Home

Swift_Parrot_MCASS user manual

1. Future threats for swift parrot the composite a Sealed Function Mak Multi way map Select values of the primary data that meet the criteria Figure 4 Example of the MCAS S Multi way tool to generate a composite data layer To view or modify how layers have been integrated click on the composite layer and view change information in the information panel Within MCAS S a mouse hover over a pixel of interest ona composite layer will result in the viewer window displaying the names and pixel values for each of the input layers that influence the selected integrated map Figure 5 If viewer window is not visible go to Edit gt show viewer Page 18 S Landscapes and polic Hu A O A E E National Environmental LANDSCAPES amp Research Program POLICY hub mature forest E mature_forest High globulus baseline Current habitat resources for swift parrot ovata_baseline 710 0723 Swift parrot current nesting wlobulus baseline 6714005 Co E j ovata_baseline Figure 5 Viewer window in MCAS S showing values of a pixel in the Current habitat resources for swift parrot composite layer When the current habitat resources and current threats composites are combined into the final composite areas containing potential refuges from predation are highlighted In this case we want to include the dark grey pixels from the layer Swift parrot current nesting habitat that DO NOT co occur wit
2. er cree e e i ee ecconcacaatcccsascncaccosncn nee ccc 27 5 References Se 606 IER E oaoa oa ccc05300c 28 Page 3 AUD Landscapes and policy National Environmental LANDSCAPES amp Research Program POLICY hub ABARES Australian Bureau of Agriculture and Resource Economics and Sciences ALA Atlas of Living Australia ANN Artificial neural networks used in BIOMOD forecasting ArcGIS Software produced by ESRI that enables spatial data to be generated analysed manipulated stored and managed BIOMOD BIOMOD is a computer platform for ensemble forecasting of species distributions enabling the treatment of a range of methodological uncertainties in models and the examination of species environment relationships Composite data Any layer that combines data from more than one layer layer in MCAS S CTA Classification tree analysis used in BIOMOD forecasting Current climate Baseline climate surfaces 1976 2005 used in BIOMOD to obtain the habitat Suitability maps for the species FPA Forest Practices Authority Tasmania Future climate Future climate projections for the period 2070 2099 under the A2 emissions scenario based on the output of three dynamically downscaled climate models UKMO HadCM3 GFDL CM2 0 and MIROC3 2 medres Used in BIOMOD to obtain the future habitat suitability maps for the species GAM Generalized additive regression model used in BIOMOD forecasting GBM Gradient boosting machi
3. provide habitat resources for nesting Data on land tenure protection and nest observations are also included for the purpose of prioritising refuges or for reporting 2 5 Data for the worked example A complete list of data layers available for this analysis is shown in Table 1 The data are grouped into five types 1 Current habitat resources potential areas where swift parrots breeding may succeed as observed by Stojanovic et al 2014 Current threats potential areas where swift parrots breeding would not succeed Future habitat resources potential areas where swift parrots breeding may succeed Future threats potential areas where swift parrots breeding would not succeed Supporting data oR Won The types of spatial data considered for the analysis included vector and raster data MCAS S requires all data for analysis to be in raster format so other data types vector point and polygon were converted eg rasterised Species data invariably comes in the form of point locations where an animal has been sighted during a particular survey Stojanovic et al 2014 Prior to conversion data observation points for individual species were buffered by 5 km to account for species movement if needed other relevant buffering distances could be calculated using the raw point data directly in MCAS S Data values in this worked example are in binary format Values of were given to grid cells that meet the criteria for example presence o
4. Climate Futures for Tasmania project suggest that temperatures will rise by about 2 9 C by the end of the century based on the high CO emissions scenario SRES A1 which triggers a decrease in cloud cover and an increase in evaporation Grose et al 2010 The Climate Futures for Tasmania also projected great changes in seasonal rainfall patterns with increased rainfall over coastal regions and reduced rainfall over central Tasmania and the north west Grose et al 2010 However the frequency of extreme events like high temperatures and strong winds is projected to increase which will increase the risk of bushfires Grose et al 2010 1 3 The MCAS S Tool The Multi Criteria Analysis Shell for Spatial Decision Support MCAS S developed by Dr Rob Lesslie and colleagues at the Australian Bureau of Agricultural and Resource Economics and Sciences Lesslie et al 2008 is a decision support tool designed specifically for non GIS users to easily explore spatial data and apply them to natural resource management and planning problems The tool is free and users require little training to analyse spatial data and assess the results MCAS S allows people with different levels of technical and scientific expertise to easily and interactively interrogate and analyse spatial data answer what if questions and build scenarios Page 6 m juo Landscapes ana policy AR National Environmental LANDSCAPES amp v Research Program POL
5. Research Centre Hobart Tasmania Hansen MC Potapov PV Moore R Hancher M Turubanova SA Tyukavina A Thau D Stehman SV Goetz SJ Loveland TR et al 2013 High Resolution Global Maps of 21st Century Forest Cover Change Science 342 6160 850 853 Lesslie R Hill M Hill P Cresswell H and Dawson S 2008 The application of a simple spatial multi criteria analysis shell to natural resource management decision making In Pettit C Cartwright W Bishop I Lowell K Pullar D Duncan D eds Landscape Analysis and Visualisation Spatial Models for Natural Resource Management and Planning Series Lecture Notes in Geoinformation and Cartography pp 73 96 Mac Nally R and Horrocks G 2000 Landscape scale conservation of an endangered migrant the Swift Parrot Lathamus discolor in its winter range Biological Conservation 92 3 335 343 MCAS S development partnership 2014 Multi Criteria Analysis Shell for Spatial Decision Support MCAS S version 3 1 User guide ABARES Canberra Phillips SJ and Dud k M 2008 Modeling of species distributions with Maxent new extensions and a comprehensive evaluation Ecography 31 2 161 175 Saunders D and Tzaros C 2011 National Recovery Plan for the Swift Parrot Lathamus discolor Birds Australia Melbourne Stojanovic D Webb MH Alderman R Porfirio LL and Heinsohn R 2014 Discovery of a novel predator reveals extreme but highly variable mortality for an endangered migratory bird Diversi
