetation > minimal use

Eucalyptus woodland (VEG_ASSOC 468)

30°20'43.38"S; 120°43'30.08"E

Chadwin_credo

Native eucalypt woodland/ open chenopod shrubland > broad scale timber harvesting > regrowth woodland and cattle grazing > regrowth woodland and sheep grazing > cattle grazing native vegetation > minimal use

Eucalyptus woodland (VEG_ASSOC 468)

30°26'3.01"S; 120°52'32.03"E

Carnage_credo

Native eucalypt woodland / open chenopod shrubland > cattle grazing native vegetation > sheep grazing native vegetation > cattle grazing native vegetation > minimal use

Eucalyptus open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506)

30°27'5.78"S; 120°50'56.12"E

Carnage_cultivati on_credo

Native eucalypt open woodland / open chenopod shrubland > rainfed cropping > irrigated cropping > sheep grazing native vegetation > cattle grazing native vegetation > minimal use

The location of the four sites is shown in Figure 1. Two plant communities are represented in these four sites (Table 2). Table 2 describes using the vegetation communities using NVIS classification system and other relevant information:  Sites A, B, C have are described as a medium height (10-30 m) woodland (10-30% foliage cover) which is dominated by two species Eucalyptus salmonophloia and Eucalyptus dundasii. The NVIS level 4 description is used hereafter to describe this plant community: Eucalyptus woodland (VEG_ASSOC 468). 

Site D has been are described as a medium height (10-30 m) open woodland (<10% foliage cover) with Eucalyptus salmonophloia as the dominant species in the upper storey and a ground layer of low height (0.5-1.0 m) open (10-30% foliage cover) chenopod shrubland, which is dominated by Maireana. The NVIS level 4 description is used hereafter to describe this plant community: Eucalyptus open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506).

It should be noted that the process of translating the original vegetation mapping (i.e. Beard) into the national vegetation classification system (i.e. NVIS) led to some inconsistencies (Judith Harvey pers comm). For example the Eucalyptus woodland (VEG_ASSOC 468) is described as having no understorey structural or floristic elements. In reality this is not correct. These gaps and mismatches have yet to be fully resolved in the WA statewide vegetation map. This is discussed further below. Table 2. Description of the two plant communities and associated information supplied by WA DEC from the geospatial vegetation database. Vegetation transformation sites VEG_TYPE STUCT_DESC

A, B, and C

4 Woodland other

FLOR_DESC

Wheatbelt; York gum, salmon gum etc. Eucalyptus loxophleba, E. salmonophloia. Goldfields; gimlet, redwood etc. E. salubris, E. oleosa. Riverine; rivergum E. camaldulensis. Tropical; messmate, woolybutt E. tetrodonta, E. mini 468 Medium woodland; salmon gum & goldfields blackbutt e8,13Mi 468.0 KUNUNULLING_468 Tree Woodland Eucalyptus woodland Eucalyptus woodland

43 Saltbush and/or bluebush with woodland or scattered tree Salmon gum & gimlet Atriplex spp., Maireana spp., Eucalyptus salmonophloia, E. salubris

VEG_ASSOC SOURCE_DES MAP_CODE SA_CODE SYSTEM_ASS NVIS L1 NVIS L2 NVIS L3 NVIS L4 NVIS L5 NVIS L6

U^Eucalyptus salmonophloia, Eucalyptus dundasii\tree\7\i U1+Eucalyptus salmonophloia, Eucalyptus dundasii\tree\7\i

506 Succulent steppe with woodland; salmon gum & bluebush e8Mi k2Ci 506.0 KUNUNULLING_506 Tree Open woodland Eucalyptus open woodland Eucalyptus open woodland / Maireana open chenopod shrubland U Eucalyptus salmonophloia\tree\7\rG Maireana sp.\chenopod\2\i U1 Eucalyptus salmonophloia\tree\7\r;G1 Maireana sp.\chenopod\2\i

Source: Judith Harvey DEC WA

Figure 1. Location of the four vegetation transformation sites in Credo Station plotted relative to the boundaries of the WA statewide vegetation map. This figure uses the NVIS 6 i.e. subassociations to show context. Appendix A presents a list of the vegetation communities for Credo Station. Source: Judith Harvey DEC WA Data curation and publishing A consistent file naming system is used for each site to comply with the guidelines required by ACEAS TERN for uploading vegetation transformation site data: Paddock name - station name - nearest rangeland monitoring sites (i.e. RMS);

A. B. C. D.

gallah-supersite_credo chadwin_credo carnage_credo cultivation_carnage_credo

Four file formats use this file name convention to compile, analyse, spatially locate and present the results for each vegetation transformation site (Table 3).

Table 3. TERN file formats for uploading vegetation transformation site data to the TERN data portals.

File accompanying this report

Purpose of the tool

Description

MS word (.doc)

Compilation template

Historic record of land use and management practices and their observed and recorded impacts on vegetation condition

MS Excel (.xls)

Assessment framework

An assessment of the impact of land use and management practices using 22 indicators, 10 attribute groups, three weighted components of vegetation condition and total vegetation status

Google earth (.kmz)

Geospatial tool

Location showing the centroid of the site.

Graphics file (.jpg)

Presentation of results

A graph depicting three time series representing the components of vegetation condition; 1. regenerative capacity, 2. vegetation structure and 3. species composition, and plus one time series representing the transformation of the vegetation community at a site

A metadata file will be prepared when this report and the accompanying file formats is accepted by the projectâ&#x20AC;&#x2122;s proponents. It is planned that all data for the four sites will be published on the TERN websites including http://aceas.org.au/portal/ and the data portal http://portal.tern.org.au/. Analytical method â&#x20AC;&#x201C; VAST-2 The VAST-2 system builds on commonly used site-based indicators of vegetation condition and landscape function (e.g. Noss 1990; Gibbons and Freudenberger 2006; Thackway and Lesslie 2008; Tongway and Ludwig 2011) to derive an historical record about changes due to management practices and their effects on the condition of native plant communities. VAST-2 uses three widely acknowledged components of vegetation condition (aggregated from 22 indicators): 1) site regenerative capacity or function (i.e. post fire, soil structure, soil hydrology, soil chemistry, soil biota and reproductive potential); 2) vegetation structure (i.e. height, cover and age classes); 3) species composition (i.e. functional groups and species richness) (Table 4).

Table 4. List of component attribute groups and indicators of vegetation condition scored at sites. Change is assessed relative to an assumed pre-European reference state. Indicators: level 1

Attribute groups: level 2

1. Area /size of fires

(a) Fire regime

Components: level 3

2. Number of fire starts 3. Soil surface water availability.

(b) Soil hydrology

4. Ground water availability 5. Depth of the A horizon

Regenerative capacity (c) Soil physical state

6 Soil structure 7. Nutrient stressâ&#x20AC;&#x201D;rundown (deficiency)

(d) Soil nutrient state

8. Nutrient stressâ&#x20AC;&#x201D;excess (toxicity) 9. Recyclers (vertebrate and invertebrate) responsible for maintaining soil porosity and nutrient recycling

(e) Soil biological state

10. Surface organic matter, soil crusts 11. Reproductive potential of overstorey structuring species

(f) Reproductive potential

12. Reproductive potential of understorey structuring species 13. Overstorey top height (mean)

(g) Vegetation structure overstorey

Vegetation structure

14. Overstorey foliage projective cover (mean) 15 Overstorey structural diversity (i.e. a diversity of age classes) 16. Understorey top height (mean)

(h) Vegetation structure understorey

17. Understorey ground cover (mean) 18. Understorey structural diversity (i.e. a diversity of age classes) 19. Densities of functional groups of overstorey species

(i) Species composition overstorey

Species composition

20. Relative number of overstorey species (richness) as a ratio of indigenous to non-indigenous species 21. Densities of functional groups of understorey species

(j) Species composition understorey

22. Relative number of understorey species (richness) as a ratio of indigenous to non-indigenous species

Assessment of change for each of the 22 indicators is relative to an assumed pre-European reference for each plant community (Thackway 2012a, b). Each indicator is treated equally and independently, and scored from 0 to 1. Indicators are hierarchically structured into attribute groups and assigned weighted scores. The above components (referred to in Step 6) are generated and finally a vegetation transformation index (vegetation status) for each year in a siteâ&#x20AC;&#x2122;s historical record (Fig. 2). The choice of indicators, their assignment to attribute groups, and the weighting of the components within the VAST-2 hierarchy reflects the expert knowledge of numerous experienced plant community ecologists (Thackway and Specht in prep.). The weighted components comprise regenerative capacity (55%), vegetation structure (27%) and species composition (18%).

