User guide for the AGUAANDES (WATERWORLD) Policy Support System

Documentation written by: 

Mark Mulligan with contributions from: 

Leonardo Saenz (KCL), Arnout van Soesbergen (KCL), Leo Zurita Arthos (KCL), Louis Parker (KCL), Rebecca Goldman (TNC), Ruth Swetnam (University of Cambridge), Sophia Burke (AmbioTEK)

Citation:

Mulligan, M. [year]  User guide for the AGUAANDES (WATERWORLD) Policy Support System [version]. Available online [https://goo.gl/gvTKq6].


AguaAndes/Waterworld is a web based policy support system (PSS) based on the FIESTA hydrological model (Mulligan and Burke 2005; Mulligan et al., 2010; Bruijnzeel, Mulligan and Scatena 2011) and previous policy support systems including DESURVEY PSS).  When used inside the Andes mountain range the system is referred to as AguAAndes and has been further developed for the Challenge Programme on Water and Food basin focal project for the Andes and the COMPANDES project.  When applied elsewhere in the world the PSS is referred to as WaterWorld.  The PSS is a testbed for the development and implementation of land and water related policies, enabling intended and unintended consequences to be tested in silico before they are tested in vivo. It incorporates detailed spatial datasets at 1-square km and 1 hectare resolution globally, spatial models for biophysical and socioeconomic processes along with scenarios for climate, land use and economic change. A series of interventions (policy options) are available which can be implemented and their consequences traced through the socio-economic and biophysical systems. The model integrates with a range of geobrowsers for immersive visualisation of outcomes. A series of policy exercises are provided and the model can be used by policy analysts, scientists and students in Spanish or English. Works best in the Mozilla Firefox browser. Your browser must have Javascript enabled and any pop-up blockers disabled.  

Models are learning tools which can provide guidance if their results are appropriately supported by experience and observations.  Look at the results of these models carefully and use them to better understand the outcomes of your baseline or intervention rather than as a numerical prediction of the present or future.  Read and understand the model descriptions carefully to be able to better interpret the behaviour of the model in response to spatial variability of input data, scenarios or policy options.  If in doubt ask us.

You can see a FAQ here

You can see a powerpoint presentation on the system here.


Index

User guide for the AGUAANDES (WATERWORLD) Policy Support System

Disclaimer

Commercial use

Fair Use

Citation and redistribution

VIDEO DEMONSTRATION QUICK START GUIDE

THEORY, the video links below describe the theory behind these systems:

PRACTICE, the video  links below describe the steps required for use:

TEXT QUICK START GUIDE

Creating an Account

Once Logged in

Step 1: Define Area.

Step 2: Prepare data.

Step 3: Start simulation. Run the PSS for your site

Step 4, 5 and 6: View results

Glossary (definitions of terms used by the PSS)

TIPS AND TROUBLESHOOTING

FULL USER GUIDE

VIDEO:

TEXT:

Your browser: Accessing the AguaAndes/WaterWorld Policy Support System

Before logging in

Creating an Account

Access levels

AguAAndes/WaterWorld functionality (Scientist interface)

VIDEO:

TEXT:

Setting up a simulation

EXAMPLE POLICY EXERCISE: Land use change

EXAMPLE SCENARIO: Climate change (documentation by Arnout van Soesbergen)

Mobile use of data offline (in the field)

Manage simulations (Scientist)

Step 5: Results : maps (Scientist)

Step 6: Results : stats (Scientist)

Step 7: Results: narrative (Scientist)

Redefine land use and cover according to your own map

AguAAndes/WaterWorld functionality (Hyperuser interface)

AguAAndes/WaterWorld functionality (Megauser interface)

Ecoengine Functions by step (links to video demonstrations)

Step 1: Define area

Step 2: Prepare Data

Step 3: Start simulation

Manage Simulations

Step 4: Policy exercises

Step 5: Results maps

Step 6: Results stats

Step 7: Results narrative

Questions and Answers

GUIDANCE FOR PRE- and POST- PROCESSING OF DATA

Conducting Hydrological Analysis Over Multiple Tiles (by Louis Parker)

Creating a Zone of Interest (ZOI) map from a  shapefile using SAGAGIS (by Mark Mulligan)

Creating a PSS grid from a  shapefile using SAGAGIS (by Mark Mulligan)

Combining tiles and clipping them to a shapefile in SAGAGIS (by Mark Mulligan)

Cutting and resampling your gridded data so that it is compatible with the PSS and ready for upload in SAGAGIS (by Mark Mulligan)


Disclaimer

No warranty or liability The developers of AguAAndes provide this system and associated data without any warranty of any kind whatsoever, either express or implied, including warranties of merchantability and fitness for a particular purpose. The developers shall not be liable for incidental, consequential, or special damages arising out of the use of the system or of any data downloaded or used online.  For the period that this system is is in beta please be aware that functionality can break and results can be incorrect - check them carefully and report any errors using the Beta, report problem link.