6. change would affect swift parrot nesting habitat we developed models of habitat suitability using future climate projections and testing for change by the period 2070 2100 centred in 2080 hereafter 2080 or end of the century We repeated the procedure to test for sugar glider s habitat suitability and the two Eucalypts that provide food and hollows for nesting habitat resources As sugar glider s proved to be a major threat to swift parrots we wanted to know to what extent glider s habitat would co occur with swift parrots by the end of the century We also included the projected future forest fire danger index FFDI as a threat to consider the impact of climate change upon the likelihood of fire damage to swift parrot nesting hollows Habitat suitability for all species was modelled using BIOMOD in the biomod2 package Thuiller et al 2009 using R software BIOMOD generates ensemble forecasting maps based on ten Statistical models including regression GLM GAM and classification CTA MDA methods maximum entropy Maxent flexible discriminant analysis FDA machine learning techniques ANN RF GBM and surface range envelope SRE We use consensus scenarios for the species by utilising climate data for two time periods 1976 2005 hereafter the baseline period and 2080 under the A2 emissions scenario The modelling techniques in BIOMOD assume that the inputs represent presence and absence of the species In this study
7. ee O se a Research Program POLICY hub 6 National Environmental LANDSCAPES amp 2 7 MCAS S package structure The swift parrot nesting habitat MCAS S datapack consists of two MCAS S project files and a set of three MCAS S data folders The project files are MCASS_ swift parrot tutorial mcas contains the completed worked example described in this document MCASS swift parrot tutorial blank mcas contains links to the data used in the worked example but the workspace is blank so that users can select and combine the layers from scratch The three data folders are Data History and KML see Table 2 The Data folder contains all of the base data layers necessary for the swift parrot nesting habitat analysis The Data folder includes four sub folders Primary Classified Overlay and Mask Most of the base data layers for this analysis are located in the Primary sub folder We have also added some useful overlays and masks to their respective sub folders Some folders are initially empty but these should not be deleted as MCAS S will write to these folders for various operations Table 2 MCAS S folder structure and descriptions adapted from MCAS S user guide V3 1 2014 Level 1 Level 2 Description Data Primary Gridded base layers each habitat resource or threat Classified Gridded data exported from MCAS S by user initially empty Overlay Line or point data used for visual reference only Tasmania boundary Mask Gridd
8. x nE aiie ey pG h h pi BE t Ba ef 4 wf 2 x Aa A TAL go a oh ik ae a5 gt Pi me fee Ls z 7 if te 1 Legend WA o Tasmania EEE Eucalyptus globulus GlobMAP x ie Gaal Nest observations amp 5km buffer i ly T Mature forests N 0 25 50 100Km A Ratha se Figure 1 Nesting habitat for the swift parrot Eucalyptus globulus distribution from GlobMap nest observations from Stojanovic et al 2014 mature forest data from Forest Practices Authority Page 9 6 Landscapes and polic Hu Le sd National Environmental LANDSCAPES amp Research Program POLICY hub Food availability is a major driver in the location of nests for a given year and as Blue gums E globulus present a very irregular flowering pattern Webb et al 2014 it makes management of nesting habitat a challenging task Over the period 2010 2012 nesting sites were geo located and information about the habitat collected to assess breeding success or failure Stojanovic et al 2014 Their results showed that a major threat to swift parrots is predation by sugar gliders Petaurus breviceps Predation risk varied dramatically across the breeding range depending on the presence of sugar gliders One of the main conclusions of Stojanovic et al 2014 is that offshore islands are an important refuge for swift parrots because sugar gliders are absent 2 3 Habitat suitability modelling To test whether or not climate
9. 2012 based on Landsat satellite data Fire frequency for the period 2000 2012 based on MODIS satellite data MCD64A1 product NASA The McArthur Forest Fire Danger Index indicates the degree of fire danger in Australian forests as a function of climatic variables including rainfall wind speed temperature amp humidity Current extent of mature forests was produced by reclassifying Forest Practices Authority data into a binary file of mature or non mature forests Globmap data buffered by 10km to inclue areas just outside foraging habitat Intersected with mature forests to highlight potential mature nesting habitat This layer provides a spatial representation of Tasmania s Reserve Estate for use as the authoritative source of information on the extent type and distribution of the comprehensive adequate and representative CAR reserve system in Tasmania Interim biogeographic regionalisation for Australia landscape classification Useful for spatial location and determining distances to from coast Page 13 LANDSCAPES amp POLICY hub http earthenginepartner S appspot com science 2013 global forest Hansen et al 2013 http earthenginepartner s appspot com science 2013 global forest Hansen et al 2013 The combined use of active fire and reflectance data enables the algorithm to adapt regionally over a wide range of pre and post burn conditions and across multi
10. ICY hul that map the implications of different conservation management interventions MCAS S has the potential to be a very useful tool for ecological management because a range of spatial data can be easily integrated to explore potential futures or the effects of management decisions particularly in workshop situations when data may be incomplete and decision making is by consensus Here we show how this tool can be used to visualise and explore spatial variability in threats and opportunities for conservation of the nesting habitat of swift parrots in Tasmania 1 4 The swift parrot nesting habitat MCAS S datapack and tutorial This users guide and data package includes an MCAS S datapack which contains spatial data for identifying swift parrot nesting habitat at risk in Tasmania plus supporting data layers such as the reserve estate and biogeographic regions The MCAS S approach allows managers to pinpoint areas that are least threatened and therefore may act as refuge or sanctuary from predation and future changes in climate The datapack does not provide recommendations or an answer as to what management activities should occur Rather we provide the data for further exploration of the issues using MCAS S The datapack consists of MCAS S formatted spatial data and sample MCAS S project files that illustrate how the data can be combined to support decision making Further data could be added by users as it becomes available The aim of this t