Figure 2. General process for tracking changes in vegetation condition over time

Figure 2 summarises the procedure used to compile, analyse and interpret information for sites. The system compiles historical records starting with explorersâ&#x20AC;&#x2122; first contacts with indigenous peoples, continuing to present day conditions including metadata records; descriptive documents and spread sheets; Google earth and other sources of remote sensing; land use histories; several branches of ecological science; landscape, vegetation and restoration; and includes relevant and credible published sources of information, and unpublished reports as well as interviews with rangeland managers. Information on where, when, and what changes in land use occurred, which rangeland management practices were used and the observed and measured effects of these practices was compiled, standardised and sequenced chronologically. This process produces a continuous string of information about the cause and effect of changes in vegetation condition. The system then presents the degree of site modification graphically so that land managers can quickly understand and assimilate the information (Thackway 2012a). The following information was compiled and/or derived for each site: 1. Location of the transformation site 2. Location of the reference site and description of reference state 3. Historical record of the disturbed site: a. Year of observed and/or measured land use and land management practices b. Year of observed, measured and/or inferred effects of land management practices on the 22 indicators hierarchically structured within 10 vegetation attributes and three components of vegetation condition, that is regenerative capacity, vegetation structure and species composition. 4. Scores of the response of the rangeland community over time: a. Each of the 22 indicators was scored between 0 and 1, using increments of 0.1, for each year of the historical record, where 1 represents a nil impact and 0 represents a complete removal or elimination.

b. Indicator scores were aggregated and weighted to assess changes in the three components of vegetation condition: regenerative capacity, vegetation structure, species composition and the total site transformation indices for each year of the historical record. 5. Reliability scores were assigned to provide transparency by accounting for quality of information e.g. published versus unpublished sources. Expert elicitation and multi-criteria analysis (MCA) were used to enable collaborators from different disciplines to contribute observational and measurement information. MCA was used as a tool to take advantage of expert knowledge and stakeholder advice, and add this information as qualitative observations. Where quantitative measurements were available this information was used to replace qualitative observations. The transformation of each site was classified relative to the classes defined in the VAST framework (Thackway and Lesslie 2006, 2008), thus enabling the decision-maker to track the pathway of a site over time commencing from reference state (i.e. VAST class I– residual /unmodified = 80–100% of the reference state). Typically over time a site may progress through one or more of the following classes: For example VAST class II– modified = 60–80%, VAST class III– transformed = 40–60%, VAST class IV– replaced and adventive = 20–40%, VAST class V– replaced and managed = 0–20%, VAST class VI– replaced and removed = 0%. Over time a site may be observed to pass into and out of these classes more than once. Assessed sites were peer reviewed in consultation with local field ecologists. Final revised site information is then submitted for publication on the Terrestrial Ecosystem Research Network’s (TERN) Australian Centre for Ecological Analysis and Synthesis (ACEAS) Data Portal and the TERN Data Discovery portal. All data for the four sites will be published on the TERN websites including http://aceas.org.au/portal/ and the data portal http://portal.tern.org.au/, for example Thackway (2012c). Rainfall information The monthly modelled rainfall data was obtained from http://www.longpaddock.qld.gov.au/silo/ for two localities Credo Station homestead and the township of Menzies. The monthly modelled data were grouped into four seasons e.g. Credo Station Autumn (Mar-Apr-May). Each season was graphed and a two year running trend line fitted (Figures 3 and 4). Seasonal average rainfall anomaly graphs were interpreted to identify and document the following information:  very high and very rainfall events above and/or below the average = 0  consecutive seasons over years where the rainfall is consistently above and/or below the average. Above average runs of years coincide major periods of regeneration and germination. Below average runs of years coincide major periods of droughts  long term changes between decades  consecutive years of ‘average years’ This information provided a context for the author to ask questions of land managers and ecologists about the pattern and seasonality of rainfall and its influence on land management practices including:  do you have good, medium and poor country and how do these respond to good and bad seasons and years?  what was the carrying capacity of particular paddock/s over the years 1900 to present?  what stock numbers are available over the years 1900 to present?  what is the fodder availability (quantity, type and quality) over the years 1900 to present?

 

   

when did the major germination events occur (e.g. Austrostipa) and when there was extensive bare ground? what were the observed consequences of major rainfall events following a long drought (e.g. converting from cattle to sheep in the 1920s because sheep did not require as much water and because there was very little in the way of reliable surface dams)? Why was Credo purchased when it was i.e. middle-end of a long drought? when did investment in infrastructure occur and why e.g. larger surface area dams and deeper dams were built following a couple of good years preceded by a string of bad years? what strategies does the land manager have for ‘seeing through’ a drought e.g. destocking sheep to Esperance, selling breeders, selling wethers, selling lambs? what use are the rangeland condition sites and are their condition states linked to rainfall and stocking rates? what is the link if any between antecedent rainfall events and the extent, severity, frequency of wildfire in good, medium and poor country?

Rainfall information was not provided to the land managers at least in the first instance as this was seen by the author to unduly lead them. This enabled land managers to check their own rainfall records. Experience of the author has shown that the modelled rainfall seasonal information in most instances usually follows very similar patterns to those records kept by land managers.

Figure 3. Credo Station average seasonal rainfall showing anomalies above and below the mean with a two year running trend line fitted. Rainfall data are derived from a modelled national 0.5 degree grid.

Figure 4. Town of Menzies average seasonal rainfall showing anomalies above and below the mean with a two year running trend line fitted. Rainfall data are derived from a modelled national 0.5 degree grid.

RESULTS Before the results for the four sites are presented the following explanatory notes are provided to assist the reader. It should be noted that the reference state for each of the site-based plant community’s is 100%. This total comprises three weighted components (indices) of vegetation condition: Regenerative capacity = 55%; Vegetation structure = 27%; and Species composition = 18%. Depending on the severity and duration of land management practices and how they have impacted the indicators of one of more of the components, the indices are reduced over time. Detailed findings underpin the four Credo Station vegetation transformation sites, these include. 16 data files that accompany this report: Historical record

Analysis score and weighting

Geolocation of the site in Google earth

Graph of vegetation transformation

1. gallah-supersite_credo

MSWord

MSExcel

.kmz

.jpg

2. chadwin_credo

MSWord

MSExcel

.kmz

.jpg

3. carnage_credo

MSWord

MSExcel

.kmz

.jpg

4. cultivation_carnage_credo

MSWord

MSExcel

.kmz

.jpg



each site’s MSWord file presents a description of several characteristics of the site including: a brief description of the natural undisturbed ecosystem of the site/area (Section 2); the reference state for this community (Section 4), the location and bioregional context of the site (Section 5); a brief history of the site/area (Section 7); and the sources of data and information used to complete description of use and management and their effects on native vegetation over time (Section 9);



each site’s MSExcel file presents several key results: scores of the 22 indicators and weighted indices for the three components of vegetation condition; (regenerative capacity; vegetation structure; and species composition);



each site’s Google earth kmz file presents the geo-location of the site;



each site’s jpg file, reproduced in Figure 5a to 5d presents a summary graphic comprising the three components of vegetation condition; (regenerative capacity; vegetation structure; and species composition) and changes in the vegetation status over time. The vegetation status in each figure is compared to the six VAST classes (Thackway and Lesslie 2008) to provide a useful standard for describing the condition classes .i.e. colour bar in each figure. VAST class I– residual /unmodified = 80–100% of the reference state, VAST class II– modified = 60–80%, VAST class III– transformed = 40–60%, VAST class IV– replaced and adventive = 20–40%, VAST class V–VI – replaced and managed = 1–20% and VAST class VI– replaced and removed = 0%. Figure 5a to 5d also show the phases in the transformation of the four Great Western Woodland (GWW) sites on Credo Station representing various combinations of

management, modification, replacement, removal and recovery in two different rangeland communities. These pathways are described below. Transformation of salmon gum woodland (VEG_ASSOC 468) (Site A) The indigenous plant community for site A, Gallah paddock and Supersite, is a Salmon gum woodland (VEG_ASSOC 468) (Figure 1). Four phases describe the transformation of this site (Figure 5a). These pathways are described below.

Figure 5a. Phases in the transformation of the Galah paddock, TERN Supersite. Reference state Salmon gum woodland (VEG_ASSOC 468); site A Figure 1). Phase 1: between 1864 and 1906 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included displacement of indigenous people, traverses by explorers and early establishment of pastoralism with shepherds. Gold prospecting and small scale gold mines were prevalent in the district. Small scale timber harvesting for local mining activities would not have obviously affected the overstorey vegetation structure, species composition and regenerative capacity. Use and management of the site had no observed effects on indicators of regenerative capacity, vegetation structure and species composition.