11. IPWE 2010 Description BIOMOD output Areas predicted to be climatically suitable for the species based on nesting observation data BIOMOD outputs summarised using model agreement technique Areas projected to be climatically suitable by 2080 for the species based on current nesting observation data BIOMOD output Areas predicted to be climatically suitable for the species based on nesting observation data BIOMOD outputs summarised using model agreement technique Areas projected to be climatically suitable by 2080 for the species based on current nesting observation data BIOMOD output Areas predicted to be climatically suitable for the species based on observation data obtained from ALA BIOMOD output Areas predicted to be climatically suitable for the species based on observation data obtained from ALA BIOMOD output Areas predicted to be climatically suitable for the species based on observation data obtained from ALA BIOMOD output Areas predicted to be climatically suitable for the species based on observation data obtained from ALA The Swift Parrot Foraging Habitat Map GlobMap is the compilation of targeted species mapping and the pre existing TASVEG It is a more accurate map with a wider coverage of breeding season foraging habitat for swift parrots Page 12 Criteria Probability values above 50 were considered suitable Probability values above 50 were considered suitable We us
12. National Environmental LANDSCAPES amp Research Program POLICY hub The swift parrot nesting habitat MCAS S datapack from the Landscapes and Policy Hub TUTORIAL prepared by Luciana Porfirio Lauren Carter Sonia Hugh and Brendan Mackey The Australian National University Griffith University 2014 The Landscapes and Policy Research Hub is supported through funding from the Australian Government s National Environmental Research Program www environment gov au nerp and involves researchers from the University of Tasmania UTAS The Australian National University ANU Murdoch University the Antarctic Climate amp Ecosystems Cooperative Research Centre ACE CRC Griffith University and Charles Sturt University CSU ISl Aster E Murdoch pf Griffith yf Charles Sturt SS University W UNIVERSITY UNIVERSITY University p ON E sd National Environmental LANDSCAPES amp Research Program POLICY hub The swift parrot nesting habitat MCAS S datapack amp tutorial Hub Publication Reference Enquiries to luciana porfirio csiro au University of Tasmania This work is copyright It may be produced in whole or in part for study or training purposes subject to the inclusion of an acknowledgement of the source It is not intended for commercial sale or use Reproduction for purposes other than those listed above requires the written permission from the authors Requests and enquiries co
13. alue of 9999 This is the number MCAS S understands as NoData so it will be displayed transparent Ocean NoData Page 27 6 Landscapes and polic Hu ON Le National Environmental LANDSCAPES amp Research Program POLICY hub 4 E 5 References ABARES 2014 MCAS S Multi Criteria Analysis Shell for Spatial Decision Support V3 1 Department of Agriculture Canberra DPIPWE 2010 GlobMap The swift parrot foraging habitat map Biodiversity Conservation Branch Department of Primary Industries Parks Water and Environment Hobart Forest Practices Authority 2010 Interim species habitat planning guideline for the conservation management of Lathamus discolor Swift Parrot in areas regulated under the Tasmanian Forest Practices System Internal report to the Forest Practices Authority Hobart Tasmania Fox Hughes P Harris R Lee G Grose M and Bindoff N 2014 Future fire danger climatology for Tasmania Australia using a dynamically downscaled regional climate model International Journal of Wildland Fire 23 309 321 Giglio L Loboda T Roy DP Quayle B and Justice CO 2009 An active fire based burned area mapping algorithm for the MODIS sensor Remote Sensing of Environment 113 2 408 420 Grose M Barnes Keoghan I Corney S White C Holz G Bennett J Gaynor S and Bindoff N 2010 Climate Futures for Tasmania general climate impacts Technical report Antarctic Climate and Ecosystems Cooperative
14. and Dud k 2008 Phillips and Dudik 2008 Relevant vegetation information juno Landscapes ana policy Nest observations amp 5km buffer Forest loss Forest gain Fire frequency FFDI future projection 2081 2100 Mature forests Mature forests amp Globmap Privately and publicly managed reserves IBRA regions Coastline Dm AV Current and future habitat resources Current threat Future habitat resources Current threat Future threat Current habitat resources Current habitat resources Current and future habitat resources amp supporting data Supporting data Supporting data 6 National Environmental Research Program Dejan Stojanovic NASA NASA Data processed by Bioregional futures NERP LaP monthly satellite data obtained from Vegetation and Fire project NERP LaP satellite data downloaded from AusCover portal Climate futures NERP LaP Modified layer originally from www fpa ta S gOV au Globmap data amp FPA Mature forests data DPIPWE http dpipwe tas gov au conserva tion developmen t planning conservation assessment tools tasmanian reserve estate spatial layer Australian Government Department of the Environment Geoscience Australia GEODATA TOPO 250K Series 3 Field work data buffered by a5km radius Global Forest Change 2000 2012 based on Landsat satellite data Global Forest Change 2000
15. as a user Unzip the MCAS S Datapack Before you start playing with our worked example the next step is to extract the datapack from its zipped folder right click and unzip to you If the datapack isn t unzipped correctly MCAS S will not be able to upload the Spatial data and a red cross will appear in each map box instead of a map Now you are right to go Page 8 Dam National Environmental LANDSCAPES amp Research Program POLICY hub 2 Worked example background Mapping potential refuges from predation and climate change for swift parrot nesting habitat 2 1 About the worked example This worked example guides you through how you might use MCAS S to consider a query for mapping potential refuges from predation and climate change for swift parrot nesting habitat which occur only in Tasmania We provide background information about the issue sketch out the essential means to end diagram and then take you through a data combining exercise 2 2 Nesting habitat for swift parrots present and future threats Suitable nesting habitat for swift parrots requires the presence of mature forests with hollows for nesting and flowering Eucalypts mostly E globulus but also E ovata as food resources Forest Practices Authority 2010 These breeding habitat resources are scattered mainly along the East coast of Tasmania Figure 1 a 7 4 i af A sy i F pl J Lan A ah p 4 Liat 3 l y A i a tet 1 i