Phase 2: between 1906 and 1936 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included establishing boundary and paddock fences and the commencement of continuous stocking with cattle initially overseen from Daveyhurst Station. Rabbits arrived in the district from the south east. Sheep replaced cattle from Credo Station in 1925. Water storages, including relatively small tanks and dams, were built with horse drawn implements. These provided only intermittent water supplies in average and good rainfall years. There was a steady decline in gold mining activity in the district. In the latter part of this phase very high sheep numbers were present in the region. A major drought, 1932-37, resulted in the deaths of large numbers of sheep in the region. It is likely that Credo station was also carrying high sheep numbers and suffered a similar fate. Grazing in the early and mid-parts of this phase, initially with cattle (1907-1923), and then with sheep, resulted in minimal effects on indicators of regenerative capacity, vegetation structure and species composition. This was because of the low stock numbers when this paddock was established and the absence of enduring surface water in prolonged dry period and droughts. As sheep numbers increased in the latter part of this phase, two indicators of regenerative capacity were impacted; the loss of surface soil (A0 horizon) through trampling, sheet and rill erosion and a reduction in the infiltration of surface water. Surface soil organic matter would also have been reduced. Grazing also resulted in the reduction of structural diversity and change in the species functional traits of the understorey. The magnitude of these effects was minimal, with sheep removing palatable species, particularly Austrostipa and the smaller and younger regenerating saltbush and bluebush. Small amount of timber harvested for local mining operations and fence construction would not have obviously affected the overstorey vegetation structure, species composition and regenerative capacity during this phase. Phase 3: between 1936 and 2007 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included the establishment and development of larger more permanent surface water storages (tanks and dams) and the installation of polyethylene piping and troughs. These practices had the effect of spreading the grazing pressure more evenly over the larger areas away from the immediate dams and tanks and also in enabling larger numbers of stock to be carried longer into drier periods. In the period from 1960 to 1990, at the onset of prolonged droughts, young and breeding stock was transported to a family property located near Esperance. During these droughts, older stock and weathers were sold rather than carrying large number of sheep into and through droughts. From 1990-2007 cattle replaced sheep. Use and management of the site during this phase had minimal effects on indicators of regenerative capacity, vegetation structure and species composition. Sheep numbers across Credo ranged from 5,600 (1973) to 14,500 (1967) depending on the quality of the season and severity of droughts. Grazing with sheep (1936-90), and subsequently with cattle (1990-2007), resulted in minimal effects on indicators of regenerative capacity, vegetation structure and species composition. The construction of larger and deeper dams combined with the installation of polyethylene piping and troughs led to a stable vegetation state and a gradual improvement in the condition of the site from the low in shown in Fig 5a in the mid-1970s. However, the Supersite is located in close proximity to Charlie Dam which was not fenced from stock during this phase. As a result continued grazing impacts are expected. Two indicators of

regenerative capacity continued to be impacted from Phase 2, the loss of surface soil (A 0 horizon) through trampling, sheet and rill erosion and a reduction in the infiltration of surface water. Surface soil organic matter would also have been reduced. Ongoing grazing resulted in low structural diversity and changed in the species functional traits of the understorey. The magnitude of these effects was relatively minimal, with sheep removing palatable species, particularly Austrostipa and the smaller and younger regenerating saltbush and bluebush. The overstorey vegetation structure, species composition and regenerative capacity were scored as unaffected throughout this phase. Phase 4: between 2007 and 2012 the vegetation status of the site remained in the unmodified class (i.e. 80–100% VAST I). This phase included the purchase of Credo Station as a proposed conservation reserve, destocking of cattle, closing of Charlie Dam and decommissioning of polyethylene piping and troughs, establishment of the TERN Supersite and the commencement of a systematic biophysical monitoring. There has not been sufficient duration of time since the cessation of sustained grazing of cattle in Phase 3 to observe any noticeable improvements in the 22 indicators and the corresponding indices of vegetation condition described in Phase 2 above. Nevertheless, the current vegetation status (i.e. 88% of the reference state) is the highest index compared to the other three Credo Station vegetation transformation sites. For this reason it has the highest ecological integrity or resilience. The current gently increasing trend would suggest a continuation of the current trajectory or where there is a return to average or above average rainfall a possible gradual increasing recovery. Summary The Gallah paddock and Supersite (Site A) consistently had high indices for total vegetation status across all four phases i.e. 80-100% range, which is equivalent to VAST class I (Thackway and Lesslie 2008):  natural regenerative capacity unmodified,  structural integrity of native vegetation community is very high, and  compositional integrity of native vegetation community is very high. The current vegetation status indices in Fig 5a are the highest of the four sites presented in this report i.e.:   

Regenerative capacity = 47% compared to 55% reference state Vegetation structure = 25% compared to 27% reference state Species composition = 16% compared to 18% reference state

Transformation of salmon gum woodland (VEG_ASSOC 468) (Site B) The indigenous plant community for site B, Chadwin paddock, is Salmon gum woodland (VEG_ASSOC 468) (Figure 1). Five phases describe the transformation of this site (Figure 5b). These pathways are described below. Phase 1: between 1864 and 1895 the vegetation status of the site remained in the unmodified class (i.e. 80–100% VAST I).

This phase included displacement of indigenous people, traverses by explorers and early establishment of pastoralism with shepherds. Gold prospecting and small scale gold mines were prevalent in the district. Small amounts of timber that harvested for local mining operations would have not affected the overstorey vegetation structure, species composition and regenerative capacity during this phase. Use and management of the site had no observed effects on indicators of regenerative capacity, vegetation structure and species composition.

Figure 5b. Phases in the transformation of the Chadwin paddock. Reference state Salmon gum woodland (VEG_ASSOC 468); site B Figure 1). Phase 2: between 1895 and 1905 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase is characterised by extensive timber harvesting to supply the gold smelters in Kalgoorlie. All healthy trees were cut and removed from the site using timber lines. Use and management had an obvious impact on the overstorey indicators of vegetation structure and a minimal impact on indicators of regenerative capacity and species composition. Phase 3: between 1905 and 1931 the vegetation status of the site was in transition from an unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I) to a modified class (i.e. VAST 60-80% VAST II) This phase included establishing boundary and paddock fences and the commencement of continuous stocking with cattle. Rabbits arrived in the district from the south east. Sheep replaced cattle in 1925. Water storages, including relatively small tanks and dams, were built with horse drawn implements. These provided only intermittent water supplies in average and good rainfall

years. There was a steady decline in gold mining activity in the district. In the latter part of this phase very high sheep numbers were present in the region. Immediately following the harvesting of the timber there was minimal pressures from grazing with domestic animals, sheep or cattle. However grazing with cattle commenced in 1906-07 and continued until 1923. This effect would have placed some pressure on the regenerating overstorey, although at that time there were initially low stock numbers when this paddock was established and the absence of enduring surface water in prolonged dry period and droughts. As sheep numbers rapidly increased in the latter part of this phase, five indicators of regenerative capacity were impacted: the loss of surface soil (A0 horizon) through trampling, sheet and rill erosion; loss of surface soil organic matter; a reduction in the infiltration of surface water; a reduction in the organic recycling of organic matter; and a reduction in the recycling of organic matter. Grazing also resulted in the reduction of structural diversity and change in the species functional traits of the understorey. The magnitude of these effects was moderate, with sheep removing palatable species, particularly Austrostipa and the smaller and younger regenerating saltbush and bluebush. Phase 4: between 1931 and 2007 the vegetation status of the site was in transition from a modified class (i.e. VAST 60-80% VAST II) to an unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included the establishment and development of larger more permanent surface water holder storages (tanks and dams) and the installation of polyethylene piping and troughs. A 1.0 km channel was dug immediately to the north of Sixty One Dam to capture and divert water from one intermittent creek into another creek, thus enlarging the catchment of that Dam. These practices had the effect of spreading the grazing pressure more evenly over the larger areas away from the immediate dams and tanks and also in enabling larger numbers of stock to be carried longer into drier periods. In the period from 1960 to 1990, at the onset of prolonged droughts, young and breeding stock was transported to a family property located near Esperance. During these droughts older stock and weathers were sold rather than carrying large number of sheep into and through droughts. From 1990-2007 cattle replaced sheep. In the early part of this phase, a major drought, 1932-37, resulted in the deaths of large numbers of sheep in the region. It is likely that Credo station was also carrying high sheep numbers. Improved water point management of Chadwin paddock during this phase lead to moderate increases in the index values for vegetation structure. This recovery coincided with the regeneration of the overstorey and with spreading the grazing pressure due to improved pasture utilisation. Associated with improved range management a small and gradual increase can be observed in the regenerative capacity. No obvious improvement was observed in species composition to the level observed in phase 2. Phase 5: between 2007-12, the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included the purchase of Credo Station as a proposed conservation reserve, destocking of cattle and decommissioning of polyethylene piping and troughs. There has not been sufficient duration of time since the cessation of sustained grazing of cattle in Phase 4 to observe any noticeable improvements in the 22 indicators. Nevertheless, the current vegetation status (i.e. 87% of the reference state) is the high. For this reason it has the second highest ecological integrity or resilience. The current tapering-off of the trend in the total vegetation status in this phase (Fig 5b) would suggest either a slowing rate of recovery in the current trajectory. However, if there was a return

to average or above average rainfall the recovery is expected to return to that observed prior to the Millennium drought. Transformation of almost treeless bluebush /saltbush shrubland a variant of Salmon Gum Woodland (VEG_ASSOC 468) (Site C) The indigenous plant community for site A, Carnage paddock, is Salmon gum woodland (VEG_ASSOC 468) (Figure 1). Information compiled in the historic record for this site indicates that this site was almost naturally treeless when the Halford family took up the lease in 1905-06. Four phases describe the transformation of this site (Figure 5c). These pathways are described below.