16. at using existing data on extant mature forest cover across the study area assuming that these forest will persist the availability of potential habitat decreased by approximately 50 We need to emphasise that these results are based on the primary data we used in the analysis Including new or revised data is likely to change the results However our main objective was to demonstrate how the MCAS S tool can assist with determining management priorities for the natural habitat of the swift parrot and to consider the interaction between multiple variables Page 26 yuo Landscapes and Pol oy National Environmental LANDSCAPES amp Research Program POLICY hub 4 Appendix Technical recommendations Extent All layers to be used in MCAS S should have the same extent In this case the extent for Tasmania is Top 5626000 Left 220000 Right 628000 Bottom 5163000 Layers to be used in MCAS S can have different resolutions however we recommend using the same resolution in particular for beginners The resolution we used is 100 x 100 metre pixels so 1 pixel 1 hectare Resolution All layers must have the same projection We reached an agreement with DPIPWE to use GDA94 Zone 55 South This ensures all pixels from north to south are comparable in terms of area Projection NoData VO data values within Tasmania were given a value of 888 No data values outside Tasmania in the ocean were given a v
17. d in MCAS S however itis important that the projection of this data matches the projection of the raster layers in our datapack In comparison masks can be used to analyse data within a specific area For example using the IBRA region mask select TSE Tasmanian South East All maps will be zoomed to that region and any further analyses will only be undertaken for the selected area To add remove a mask select the Mask tab on the Map menu bar and then check uncheck the relevant box 3 5 Reporting MCAS S gives the option to generate reports using the Reporting tool The reports give for example the total number of hectares in a selected region where the criteria are met In this example we can determine the total number of hectares that are considered as potential refuge from predation To compile a report right click on top of the map you are interested in and select reporting The Reporting window will appear see Figure 10 and there you can select the reporting region in our example we will use Tasmania For the type of report Report as we will use cell counts because our pixels are 100x100 m so 1 pixel 1 hectare All layers in this example have the same resolution so any layer can be chosen and the outcome should be the same If you chose to report by land tenure a table will appear with a list of the reserve names In this example we chose to report for the whole of Tasmania so total the number
18. dation and climate change for swift parrot nesting habitat M MUMMMas soscseseceeessssersseeresssessseseresecees 9 2 1 About the worked example 20 gllMecssccnscccreceess MQMMMMCeerocssssccnsccoeeccesssscsnsccoeeees 9 2 2 Nesting habitat for swift parrots present and future threats cessseccssseeeesssseeeeees 9 2 3 Habitat suitability Mocelling MQMMMbsessssscsssersteccrserscscrserss se MEM sores scseessseesees 10 ee FT Ga se 11 2 5 Data for the worked examp A NR MEMS ccsccnsorsccsscsscsccsssonscesccssonsccssossoones 11 2 6 Means to end diagram for current and future threats seessseeesssseeesssoessseeessseoessseeesssee 14 2 7 MCAS S packe ure SQM fll e MM eccrscccsserssscsscessscsseessscesees 15 3 Worked exam pia structiori sii Y SDM Cccesccesssscenscccesscesssccenscoesssesesece 16 COME Primary elec e Seeereeeerneeee Renee er eerie re ree tee ee 16 3 2 o ORE T E E 17 3 3 Modifying primary and COMPOSite data lAYESS seeessoossssoossssoosssssosssssossessosssssesssssesseo 23 3 4 Ways and mMasSh ih SgBBDS 0n ae 5 0000essscssssoeseseseessses 24 STE E RE ooo oao oo AE 24 3 6 Viewing PEBIBAS outsidai MCAS S eesseoesecoseeeeseoeessecesoeeesseessseessoeesesoessoeessoosssrossoressoss 25 3 7 Management conclusions from the worked example sssscccssseccsssscssssressssessssseeeeee 26 4 Appendix
19. dequately represent fine scale data To view files at the full resolution in Google Earth the user must export the layer from MCAS S see 3 below and convert to kml in an alternative application such as ArcGIS Page 25 6 Landscapes and polic Hu ON Le National Environmental LANDSCAPES amp Research Program POLICY hub 4 E 2 Save image Allows you to save the map legend and or histogram of the layer of interest 3 Export Creates a GEOTIFF or ASCII file of the layer and saves it to the Data Classified This procedure requires the creation of a small metadata file known in MCAS S as a tip file See Appendix 10 2 for details on completing a tip file metadata record 3 7 Management conclusions from the worked example This exercise highlighted areas that may currently act as refuge from predation for swift parrot nesting habitat About 31 of that area is already under the Tasmanian reserve system and the remaining locations may be interpreted as priority areas for conservation Figures 3 and 7 Our models project an approximate 23 reduction in climatically suitable area for swift parrot nesting habitat by 2080 predominantly along the north east coast of Tasmania However it is important to note that not all projected potential nesting habitat is necessarily available for nesting Some areas may be climatically suitable but may not support mature forest When we filtered our projections of future nesting habit