Figure 5c. Phases in the transformation of the Carnage paddock. Reference state Salmon gum woodland (VEG_ASSOC 468); site C Figure 1). Phase 1: between 1864 and 1905 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included displacement of indigenous people, traverses by explorers and early establishment of pastoralism with shepherds. Gold prospecting and small scale gold mines were prevalent in the district. Use and management of the site had no observed effects on indicators of regenerative capacity, vegetation structure and species composition. Phase 2: between 1905 and 1931 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I).

This phase included establishing boundary and paddock fences soon followed by the commencement of continuous stocking with cattle. This paddock was recognised as having a naturally high level of pastoral productivity. At the commencement of the pastoral lease it was used for grazing cows and calves and later ewes and lambs. Grazing, initially with cattle (19071923), and then with sheep (1925-31), saw this paddock maintaining a higher numbers of stock than salmon gum country further to the north in Credo Station. The proximity of this paddock to permanent surface water in Rowles Lagoon enabled larger numbers of cattle or sheep to be kept on the land for longer in dry periods and in prolonged droughts. In addition, this paddock was the holding or â&#x20AC;&#x2DC;bulk-upâ&#x20AC;&#x2122; paddock for sheep mustered across Credo prior to the sheep being moved to the sheering sheep located at Credo Station homestead. Early in this phase, rabbits arrived in the district from the south east. Initially it was feared that rabbits would cause major environmental damage as they had in the rangelands of News South Wales, South Australia, Victoria and the south west of Queensland. Despite judicious preparations made to halt and/or limit the impact of rabbits, including erecting rabbit proof fences and purchasing rabbit warren fumigation equipment (stored at the Credo Station shearing shed), rabbit numbers did not build-up nor did they cause major environmental damage. In the latter part of this phase there was a steady decline in gold mining activity in the district and sheep number had reached very high numbers across the region. The impacts of high numbers of continuously grazing by cattle and sheep resulted in steep decline in several indicators of regenerative capacity, vegetation structure and species composition. Through grazing, bowsing and trampling three indicators of regenerative capacity were impacted resulted in; decreases in recycling of organic matter into the soil, decreases in the surface soil organic matter and the persistence of more bare soil for longer, and decreases in the reproductive potential of the shrub and ground layer. Ongoing intensive grazing and browsing also resulted in the reduction of structural diversity and change in the species functional traits of the understorey. The magnitude of these effects was moderate, with sheep removing palatable species, particularly Austrostipa and the smaller and younger regenerating saltbush and bluebush. As a consequence hardier shrubs thrived and short lived species occupied the inter-plant spaces between the chenopod shrubs. Isolated and scattered tree stumps have been recorded in the paddock among the shrub dominated vegetation. It is likely that these trees were harvested for fence construction. High grazing pressure would have prevented the overstorey species from re-establishing. Phase 3: between 1931 and 2007 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). A major drought, 1932-37, resulted in the deaths of large numbers of sheep in the region. It is likely that Credo station was also carrying high sheep numbers. This phase included the establishment and development of larger more permanent surface water holder storages (tanks and dams) and the installation of polyethylene piping and troughs. These practices had the effect of spreading the grazing pressure more evenly over the larger areas away from the immediate dams and tanks and also in enabling larger numbers of stock to be carried longer into drier periods. In the period from 1960 to 1990, at the onset of prolonged droughts (e.g. 1968-74), young and breeding stock was transported to a family property located near Esperance. During these droughts older stock and weathers were sold rather than carrying large number of sheep into and through droughts. From 1990-2007 cattle replaced sheep. The development of improved in water point management resulted in more even grazing pressure, and the removal of stock during prolonged dry periods and droughts led to a gradual and slight improvement in the condition of the several indicators of vegetation structure including the height, foliage cover and structural diversity of the shrub layer from 1932-2007.

Phase 4: between 2007 and 2012 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I). This phase included the purchase of Credo Station as a proposed conservation reserve, destocking of cattle and decommissioning of polyethylene piping and troughs. There has not been sufficient duration of time since the cessation of sustained grazing of cattle in Phase 4 to observe any noticeable improvements in the 22 indicators. Nevertheless, the current vegetation status (i.e. 83% of the reference state) is high. For this reason it has the third highest ecological integrity or resilience. Transformation of Eucalyptus open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506) (site D) The indigenous plant community for site A, Carnage paddock, is Salmon gum open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506) (Figure 1). Four phases describe the transformation of this site (Figure 5d). These pathways are described below.

Figure 5d. Phases in the transformation of the Carnage cultivation paddock. Eucalyptus open woodland / Maireana open chenopod shrubland (VEG_ASSOC 506); site D Figure 1). Phase 1: between 1864 and 1906 the vegetation status of the site remained in the unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I).

This phase included displacement of indigenous people, traverses by explorers and early establishment of pastoralism with shepherds. Gold prospecting and small scale gold mines were prevalent in the district. Use and management of the site had no observed effects on indicators of regenerative capacity, vegetation structure and species composition. Phase 2: between 1906 and 1965 the vegetation status of the site went through a very major transformation, transitioning from an unmodified class (i.e. 80â&#x20AC;&#x201C;100% VAST I), modified class (i.e. VAST 60-80% VAST II), transformed .class (VAST 40-60% VAST III), to adventive class (VAST 20-40% VAST IV) in the early part of this phase. This phase included fencing (40 acres or 18 ha paddock) and then using draught horses and hand axe ringbarking to clear of all trees and shrubs. Felled trees and shrubs were heaped and burned. The ground was ploughed and sown to wheat using horse-drawn ploughs and sown by hand and later mechanical seeders. Wheat was harvested using horse-drawn harvesters. Harvested wheat was fed to working farm animals. In the 1940s onwards horses were replaced for tractors. Rainfed cropping was regularly practiced when there was enough soil moisture. This involved ploughing several times to reduce the burden of regenerating grasses and shrubs before the ground was sown to wheat. When the soil moisture was too low the paddock was opened up and cattle or sheep were allowed to freely graze stubble and any regenerating native grasses and shrubs. From 1954-65 the paddock was furrow irrigated with water pumped from Rowles Lagoon. This practice was ceased in 1966. The effects of these land management practices had major effects on most indicators of regenerative capacity, vegetation structure and species composition. Effectively these components were prevented from functioning by using intensive soil management that repeatedly removed regenerating grasses and shrubs from the site. In terms of regenerative capacity, all regenerating native vegetation was actively removed, thereby obviously diminishing the reproductive potential of the overstorey and understorey. Other indicators of regenerative capacity were obviously affected but to a lesser extent, including soil hydrology, soil structure and soil nutrients. A common feature of the soil surface was bare ground due to the loss of indigenous soil organic matter and frequently occurring dry periods and droughts. All or most recyclers responsible for recycling soil organic matter were also obviously impacted by regular ploughing. Phase 3: between 1965 and 2007 the vegetation status of the site went through a moderately rapid transformation, transitioning from an adventive class (VAST 20-40% VAST IV) to a transformed class (VAST 40-60% VAST III). This phase included removing the paddock fences and the recommencing continuous stocking with initially with sheep (1965-1990) and then with cattle (1990-2007). No active revegetation practices were used to promote or encourage the regeneration of the former plant community. Windblown seeds naturally regenerated the site, forming a simplified plant community compared to the reference state. This transformation was observed in the gradual improvements in height, cover density and composition of the regenerating species. The process of regeneration was associated with relatively high numbers of stock in the paddock and the occurrence of regular average or above average rainfall events. Phase 4: between 2007 and 2012 the vegetation status of the site remained in a transformed class (VAST 40-60% VAST III).