20. ed climate data from 3 GCMs that were summarised in one map using model agreement techniques Probability values above 50 were considered suitable Probability values above 50 were considered suitable We used climate data from 3 GCMs that were summarised in one map using model agreement techniques Probability values above 50 were considered suitable Probability values above 50 were considered suitable We used climate data from 3 GCMs that were summarised in one map using model agreement techniques Probability values above 50 were considered suitable Probability values above 50 were considered suitable We used climate data from 3 GCMs that were summarised in one map using model agreement techniques http dpipwe tas gov au Pages document aspx pat h Documents GlobMap p df Justification Phillips and Dud k 2008 Model agreement was the models summary technique where pixels in which all models predict values above 50 are displayed This technique was chosen by land managers from the Department of the Environment after a meeting in 2013 Phillips and Dud k 2008 Model agreement was the models summary technique where pixels in which all models predict values above 50 are displayed This technique was chosen by land managers from the Department of the Environment after a meeting in 2013 Phillips and Dud k 2008 Phillips and Dudik 2008 Phillips
21. ed data used to restrict the area of analysis The extent of Tasmania History Cached project files initially empty KML GoogleEarth kml files exported from MCAS S initially empty Page 15 UD vicy H Landscapes ana policy National Environmental LANDSCAPES amp Research Program POLICY hub 4 E 3 Worked example instructions To gain the most from the worked example open the MCAS S file in your unzipped folder and follow what we do in the text by having a play on the MCAS S screen Open the file MCASS swift_parrot_tutorial mcas 3 1 Primary input data The file MCASS_ swift parrot _tutorial mcas contains a worked example using most of the primary input datasets provided in the swift parrot nesting habitat datapack Figure 3 This worked example is based on the model structure shown in the means to end diagram in Figure 2 The primary input layers are the smallest maps shown and each has three categories NoData within Tasmania light grey NoData areas outside Tasmania eg the ocean are displayed as transparent Areas that meet the conditions associated with a particular layer are shown in colour and may have a single value eg swift parrot current nesting habitat layer or have a single to multiple classes categorical data eg the Tasmanian Reserve System layers can be combined into one layer showing private and public reserves To zoom into a layer you have three options change the in
22. elete Note you cannot undo a delete You ll need to drag the map back in from the Primary Input Data tab if you make an error e Change the input layers for an existing composite data layer Modify connections by changing the input layers on the information panel or by modifying how the layers are integrated e g by creating a Composite map using this option on the Map menu bar instead of a Two Way or Multi Way and selecting and completing either Manual Function or AHP weighting options to combine layers Page 23 D Landscapes and Policy NL EA National Environmental LANDSCAPES amp Research Program POLICY huly 4 E e Add layers from the primary folder Click and drag layers from the Primary Input Data tab e Create new integrated layers Click and drag from either the Composite Two Way Reclass or Multi Way tabs to create a new blank map Create connections by selecting input layers on the Information Panel See the MCAS S user guide for more information about these tools 3 4 Overlays and masks The swift parrot nesting habitat datapack contains some overlay and mask files including e IBRA regions e reserves Overlays added to the maps are displayed on screen primarily as a useful visual reference as they do not influence any of the calculations occurring in the model Any line or point data can be added to the overlay folder and viewe
23. f nesting habitat resources NoData values were used for grid cells that do not meet the criteria Justification for each criterion are provided in Table 1 Page 11 o icy Huo Landscapes and Po National Environmental 4 Research Program LANDSCAPES amp POLICY hub Table 1 Primary input layers available for swift parrot nesting habitat analysis Data layer Swift parrot current nesting habitat Swift parrot future nesting habitat 2080 Sugar glider current suitable habitat Sugar glider future suitable habitat 2080 E globulus current suitable habitat E globulus future suitable habitat E ovata current suitable habitat E ovata futuret suitable habitat GlobMap Type Current habitat resources Future habitat resources Current threat Future threat Current habitat resources Future habitat resources Current habitat resources Future habitat resources Current and future habitat resources Source Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP Bioregional futures and Climate futures NERP LaP DPIPWE DP
24. for integrating layers Details for how to alter these options can be found in the MCAS S User Guide 2014 Section 6 Explore and Combine data In this worked example a very simple rule set is used to integrate all of the data layers into three composites for current data 1 Current habitat resources for swift parrot 2 Current threats to swift parrot 3 Current potential refuges from predation Using the Multi way from MCAS S the composite data called Current habitat resources for swift parrot Figure 4 shows areas that are e Covered by mature forests all of the 3 categories Page 17 s Landscapes and Pol oia NL EA National Environmental LANDSCAPES amp Research Program POLICY huh nest observations with 5km buffer zone and values gt 0 5 from the Swift parrot potential nesting habitat layer habitat suitability map using current climate data the BIOMOD outcome In a similar fashion the second composite data layer Current threats to swift parrot is created using the Multi way tool selecting e Sugar glider current suitable habitat e Forest loss and e Forest gain Source Layers Current habitat resources for swift parrot A Eglobulus_model_aggrement is Eovata_model_aggrement v Forest loss B vy forest_gain_tas Select primary Future FFDI Future habitat resources for swift parrot data to create Future potential refuges for swift parrot
25. ft parrot nesting habitat MCAS S datapack 1 1 The swift parrot The swift parrot Lathamus discolour is an Australian nectarivorous bird that is listed as endangered under the Tasmanian Threatened Species Protection Act 1995 and endangered on the Commonwealth s Environment Protection and Biodiversity Conservation Act 1999 It has a regular annual migration pattern with the entire population migrating from south eastern Australia to Tasmania from September to March as Tasmania is the only location where the species breed The nesting habitat is characterised by a very specific type of tree hollow only found in the mature forests of Tasmania mostly Blue gums Eucalyptus globulus which also provides a food resource during the breeding season Swift parrots are threatened by broad scale habitat loss Mac Nally and Horrocks 2000 which is likely to interact with other threatening processes such as heightened predation The Tasmanian breeding range of swift parrots is also under ongoing pressure from habitat loss Saunders and Tzaros 2011 and it may be impacted by future climate projections A management priority in Tasmania is to detect areas for conservation to secure nesting habitat resources for the species in the long term However there are knowledge gaps in the species ecology Forest Practices Authority 2010 which adds further complexity to the issue 1 2 Climate change in Tasmania direct and indirect threats Climate projections from the
26. h the dark pixels from the Current threats to swift parrot This is achieved by using the Multi way tool and checked the dark grey pixels from the habitat maps that for some reason appear in RED in the display and the light grey pixels from the threats map that appear in BLUE in the display Figure 6 Page 19 s Landscapes and Pol oia n A EA A 6 National Environmental LANDSCAPES amp ew Research Program POLICY hub E File Edit Help Layer name Potential refuges from pr Source Layers 5km buffer around nest observations M Current threats to swift parrot Fire frequency Forest loss Mature forests Select the 2 composite layers Sugar glider current suitable habitat M Swift Parrot current nesting habitat Swift parrot potential nesting habitat Function Mask T Multi way map Select values of the primary Et ai soe gt ee eee ee data that meet iii the criteria We want areas where the maps intersect has Figure 6 Example of the MCAS S Multi way tool to generate final objective layer A likely modification to this worked example would be to change the multi way maps to composite maps to allow weightings to be added for individual primary data layers For example some threats may be more important in terms of their negative impacts on the swift parrot nesting habitat than others so they could be ranked in MCAS S based on expert knowledge More inf
27. icy hub at the Australian National University Professor Brendan Mackey is Director of the Griffith Climate Change Response Program based at Griffith University He is a member of the International Council for the International Union for Conservation of Nature IUCN and serves on various governmental advisory bodies Acknowledgements The report is an output from the Landscapes and Policy Research Hub The hub is supported through funding from the Australian Government s National Environmental Research Program and involves researchers from the University of Tasmania UTAS The Australian National University ANU Murdoch University the Antarctic Climate and Ecosystems Cooperative Research Centre ACE CRC Griffith University and Charles Sturt University CSU Page 2 ana policy HUD landscapes Da Gs National Environmental LANDSCAPES amp ee Research Program POLICY huh Table of Contents 1 Introduction to the swift parrot nesting habitat MCAS S datapack cssscccssseeeesssseeeeeee 6 ti WMS SV VIE ANN O eae excuse pecestseyseeent copuacenacuseepecesteegseeee oops ese pecrtene eae eae 6 1 2 Climate change in Tasmania direct and indirect threats csssccsssssecssssssesscssesesssseeees 6 Mage MU Ws OO ecdectesaieatce tenon A 6 1 4 The swift parrot nesting habitat MCAS S datapack and tutorial csseecsssseeessssseees 7 2 Worked example background Mapping potential refuges from pre
28. ive and the types of data that were used to meet the objective Comprehensive primary data that reflects species ecology is crucial to a meaningful outcome The data were combined in the MCAS S example to explore spatial variability in potential refuges from predation and climate change for swift parrot nesting habitat in Tasmania For other projects we strongly recommend developing a means to end diagram prior to getting data into MCAS S This will ensure that the model structure is appropriate to the project aims and for the required data to be identified formatted and entered into the datapack Each MCAS S layer can be exported to a GIS such as ArcGIS or Google Earth or saved as an image Primary data Composite data Objective Swift parrot potential nesting nabitat Nest observations amp 5km buffer Habitat resources for swift parrot Mature forest Map of potential refuges for swift parrot nesting habitat Eucalypts from predation amp climate Sugar glider change suitable habitat Threats to swift parrot Forest loss Figure 2 Means to end diagram to obtain the map of potential refuges or sanctuaries for swift parrot nesting habitat from predation and climate change Boxes on the left represent primary input layers and all other boxes represent different levels of spatial integration The red box represents the final composite product that meets the original objective Page 14 O Landscapes and Pole
29. ncerning reproduction rights should be addressed to Communications Manager Landscapes and Policy Hub Private Bag 141 Hobart Tasmania 7001 Tel 61 3 6226 6276 Email Landscapes Policy utas edu au Purpose of Document The aim of this users guide is to describe the spatial data in the datapack and provide instructions on how the layers can be analysed and explored in MCAS S for decision support The instructions take the form of a worked example You will need to download and install MCAS S software to your desktop or laptop The worked example then steps you through using MCAS S and how to open an MCAS S package to combine layers This guide is not intended to be a comprehensive guide to using MCAS S For more detailed information on how to use MCAS S including how to format spatial data for input to MCAS S please see the ABARES website www abares gov au mcass and User Guide ABARES 2014 The report is an output of the Landscapes and Policy Research Hub Please cite the document as follows Porfirio L Carter L Hugh S amp Mackey B 2014 The swift parrot nesting habitat MCAS S datapack Landscapes and Policy Hub Hobart Tasmania About the Authors Dr Luciana Porfirio was a post doctoral fellow in the Landscapes and Policy hub at the Australian National University currently a OCE Pos tdoctoral Fellowship at CSIRO Oceans and Atmosphere Flagship Lauren Carter and Sonia Hugh are both Spatial Analysts in the Landscapes and Pol
30. nes used in BIOMOD forecasting GCM Global Climate Models used to project the future climate GIS Geographic Information Systems are tools used to capture store manipulate analyse and manage spatial data GLM Generalized linear regression model used in BIOMOD forecasting Habitat The ability of an area to support the survival and reproduction of a focal species suitability Our models calculate suitability of a pixel based on climate information also known as the climatic envelope for a given species IBRA Interim Biogeographic Regionalisation for Australia undertaken by the federal Department of the Environment Information Left hand panel in MCAS S used to edit maps Panel Input layer Data layer contained within the primary folder of an MCAS S project LaP Landscapes and Policy Hub Mask A type of data layer used in MCAS S to restrict analyses to a defined area Mature forest Mature or old growth forest is characterised by a forest that has attained great age without significant disturbance and thereby exhibits unique ecological features Mature forests present a diverse and complex structure characterised by a multi layered canopy tree heights and diameters MCAS S The Multi Criteria Analysis Shell for Spatial Decision Support is a decision support tool designed specifically for non GIS users to integrate spatial data Page 4 O Landscapes and Pole ee O se a National Environmental Research Program LANDSCAPES am