There has not been sufficient duration of time since the cessation of sustained grazing of cattle in Phase 4 to observe any noticeable improvements in the 22 indicators. The current vegetation status (i.e. 47% of the reference state) is moderate. For this reason it has the lowest ecological integrity or resilience of the four Credo Station vegetation transformation sites. DISCUSSION "History" is the template on which the present is founded, but what we know of history is determined by human knowledge and the interpretation (or filtering) of past events. As we probe more deeply into the past, our knowledge becomes more fragmentary, uncertainty increases. Yet history holds precious clues to why things are the way they are today, and therefore how we might act to keep them that way or alter them, and how we might avoid the mistakes of the past. History provides essential lessons for the conservation and management of natural resources. (Romme et al. 2012, p4) The above results illustrate phases in the transformation pathways of four sites located on Credo Station, Great Western Woodland (GWW). Insights generated from the results can be used by decision-makers to answer the three questions posed at the start of the report. General synopsis - sites and landscape Pre-pastoral activities Prior to mining and pastoral use, the area, which is demarcated as Credo Station, formed part of the homelands of the Meru indigenous people. The largest extent of reliable fresh water was associated with Rowles Lagoon. Apart from this lagoon, surface water throughout the district was extremely rare and highly valued by the Meru language speakers. Exploration for gold preceded pastoral activities from around 1870-1920s. During this time 190515 eucalypt trees were extensively cut and collected into tram carriages over large areas of central and southern Credo. This area has been mapped and described as Eucalyptus woodland (VEG_ASSOC 468), Figure 1. Timber was extracted using tram lines known as ‘timber lines’. Harvested trees were transported by tram to Kalgoorlie where they were burnt as fuel wood in smelters to extract gold ore mined in or near Kalgoorlie. It is unlikely that fire was used to ‘clean up’ the waste timber, which was left on the ground to decay. Areas cut for fuel wood were allowed to naturally regenerate. Water for gold miners and tree cutters during this phase water would probably have been transported on the tram lines from Rowles Lagoon. During this phase local gold miners would have also harvested local trees for use in mine shafts and as pit props. It is unlikely that fire was used to ‘clean up’ the waste timber, which was left on the ground to decay. Areas cut for mine timbers were allowed to naturally regenerate. Grazing native pastures -Sites A, B and C Grazing followed gold exploration and mining. Credo station was first settled in the early 1900s by the Halford family who arrived from South Australia with cattle that they had walked overland. The Credo Station homestead was established on an intermittent creek to the south-west of Rowles Lagoon. The lagoon provided permanent fresh water for the small cattle herd. In these early years, 1906-09, cattle were overseen by shepherds, as there were no fences until around

1910. The station was progressively fenced, commencing in the south then moving north, using fence posts cut off the fence line or obtained from nearby. Over the time that Credo was managed as a pastoral station, from 1906-2008, there were three periods of rangeland management; cattle (1906-1922), sheep (1925-1991), and finally cattle (1991-2008). Credo was officially destocked in 2008 after it was purchased as a proposed future nature conservation reserve. Fire as a naturally occurring disturbance agent, or as a management tool used in managing the salmon gum woodlands of the Great Western Woodlands is rarely mentioned in the literature or by land managers. Dell et al. (1985) noted that presence of fire in salmon gum woodland is relatively rare i.e. 1 in 100 years. Under natural condition, it is considered that wildfire may occur in salmon gum woodlands after several years of above average rainfall and the build-up of large swards of Austrostipa. Pastures of Credo Station were managed with herds of cattle and flocks of sheep grazing native grasses (e.g. Austrostipa) and chenopod shrubs (saltbush and blue bush). Anecdotal evidence suggests that grass cover and biomass tended to be more abundant in the early years prior to around 1930s i.e. before Credo had developed an extensive system of water points; mainly dams and tanks, which enabled a higher stocking rate to be carried not only in average years but also longer into droughts. It has been suggested that the presence of these large areas of grasses is now a rare event, being observed only after extended wet periods where germination and establishment of grasses occurs more rapidly than can be eaten by sheep and cattle. The contemporary grass cover is apparently less dense and less extensive than expected prior to the 1930s. Further, anecdotal evidence suggests that since the 1930s flocks and herds have become largely dependent on grazing chenopod shrubs (saltbush and blue bush). While plausible this information was difficult to classify using the VAST-2 indicators, and as a result it was not used. What seems more plausible is that the Halford family over the years of pastoral management of Credo were conservative in their carrying capacity of flocks and herds with perhaps the exception of the major droughts in the 1937 and 1968. It would appear that many stock died in the 1937 drought, as they did over large areas of the Great Western Woodlands. This lesson led the Halford’s to develop an extensive system of large and deep dams and tanks followed in the 1990s by a network of polythene pipes and troughs to improve pasture utilisation. After the 1968 drought, the Halford’s purchased a property near Esperance which has been used to agist Credo’s breeders and young stock during prolonged dry periods and droughts. Figures 5.1-5.3 reflect this interpretation. These sites have either remained the same since the 1930s or have generally shown a gradual improvement in condition. Available evidence suggests that the Halford family understood the benefits of spreading grazing pressure associated with well-spaced reliable water supplies and pasture spelling through flock and herd agistment off Credo Station. The best, i.e. most productive, pastoral country on Credo Station is found to the east of Rowles Lagoon (Jeff Halford pers comm). This is comprised of almost treeless chenopod saltbush and bush shrublands gently rolling plains and depressions. During average and good years ewes and young sheep were kept on these pastures (1925-1991). Credo Station is characterised by a central diagonal ‘spine’ running north-west south-east comprising salmon gum woodlands on lower slopes and interspersed with greenstone rises with shallow skeletal soils. This country was regarded as less productive and for in the early years of pastoral development of Credo Station had limited reliable supplies of surface water. Compared to the almost treeless chenopod shrublands to the south Credo, paddocks in this central spine i.e.

Salmon gum and greenstone country, carried fewer sheep per hectare than in the south around the permanent water of Rowles Lagoon (Jeff Halford pers comm). Numerous Western Australia Rangeland Monitoring Sites (WARMS) were established on Credo Station commencing in 1988. While an invaluable record of change, it was challenging to ascertain the likely impacts of grazing on the 22 indicators for the three vegetation transformation sites i.e. sites A, B and C. An example of how the WARMS data was interpreted is shown in Appendix B (Ron Hacker pers comm). In addition, anecdotal evidence was provided to the author concerning the ongoing impacts of heavy stocking rates on soils associated with the severe droughts in the mid-1930s and late-1960s. It was suggested that soil surfaces on the mid and upper slopes of central and northern Credo had been eroded and become compacted. It was argued that this has contributed to increased run-off and leading to localised gully erosion or landscape etching. While plausible this information was difficult to classify using the VAST-2 indicators, and as a result it was not used. Recovery from cropping – Site D Credo had a small area of land set aside for cropping located to the south of Rowles Lagoon. Established in 1906 it was used as a wheat paddock, producing fodder for working farm animals. This area comprised two small paddocks of some 40 acres. Crops were sown when the seasons were favourable. In drought years the soil was left bare. When not cropped the paddock was ‘opened up’ to stock from the Carnage Paddock. The cropped areas were briefly irrigated from Rowles Lagoon. However, this ceased in mid-1960s. The fences surrounding the crop paddocks were removed and the areas were allowed to naturally regenerate from windblown seeds. Figure 5.4 shows a relatively rapid recovery of structure and composition although full recovery will take many years. Post- pastoral activities In 2007 Credo Station was purchased as a future conservation reserve. While the station was destocked in 2008, today small herds of feral cattle can be found near dams and tanks. In a field survey in 2012, the author noted that the majority of the chenopod shrublands on Credo Station were mature or over-mature evenly spaced of shrubs, which commonly showed evidence of dead rank stems. Presumably this was evidence of former heavy grazing and/or of drought. It was not possible to ascertain what the cause of shrub condition was. Perhaps of greater importance was that the author observed minimal evidence of different life stages of chenopod shrubs i.e. no young seedlings or immature shrubs. Perhaps this observation is premature given the short period since Credo was destocked and there has been insufficient time and good seasons of rains to enable broad scale germination and establishment. Interpreting change and effect Ideally, monitoring involves the regular collection of information over a number of years using the same methods of observation and measurement, but this has not happened in most places (Thackway 2012a). While the WARMS sites on Credo Station, established in 1988 have been monitored three times, the data is difficult to interpret using the VAST-2 indicators because their primary purpose is grazing management. Despite this Ian Watson and Hugh Pringle (person comm) both noted that there had been a general improvement in the condition of rangelands pastures monitored at