31. of cells that met the criteria are shown as a single row in the report Figure 11 Reports can be saved as csv files and opened in a spreadsheet program for further analysis and data manipulation Page 24 AID Landscapes ana policy 2 amp National Environmental LANDSCAPES amp w Research Program POLICY hub eee ee Sand nes observadors w Currert threats to sart parrot Fire teauency Forest kes Mature forests Sugar ghder current sutable habitat w Sarh Parrot current nessng hatetat Sarh partot potersal nestng habetst Function eon K 1 Right click on 2 top of this map 2 Note that we want a report for all Tasmania region section using Cell counts because our cell are 100x100m 1 hectare and the resolution in this example doesn t matter as all layer have the same resolution 3 Then click generate Figure 10 How to generate a report in MCAS S OF Export Layer Save Figure 11 MCAS S sample report 3 6 Viewing results outside of MCAS S MCAS S allows users to export layers for use in other applications The available options for exporting are accessed via a right click on the layer of interest and include 1 Show in Google Earth Note at time of publication there was a maximum resolution 1000 1000 cells for viewing MCAS S layers in Google Earth Google Earth will render a lower resolution version that may not a
32. ormation about how to add weightings to composite layers can be found in the MCAS S user guide MCAS S development partnership 2014 We compiled other relevant primary data that can be used in this exercise For example add into the MCAS S panel the maps of private and public reserves listed in Table 1 combine them into a single map called reserves then combine that map with the Potential refuges from predation to see areas that are already under the reserve system Figure 7 Another option is to use the new reserve map for reporting in that case the layer should be exported automatically saved in the Classified folder and then transferred to the Mask folder Page 20 s Landscapes and Policy Hu National Environmental LANDSCAPES amp Research Program POLICY hub Layer name Potential refuges in ther Source Layers Sken buffer around nest observabors A Current treats 1o swi parrot v Potential refuges tom predabon Privately managed land Publicly managed land w Reserves b AD a aats ge tom eer Reserves TO Other relevant layers to Potential refuges in the reserve system Publicly managed land include in the example Figure 7 A snapshot of the MCAS S example when reserve layers are incorporated For our climate change example another simple rule set is used to integrate all of the data layers into two future composites future habitat resources and fut
33. p POLICY hul Glossary continued MCAS S___ A series of folders and data layers organised and formatted ready for use in an datapack MCAS S project MDA Mixture discriminant analysis used in BIOMOD forecasting Model A technique used to summarise the output of multiple habitat suitability models agreement in which the pixels where ALL models predicted suitable are shown on the map Multi way map A composite map layer created using the multi way function to combine more than two input layers NERP National Environmental Research Program funded by the Australian Government Overlay A data layer used in MCAS S as a visual reference which does not influence any calculations occurring in composite maps P4 Project 4 within LaP also known as the Bioregional Futures Project RF Random forest used in BIOMOD forecasting Tip file A short metadata record used to describe data layers exported from MCAS S Two way map A composite map layer created using the two way function to combine two input layers Viewer window A pop up window within in MCAS S that shows the value of a pixel in a map plus values of all input layers to that map upon mouse hover over the pixel of interest The viewer opens by default but if closed can be opened via the Edit tab Page 5 6 Landscapes and polic Hu ON NL National Environmental LANDSCAPES amp Research Program POLICY huly 4 E 1 Introduction to the swi
34. p Oy ear Cisggties Compa Tece Recim Wuhan Ui eat D Dra i 5 SP curent mutable m E __ 3 Matre Globmap buffer LE ee CURREN I habitati resource f r Se bardine 3 a i E globulus baseline fas Fi i at j a j f A g Combined baseine ii ko pa oe i n an pear E ovata_baseline forest_gun tas S T a f gt Ta ka 4 ics lon as SSS od CURRENT threats Curent potential refuges pidea amet habitat saitability SP furure_suitable 12080 r a a es Mature Glebmap butte S q a y m Ee E gl cbul per m e a seepage aame globulus model agreement _ z zi n a FUTURE bbita resources ri p i hag a gt wW Combined Model agreement Se i E ovate modil permei s _ i Ai g A F EFE Future FFDI i if j FUTURE threats Gide future motahle 2080 Future potential refuges Figure 9 Current and future potential refuges from predation and climate change for swift parrots 3 3 Modifying primary and composite data layers Clicking on any layer within the worked example will show details about the layer in the left hand information panel Many of the options within this panel can be altered including the layer name colour scheme the number and colour of classes and the names of the categories or classes within each layer Other ways of modifying results are to e Delete layers Right click then d
35. ple ecosystems Text from Giglio et al 2009 http dx doi org 10 1016 j rse 2008 10 006 Classes extreme severe and very high used to mask future suitable habitat for the focal species Dataset not included as it was provided by the FPA under licence We focus on the co occurrence of reserves and suitable nesting habitat Stovanovic et al 2014 Forest loss is a major threat for swift parrot nesting habitat Forest gain may represent an opportunity for swift parrot nesting habitat in the future Depending on its intensity fire can cause damage to mature forest s structure reducing the availability of nesting habitat Future climate change is projected to alter fire regimes with an increase in fire danger amp changed seasonality of fires which may impact Swift parrot habitat Fox Hughes et al 2014 Mature forests provide food and nesting habitat resources FPA 2010 notes the Eastern breeding range was buffered by 10km to include areas of mature forest just outside the foraging habitat Areas that are projected to be suitable and are under a private or public reserve system provide an opportunity for management na na na na y AUC and Policy t Landscapes aR National Environmental LANDSCAPES amp w Research Program POLICY hub 2 6 Means to end diagram for current and future threats The means to end diagram illustrated in Figure 2 shows the main object