WARMS sites in the Great Western Woodlands and that this was attributed to improved stock management. This improvement would appear to be related to more sustainable pasture utilisation practices since the severe droughts of the 1930s and 1960s. In order to place the WARMS sites in a broader ecological context on Credo and in the Great Western Woodlands, arguably it is necessary to calibrate these sites to a broader range of ecological attributes, such as those found in VAST-2. One of the benefits of utilising the 22 indicators in the VAST-2 is it helps land managers and ecologists to raise questions about how use and management of the native vegetation is reflected in the response of the plant communities. Resilience of rangeland ecosystems Where land management practices have affected the physical character of a site then the site may or may not be able to recover. For example, if excessive fire frequency has allowed the soil to be removed down to bedrock then recovery would be possible only in geological timescales. If soil fertility levels have changed, e.g. decreased by constant cropping following conversion of rangeland cover, then it may be several decades before fertility levels return to pre-intervention levels (Trudgill 1977). Where land management practices have affected the regenerative capacity of the site—that is fire regime, soil health, reproductive potential of the overstorey and/or understorey—the resilience of the site is strongly affected. Such a site will recover more slowly from disturbance than will an unaffected site. In contrast, where land management practices have affected vegetation structure and/or species composition, then the effect on the rate of recovery is relatively less. VAST-2 provides a mechanism to track these three components of rangeland change, that is regenerative capacity, vegetation structure and species composition. Resilience Alliance (2010: p. 51) defines resilience as ‘The capacity of a system to absorb disturbances and reorganise while undergoing change so as to retain essentially the same function, structure, identity, and feedbacks’. VAST-2 provides a gradient of measured, observed or inferred levels of resilience of a site. Davidson et al. (2011) define ‘resilient lands’ as areas of native vegetation that conform to the definition of VAST Classes I, II or III. Areas with moderate to high resilience are amenable to prescribed natural regeneration (Clewell and McDonald 2009). Prescribed natural regeneration is intentionally-allowed natural regeneration, planned and managed for ecosystem restoration. This approach requires knowledge of resilience processes, and control of land management activities to encourage restoration. In areas where natural resilience has been significantly altered or removed, the capacity of the site to naturally stimulate the initiation of latent ecological regeneration processes is also removed; such sites conform to the definition of VAST Classes IV, V or VI. These areas are classified as ‘non- resilient lands’ and require ‘reconstruction’ (Davidson et al. 2010). McDonald (2000) and Tongway and Ludwig (2011) note that stimulating or re-establishing native vegetation on these ‘non-resilient lands’ requires ‘assisted regeneration’, that is management interventions (such as targeted weed control, log installation, and ⁄ or introduction of plant propagules by, say, direct seeding) to stimulate the initiation of latent natural processes. Resilience Alliance (2010) note that all sites exhibit dynamic responses reflecting the interaction between natural and anthropogenic processes, hence each resilience assessment is time bound. From the information in this paper the above four sites can be ordered from highest to lowest based on the current resilience; Figure 5.1 to 5.4 shows that site A has the highest resilience, through to site D with the lowest resilience. The above results help land managers and ecologists understand history of disturbance, the response of indicators and the current trend of site development. Such information is of key interest to a wide range of stakeholders. For example, people compiling the state of the rangelands (e.g. Bastin and the ACRIS Management Committee 2008), ‘State of the Forest’ and the

â&#x20AC;&#x2DC;State of the Environmentâ&#x20AC;&#x2122; reports could use it to monitor recovery in height, cover and/or age class following anthropogenic. Benefits for land managers The above results describe transformation pathways for four rangeland sites that have been subject to in excess of 100 years of management practices. This information is directly relevant to land managers, specifically those involved in replacement, removal or recovery of native vegetation, such as conservation managers (Wiens et al. 2012). Decision-makers can gain a deeper understanding of the main factors influencing a site through a closer examination of the three components of condition, 10 attribute groups and 22 indicators provided in the VAST 2 system. This information, along with an understanding of rainfall patterns (long-term average rainfall records are available from the Bureau of Meteorology) can be used by decision-makers to consider what land management options might be available to deliberately change a plant community to a desired condition class. Application of the system to the four sites in this report demonstrates its value to: o

o o o

integrate information coming from an extensive network of land managers, research scientists, and the wider community regarding land management and the responses of selected plant communities at sites over time provide a consistent national approach for reporting on changes in condition of native plant communities over time consistently describe and collate temporal patterns and processes associated with the use and management of Australiaâ&#x20AC;&#x2122;s vegetated landscapes, with regard to specific geolocations collate, review and revise a national repository of historical records of changes in land management and their effects on the condition of plant communities and to assess the suitability of this information to inform research priorities.

Close analysis of the VAST-2 data highlights the critical importance of tracking the interactions between historic rainfall and land management to the responses of each plant community as seen in the indicators and the components of vegetation condition (Figure 2). Understanding these interactions is critical to developing an ecological understanding of a site as well as change and trend. The VAST-2 system offers very useful insights into the effects of historic use and management on the condition of a site over time. The visual presentation of the results allows decision-makers to quickly understand and assimilate complex ecological processes and their effects on degradation, restoration and regeneration. The system provides a tool for identifying what component needs to be manipulated to improve vegetation condition, demonstrating progress toward the desired vegetation condition, and selecting sites which represent least-cost options for future land use changes. It also highlights the importance of an accounting system that can be used to track the sustainable use and management of native vegetation across all land use types and has relevance for managing biodiversity. Scaling up to the landscape level Monitoring, mapping and modelling changes in vegetation condition over time at landscape scales presents major challenges for spatial scientists. The VAST (Vegetation Assets, States and Transitions) framework (Thackway and Lesslie 2006 and 2008;) has been used to generate consistent national datasets for reporting changes in vegetation condition at the landscape scale (Lesslie et al. 2010). However without an enhancement VAST cannot track and routinely report changes in condition over time for the same site. Hence VAST-2 was developed out of a need to

implement a system to routinely tracking the causes of transitions between condition states. VAST is more spatially oriented whereas VAST-2 is more temporally oriented. The vegetation transformation sites in Credo Station show that similar land management practices were used across the station at the same time in the historical record e.g. wood-line timber harvesting and grazing. Broadly the same land use and management practices can be expected at the bioregion level in the historical record. This is also true today. A more comprehensive and representative set of VAST-2 sites sampled across the landscape than currently available would provide valuable insights into spatially extending land management practices and their observed effects on the components of vegetation condition across bioregions; vegetation structure, species composition and regenerative capacity or ecological function. More vegetation transformation sites with records of use and management and the observed and measured effects on indicators of regenerative capacity, vegetation structure and species composition are needed before models can be developed to map the transformation of Credo Station and the Great Western Woodlands. Without a strategic approach to the collection of this historical record at a broader landscape level we currently cannot with any confidence attribute dynamic changes in native vegetation observed in remote sensing imagery e.g. Landsat and MODIS dynamic land cover to changes in land management practices. Hence, there is a need for more comprehensive sampling at the landscape level sites that document cause and effect to underpin spatial and temporal extrapolation landscape changes to the landscape and bioregion levels. Site to regional levels Credo Station also sits in a broader regional context. To deal with the need to roam between site and regional levels a hierarchy of mapping scales has been developed encompassing the patch (within which sites may be sampled), the immediate surrounds and the broader landscape (Thackway and Lesslie 2008). Such an approach could be applied to the Great Western Woodlands. The ‘site’ and ‘immediate landscape’ context are described and mapped using a VAST spatial dataset (e.g. Lesslie et al. 2010) that is Intact (I), Modified (II), Transformed (III), ReplacedAdventive (IV), Replaced-Managed (VI), and Removed (VI). The broader landscape is described and mapped using a four landscape alteration levels dataset (Mutendeudzi and Thackway 2008), (i.e., Intact, Fragmented, Variegated, and Relictual). This broader landscape level is derived using VAST dataset (e.g. 100 m X 100 m raster) and GIS spatial analysis. Landscape alteration levels are defined as the proportions (i.e., % area) of each condition state measured in a 500 m moving window radius using FRAGSTATS software (Thackway and Lesslie 2008). Access to information for the three spatial levels would enable decision-makers to develop priorities and options for vegetation management and investment options. For example, least cost options can be developed by targeting land parcels with multiple patches of vegetation (i.e. VAST II and/or III) which are large in size, relatively well connected to other large patches, and least modified. Within the broader context (i.e. landscape alteration levels), the same parcels could be assessed to determine which are located in least fragmented and modified landscapes. CONCLUSIONS The process of documenting the transformation of plant communities using historic record of land use and management at sites and assessing the effects on vegetation condition need not be a complex task. More sites in the same two plant communities or in new plant communities could be similarly documented and assessed, if required. The process of compiling and assessing the four vegetation transformation on Credo Station, for two plant communities, engaged a wide range of collaborators and stakeholders including