36. terface magnification first option on map menu bar above the right hand side workspace double click on a map Esc to exit or right click and then select Show in Google Earth Familiarise yourself with the different input layers by clicking on each individually and then viewing properties of the data in the left hand information panel At this point it would also be useful to review the information about the different input layers in Table 1 Page 16 yuo Landscapes and Poli oy National Environmental LANDSCAPES amp Research Program POLICY hub zd A A E E H g Composite Two way Reclass Multr way Overlay Mask ta SP_current_suitable Primary Input Data Mature Globmap buffer CURRENT habitat resources i E globulus_baseline Combined baseline E ovata_baseline forest_gain tas CURRENT threats forest_loss_tas Current potential refuges glider_current_habitat_suitability Figure 3 Structure of the swift parrot current nesting habitat MCAS S model Colour indicates areas that meet nesting habitat or threat values Table 1 Composite layers are monochrome with dark grey indicating areas that meet the criteria The final composite shows areas that are currently suitable for nesting habitat and are comparatively free of threat 3 2 Composite data Any output that combines data from more than one layer is a composite or integrated layer There are several options within MCAS S
37. ty and Distributions 20 1200 1207 Page 28 Huo Landscapes ana policy National Environmental LANDSCAPES amp Research Program POLICY hub E Thuiller W Lafourcade B Engler R and Araujo M 2009 BIOMOD A platform for ensemble forecasting of species distributions Ecography 32 369 373 Webb MH Wotherspoon S Stojanovic D Heinsohn R Cunningham R Bell P and Terauds A 2014 Location matters Using spatially explicit occupancy models to predict the distribution of the highly mobile endangered swift parrot Biological Conservation 176 99 108 Page 29 D Landscapes and Policy Le EA
38. ure threats to swift parrot see Figure 8 Using the Multi way from MCAS S the composite data called Future habitat resources for swift parrot shows areas that are e covered by mature forests assuming that if not affected by land use change mature forest should provide the same habitat resources by the end of the century e nest observations with 5km buffer zone and e Swift parrot future nesting habitat model agreement layer calculated using the future climate projections by 2080 the BIOMOD outcome see Table 1 In a similar fashion the second composite data layer Future threats to swift parrot is created using the Multi way tool selecting e Sugar glider future suitable habitat and e FFDP Page 21 s Landscapes and polic Hu O A A EE National Environmental LANDSCAPES amp Research Program POLICY hub SP_future_suitable_2080 Mature Globmap buffer Future potential refuges Glider_future suitable 2080 Figure 8 Future refuges from predation and climate change The current and future potential refuges for swift parrots can be compared on screen and then combined into a single map showing areas that are currently deemed to be refuges and are projected to remain as refuges by the end of the century Figure 9 Page 22 icy RHUL Landscapes and Poll Da 6 National Environmental AW Research Program LANDSCAPES amp POLICY hulp oR SHH am
39. utorial is to describe the spatial data in the datapack and provide instructions on how the layers can be analysed and explored in MCAS S for decision support The instructions take the form of a worked example You will need to download and install MCAS S software to your desktop or laptop The worked example then steps you through using MCAS S and how to open an MCAS S package to combine layers This tutorial is not intended to be a comprehensive guide to using MCAS S For more detailed information on how to use MCASSS including how to format spatial data for input to MCAS S please see the ABARES website www abares gov au mcass and User Guide MCAS S development partnership 2014 Page 7 HUD Landscapes and Poia National Environmental LANDSCAPES amp ew Research Program POLICY hub ke Installing MCAS S Software To use the datapack you need to download and install the free MCAS S software ABARES 2014 onto your Windows computer 1 Go to MCAS S home page www abares gov au mcass 2 Go to the right hand side menu and click on 3 Anew page opens with two download options a MCAS S Version 3 1 installer 150MB takes about 10mins b MCAS S Version 3 1 user guide 10MB 4 Download the installer arrives zipped then unzip takes about 10mins If you are new to MCAS S also download the MCAS S user guide 5 Once downloaded follow prompts to install requires Quicktime to also be installed 6 Register
40. we used presence only data and randomly generated pseudo absences in BIOMOD for each species We used 70 of the data to run the ensemble forecasting and the remaining 30 was used to run evaluation models Models were evaluated with a relative operating characteristic ROC The bioclimatic variables used in BIOMOD were selected using principal component analysis PCA to select uncorrelated variables R lt 0 6 that accounted for more than 95 of the variance We present the ensemble SDMs as binary maps of climate suitability using a threshold value of 0 5 Phillips amp Dudik 2008 Page 10 AWD Landscapes ana policy 6 National Environmental LANDSCAPES amp Se Research Program POLICY huh 2 4 Aim The MCAS S datapack can be used to identify locations that could act as refuges for swift parrots in Tasmania by integrating spatial data on the location of landscape features that may influence the availability of habitat resources during the breeding season The types of spatial data considered for the analysis included habitat suitability maps based on current and future climate projections mature forest extent the forest fire danger index FFDI and forest loss and gain during 2000 2012 Forest loss and gain are considered a threat to swift parrots because both represent loss of mature forests Forest gain mainly highlights forests converted to plantations or those recovering from being dramatically burnt which are unlikely to

Swift_Parrot_MCASS user manual

Contents

Download Pdf Manuals

Related Search

Related Contents