indigenous people, ecologists, academics, land managers, environmental historians and remote sensing specialists. A diverse assemblage of information was synthesized and sequenced and evaluated to help build a consensus among these specialists regarding how best to use information within the constraints of a multi criteria analysis for two selected plant communities over time. Quantitative and qualitative data and information as well as personal communications were compiled and evaluated to determine changes and trends in the 22 indicators of vegetation condition. The completed Credo Station vegetation transformation graphs, comprising information on regenerative capacity, vegetation structure and species composition, help in ‘telling the story’ of how land use and management, has affected vegetation condition over time. These easy to understand graphs enable ecologists and land managers to explore options for future use and management of each site. This information is also useful for environmental planners, educators, historians, industry groups and the wider public. The VAST-2 system, that underpins the four sites, offers certainty for land managers and planners through providing a nationally consistent approach for assessing and reporting vegetation condition at sites. As a standardized tool, VAST-2 delivers ecologically meaningful information to assist decision makers to track and understand complex ecological processes including modification, degradation, restoration and regeneration. That system provides a framework for identifying potential risks and barriers to achieving success, demonstrating progress toward vegetation condition targets, and in selecting sites which represent least-cost options for future land use changes. It also highlights the importance of an accounting system that can be used to track the sustainable use and management of native vegetation across all land use types and has relevance for conservation and management of natural resources. Additional information about the VAST-2 system can be found at http://www.vasttransformations.com/ References The reader is referred to additional references included in the historical record for each site: see Section 9 Sources of data and information used to complete description of use and management and their effects on native vegetation over time in the four accompanying MSWord files. References in those documents have not been duplicated here because many are specific to a site. Bastin G and the ACRIS Management Committee (2008). Rangelands 2008 — Taking the Pulse, Published on behalf of the ACRIS Management Committee by the National Land & Water Resources Audit, Canberra. Cocks, D. (2000) Scenarios for Australian landscapes. In: Hamblin, A. (ed.) Visions of Future Landscapes. Proceedings of 1999 Australian Academy of Science Fenner Conference on the Environment, 2–5 May 1999. Bureau of Rural Sciences, Canberra. pp. 75–81. Curtis, A., Byron, I. and McDonald, S. (2003) Integrating spatially referenced social and biophysical data to explore landholder responses to dryland salinity in Australia. Journal of Environmental Management 68: 397–407. Clewell, A. and McDonald, T. (2009) Relevance of natural recovery to ecological restoration. Ecological Restoration 27: 122–123. Davidson, I., Sheahan, M. and Thackway, R. (2011) An innovative approach to local landscape restoration planning: lessons from practice. Ecological Management and Restoration 12: 175–188. http://onlinelibrary.wiley.com/doi/10.1111/j.1442-8903.2011.00607.x/pdf [accessed on 31 January 2013]. Gibbons, P. and Freudenberger, D. (2006) An overview of methods used to assess vegetation condition at the scale of the site. Ecological Management & Restoration 7: S10–S17. doi: 10.1111/j.1442-8903.2006.00286.x Hobbs, R.J. and Hopkins, A.J.M. (1990) From frontier to fragments: European impact on Australia’s vegetation. In: Saunders, D.A., Hopkins, A.J.M. and How, R.A. (eds) Australian Ecosystems:

200 Years of Utilization, Degradation and Reconstruction. Proceedings of the Ecological Society of Australia 16: 93–114. Kirkpatrick, J.B. (1994) A Continent Transformed: Human Impact on the Natural Vegetation of Australia. Meridian-Australian Geographical Perspectives. Oxford University Press. Lefroy, T., Hobbs, R.J. and Hatton, T., (2000) Effects of changing vegetation on hydrology and biodiversity. Pp. 38–51 in Hamblin, A. (ed.) Visions of Future Landscapes. Proceedings of the Australian Academy of Science 1999 Fenner Conference on the Environment. Bureau of Rural Sciences, Canberra. Pp. 38–52. Lesslie, R., Thackway, R. and Smith, J. (2010) A National-Level Vegetation Assets, States and Transitions (VAST) Dataset for Australia (version 2). Bureau of Rural Sciences, Canberra. Maltby, E., Holdgate, M., Acreman, M.C. and Weir, A. (1999) Ecosystem Management: Questions for Science and Society. Royal Holloway Institute for Environmental Research, University of London, Egham, U.K. McDonald T. (2000) Strategies for the ecological restoration of woodland plant communities: Harnessing natural resilience. Pages 286–297 in Hobbs, R.J and Yates, C. J. (eds) Temperate Eucalypt Woodlands in Australia: Biology, Conservation, Management and Restoration. Surrey Beatty and Sons, Chipping Norton. McIntyre, S. and Hobbs, R. (1999) A Framework for conceptualising human effects on landscapes and its relevance to management and research models, in Conservation Biology 13: 1282– 1292. Mutendeudzi, M. and Thackway R. (2010) A method for deriving maps of landscape alteration levels from vegetation condition datasets. Bureau of Rural Sciences, Canberra. Noss, R.F. (1990) Indicators for monitoring biodiversity: a hierarchical approach. Conservation Biology 4: 355–364. Noss, R.F. and Cooperrider, A.Y. (1994) Saving Nature’s Legacy—Protecting and Restoring Biodiversity. Island Press, Washington. Resilience Alliance (2010). Assessing resilience in social–ecological systems: Workbook for practitioners. Version 2.0. http://www.resalliance.org/3871.php [accessed on 31 January 2013]. Resilience Alliance (2010) Assessing Resilience in Social–Ecological systems: Workbook for Practitioners. Version 2.0. http://www.resalliance.org/3871.php [accessed on 31 January 2013]. Romme, W.H., Wiens, J.A., and Safford H.D., (2012). Setting the stage: theoretical and conceptual background of historical range of variation, pages 3-18. Chapter 1 in Wiens, J.A., Hayward, G.D. Safford, H.D., Giffen, C. (Eds), Historical Environmental Variation in Conservation and Natural Resource Management. Wiley-Blackwell, UK. 352 pp. Safford H.D., Wiens, J.A., and Hayward, G.D., (2012). The growing importance of the past in managing ecosystems of the future, pages 319-27. Chapter 24 in Wiens, J.A., Hayward, G.D. Safford, H.D., Giffen, C. (Eds), Historical Environmental Variation in Conservation and Natural Resource Management. Wiley-Blackwell, UK. 352 pp. Thackway, R. (2012a). Tracking the Transformation of Vegetated Landscapes, Handbook for recording site-based effects of land use and land management practices on the condition of native plant communities. Version 2.3, October 2012. Westerlund Eco Services, Rockingham, Western Australia, p. 56. Thackway, R. (2012b). Tracking the Transformation of Vegetated Landscapes using the VAST-2 system. VAST Transformations, Canberra. 8 pages. http://www.vasttransformations.com/#!vast-2-system/c4x [accessed on 28 January 2013]. Thackway, R. and Lesslie, R. (2006). Reporting vegetation condition using the Vegetation Assets, States and Transitions (VAST) Framework. Ecological Restoration and Management 7(1):S53–S62. Thackway, R. and Lesslie, R. (2008). Describing and mapping human-induced vegetation change in the Australian landscape. Environmental Management 42: 572–590. http://www.springerlink.com/content/w318w7221202v2v8/ Thackway, R. and Specht, A. (in prep). A system for capturing the effects of land use on vegetation condition.

Tongway, D. and Ludwig, J. (2011). Restoring Disturbed Landscapes: Landscape Ecology. Island Press, USA. 200 pp. Trudgill, S.T. (1977). Soil and Vegetation Systems Contemporary Problems in Ecology. Clarendon Press, Oxford. Zellmer, A.J., Allen, T.F.H. and Kesseboehmer, K. (2006). The complexity of ecological complexity: a protocol for building the narrative. Ecological Complexity 3: 171â&#x20AC;&#x201C;181. Wiens, J.A., Hayward, G.D. Safford, H.D., Giffen, C. (Eds), (2012). Historical Environmental Variation in Conservation and Natural Resource Management. Wiley-Blackwell, UK. 352 pp.

Appendix A List of vegetation communities shown in Figure 1. VEG_ ASSOC

STUCT_DESC

FLOR_DESC

SOURCE_DES

MAP_COD E

SA_COD E

SYSTEM_ASS

Woodland other

Medium woodland; goldfields blackbutt & red mallee

e13,22Mi

502.1

BARLEE_502

Woodland other

Medium woodland; salmon gum & gimlet

e8,34Mi

8.0

KUNUNULLIN G_8

Woodland other

Medium woodland; salmon gum & goldfields blackbutt

e8,13Mi

468.0

KUNUNULLIN G_468

Woodland other

Medium woodland; salmon gum

e8Mi

936.0

JACKSON_936

Low woodland, open low woodland or sparse woodland

Mulga Acacia aneura and associated species.

Low woodland; mulga & red mallee

a1e22Li

504.0

JAURDI_504

Low woodland, open low woodland or sparse woodland

Mulga Acacia aneura and associated species.

a1c2eLi

20.2

BARLEE_20

2901

Scrub with open woodland or scattered trees

Wattle with York gum, casuarina, mulga Acacia spp. with Eucalyptus loxophleba, Allocasuarina spp. Acacia aneura.

Low woodland; mulga mixed with Allocasuarina cristata & Eucalyptus sp. Mosaic: Medium woodland; Allocasuarina cristata & goldfields blackbutt Shrublands; Acacia quadrimarginea thicket

c2e13Mi/ a14Sc

2901.1

BROAD ARROW_2901

1413

Thicket

Wattle, casuarina and teatree acacia-allocasuarina-melaleuca alliance.

Shrublands; acacia, casuarina & melaleuca thicket

acmSc

1413.1

KUNUNULLIN G_1413

502

468

936

504

VEG_T YPE

1413

Thicket

Wattle, casuarina and teatree acacia-allocasuarina-melaleuca alliance.

Shrublands; acacia, casuarina & melaleuca thicket

acmSc

1413.0

BARLEE_1413

a33,34,35 Sc

435.4

JAURDI_435

a33,34,35 Sc

435.0

KUNUNULLIN G_435

435

Thicket

Wattle, casuarina and teatree acacia-allocasuarina-melaleuca alliance.

435

Thicket

Wattle, casuarina and teatree acacia-allocasuarina-melaleuca alliance.

Shrublands; Acacia neurophylla, A. beauverdiana & A. resinomarginea thicket Shrublands; Acacia neurophylla, A. beauverdiana & A. resinomarginea thicket

484

Thicket

Wattle, casuarina and teatree acacia-allocasuarina-melaleuca alliance.

Shrublands; jam thicket

a19Sc

484.0

BARLEE_484

538

Scrub, open scrub or sparse scrub

Wattle, teatree & other species Acacia spp. Melaleuca spp.

Shrublands; Acacia brachystachya scrub

a15Si

538.0

BARLEE_538

Scrub, open scrub or sparse scrub

Wattle, teatree & other species Acacia spp. Melaleuca spp.

anSi

40.0

KUNUNULLIN G_40

555

Shrub-steppe

Hummock grassland with scattered shrubs or mallee Triodia spp. Acacia spp., Grevillea spp. Eucalyptus spp

e22Si t8Hi

555.3

KUNUNULLIN G_555

483

Spinifex complexes

Hummock grassland with scattered low trees over dwarf shrubs or mixed short grass and spinifex mixed species, Triodia spp.

Shrublands; acacia scrub, various species Hummock grasslands, mallee steppe; red mallee over spinifex, Triodia scariosa Hummock grasslands, mixed sandplain - open mallee over sparse dwarf shrubs with spinifex ; red mallee mallee & mixed sparse dwarf shrubs over Triodia basedowii

e22Sr t2Hi xZp

483.1

BARLEE_483

Saltbush and/or bluebush with woodland or scattered tree

Salmon gum & gimlet Atriplex spp., Maireana spp., Eucalyptus salmonophloia, E. salubris

Succulent steppe with woodland; salmon gum & bluebush

e8Mi k2Ci

506.0

KUNUNULLIN G_506

529

Saltbush and/or bluebush with scattered low trees

Mulga, other wattle, casuarina Atriplex spp. Maireana spp. with Acacia aneura, A. papyrocarpa, Allocasuarina cristata

Succulent steppe with open low woodland; mulga & sheoak over bluebush

a1c2Lr k2Ci

529.2

BARLEE_529

389

Saltbush and/or bluebush with scattered low trees

Mulga, other wattle, casuarina Atriplex spp. Maireana spp. with Acacia aneura, A. papyrocarpa, Allocasuarina cristata

Succulent steppe with open low woodland; mulga over saltbush

a1Lr k1Ci

389.2

BARLEE_389

676 125 128

50 51 54

Samphire Salt lake, lagoon, clay pan Rock

Tecticornia spp. communities in saline areas

Succulent steppe; samphire Bare areas; salt lakes Bare areas; rock outcrops

k3Ci sl r

676.0 125.0 128.0

BARLEE_676 GENERAL GENERAL

506

Appendix B Credo Rangeland monitoring site 9 (Assessment by R B Hacker: 29 November 2012) Groups (10) Fire regime Soil hydrology Soil physical state Soil chemical state Soil biological state Reproductive potential Overstorey structure

Understorey structure

Overstorey composition Understorey composition

Indicators (22) Fire burnt area Number of burns Surface water Ground water Depth of A horizon Soil structure decline Soil nutrient rundown Soil nutrient excess or toxicity Soil invertebrate recyclers Soil organic matter Overstorey reproductive potential Understorey reproductive potential Overstorey top height Overstorey foliage projective cover Overstorey structural diversity Understorey top height Understorey ground cover Understorey structural diversity Overstorey species functional groups Overstorey species richness Understorey species functional groups Understorey species richness

Reference

Not relevant Not relevant Unmodified Unmodified Unmodified Unmodified Unmodified Unmodified Unmodified Unmodified Not relevant Unmodified Not relevant Not relevant Not relevant Unmodified Unmodified Unmodified Not relevant Not relevant Unmodified Unmodified

1988

1992

1998

2008

9.5 10 9.5 9.5 8.5 10 7 3

3/10

7/3 5/10 10

8.5/6 5/10 10

8/6 5/10 10

na (no photo) na (no photo) na (no photo)

4.5

5.0

Explanatory Notes Soil Hydrology: As far as can be seen the site has not been affected by water diverting earthworks – either above or below ground; There may be some increase in run-off relative to the benchmark but the site is relatively flat, low in the landscape and shows little evidence of accelerated water erosion. Soil Physical State: There is little evidence from the photos of any major wind scalding – perhaps a few minor patches; structural decline due to trampling will not be marked at the stock densities typical of the area. Soil Chemical State: Soil nutrient rundown. There are several interacting factors here. 1. A general depletion of nutrients due to animal offtake; 2. Loss of cryptogamic crusts under grazing which will result in lower levels of Total N and Organic C in the surface soil (perhaps only 0-2 cm); however this will not translate into an equivalent loss of nutrients for higher plants; 3. Loss of topsoil by wind scalding – probably the most important means by which nutrients are lost – there is little evidence of this on the site; 1

22 Indicators are defined in the Handbook at www.vasttransformations.com Reference is defined as ‘best on offer’ or unmodified or unchanged 3 Score change out of 10 using 0.1 increments. Where 10 = no change from reference, 0 = total removal of component or function relative to reference state. 2

4. Recycling of nutrients from depth by deep rooted shrubs 5. Enhanced growth of annuals associated with reduced density of shrubs which can act to increase levels of total N and organic C in the surface 10cm or so (and has been documents in a degraded Chenopod shrubland in the North Eastern Goldfields) 6. Establishment of naturalized Medicago spp (but this has been included in the reference site description). Overall I doubt that the site is severely depleted in nutrients or that regeneration would be limited by soil chemical fertility. Soil Biological State: Soil invertebrate recyclers: There would be little soil compaction although the cryptogam crust will have been largely removed. With depleted shrub density there would also be reduced area of leaf litter under canopies which may be the main effect. Soil organic matter: I scored this low because of the reduction in cryptogamic crusts but as noted above enhanced growth of annuals can act to increase surface organic matter. Reproductive potential: Understorey reproductive potential: I’ve given this as a two part score – first number is for the mid storey and the second for the ground storey. The ground storey will respond well to seasonal conditions, as evidenced by the site photos, and there is no reason to suppose its reproductive potential would be less than the reference. The midstorey has probably lost some of its reproductive potential - Maireana sedifolia seems to establish very infrequently in these communities; after probably a fairly prolonged absence or presence at quite low density, the same may be true of Atriplex vesicaria and A. bunburyana. The site data indicate that Maireana pyramidata, an increaser, did recruit between 1988 and 1992 so this species may still be capable of filling the gaps in the shrub layer. I have taken a view based more on the capacity of depleted components of the reference site to increase. Understorey structure: Understorey top height I have given a two part index – first figure for the midstorey and the second for the ground storey. The ratings basically just reflect the changes in seasonal conditions that seem to be reflected in the photos. Understorey ground cover Again a two part index as above. The rating for the mid storey reflects the substantially reduced shrub density of the site; although M. pyramidata did recruit between 1988 and 1992 the overall cover of the shrub layer is still quite low so I didn’t change the index. The score of 10 for the ground storey just reflects it capacity to respond to seasonal conditions probably as well as the benchmark (in fact it could be more responsive given the increased growth of annuals with 35

reduced shrubs, and the fluctuations of annual species in relation to seasonal conditions). Understorey structural diversity The photos indicate arrange of plant sizes on the site and the transect data indicate recruitment between 1988 and 1992 particularly so no need to downgrade the site relative to benchmark. Understorey composition: Understorey functional groups: The community is considerably reduced relative to benchmark; the recruitment of M. pyramidata between 1988 and 1992 is the reason for the increase in the index (given it is based on density and not cover as above). Iâ&#x20AC;&#x2122;m assuming the increase in density was retained after 1992. Understorey species richness: The high score indicates that many, probably most, of the species present in the benchmark are still present though at reduced density in many cases.