Commercial use


This tool uses datasets from suppliers who make these available for non-commercial use only  (e.g. the WDPA, IBA and EBA datasets).  This means that these datasets cannot be used if your project is a commercial one.  Most definitions of 'commercial use' include a) any use for profit or to generate revenue, or b) any use by an individual or entity operating within or on behalf of or to the benefit of or to assist the activities of any entity other than a not-for-profit organisation.  

If you need to use the tool for commercial use and do not have license to then you must license and use the commercial-use version, which excludes these datasets.




Fair Use

Each registered user can run one simulation concurrently.  Please do not open multiple accounts and tie up our servers running multiple runs simultaneously.  If you make significant and prolonged use of this system in your organisations operations you should consider funding a server to improve the service provided to all users.  Each user can store 3 simulations on our servers (since each simulation takes 5GB of space).  Please delete simulations once you have finished downloading results from them.


Citation and redistribution

A lot time and effort has gone into producing this PSS and making it freely available.  Please cite the source of any results, data, images etc that you use from this system as Mulligan, M [year] Results from the WaterWorld Policy Support System [version number]. http://www.policysupport.org/links/waterworld Kings College London also request reprints of any publications and notification of any redistributing efforts.

WaterWorld is frequently updated so please do not redistribute datasets that you have taken from WaterWorld that would invariably be out of date very soon, but rather point interested users to WaterWorld so that they can generate those datasets anew.  If you wish to redistribute WaterWorld data for any reason , please contact us.  Users are prohibited from any commercial, non-free resale, or redistribution.


VIDEO DEMONSTRATION QUICK START GUIDE

THEORY, the video links below describe the theory behind these systems:

Presentation:Comparison of ES tools (0h1m13s)

    Making Science accessible - advantages and objectives of WW /CN (0h5m35s)

WaterWorld Scientific principles and processes (0h10m30s)

    WW v2.7 Scientific principles and processes (0h12m36s)

PRACTICE, the video  links below describe the steps required for use:

  Logging in and choosing the interface: Scientist and Licensed User, creating accounts (0h38m24s)

    Step 1: Define area (0h13m55s)

    Step 2: Prepare data (0h16m1s)

    Step 3: Start Simulation (0h16m54s)

    Step 4: Policy Exercises : Climate change scenarios (0h18m1s)

    Step 5: Results maps - options - GIS (0h18m56s)

    Step 6: Time series and Step 7: Narrative (0h21m14s)

    Documentation and training (0h22m34s)

    Summary of Functions, Outputs, Scenarios and Interventions (0h23m14s)


If you prefer text use the:

TEXT QUICK START GUIDE

written by Mark Mulligan, Rebecca Goldman, Sophia Burke

Creating an Account 

Sign up: Click “Create an Account” link in the left hand Nav Bar and complete the details

  1. The system will send an email to the account you have entered. Once you get the email, click the link and log in.  If the email does not arrive within a few minutes check your spam folder and if it is not there contact us and we will confirm your account manually.
  2. The default for log in is “policy analyst interface” – choose scientist interface to log in with more details or superuser to enable the latest features but this is in development so liable to interface bugs.
  3. The default interface is google maps.  If you prefer change preference to google earth but this will load more slowly than google maps. Note the user can change to google earth in the next stage once logged in

Once Logged in 

Logged in as Policy analyst:  When logged in as policy analyst there are only three steps: define area, policy exercises and outcomes.  Define area allows the user to define the study area as explained below.  Policy exercises allow the simulation of a baseline and then any scenarios or interventions (combining the Start Simulation and Policy Exercises functions in the scientist interface).  Outcomes incorporates the results: maps, results: stats and results: narrative functions under the scientist interface.         

Logged in as scientist and above:

Step 1: Define Area.

[necessary when logged in as policy analyst or any other user type] 

  1. Move the Google map until the crosshair in the centre of the map is above your area of interest.  The defined area for simulation will be the “tile” (square) area containing this point. The size of the tile can either be a ten degrees (low resolution) or one degree tile (high resolution), according to the resolution selected. To search for a location type the location name or coordinates (latitude, longitude) into the Location box above the map and click Go>. Note that western hemisphere longitude start with a minus and southern hemisphere latitudes start with a minus. Use decimal degree i.e. -2.65727 not degrees, minutes and seconds.  
  2. Provide a short name for the run in the box labelled Run name.
  3. Choose whether the run will be tiled 1km resolution (over a 10 degree tile) or tiled 1-hectare resolution (over a 1 degree tile).

Step 2: Prepare data.

Getting all data about your site [necessary when logged in as scientist and above]

  1. The system will list the datasets available for the simulation.  If some are not available they will appear as red crosses in the Available column.  If any have not been selected by a previous user you will need to build (extract and format) them so click 'Build missing map tiles'. When they are built you will be asked to click 'prepare data' again to list them.  You can look at the input datasets by clicking the + to view the datasets and clicking  geobrowse for the dataset of your choice. Once all input datasets are ready, click “Copy data to your workspace” at the bottom of the page.  This will copy all the data to your workspace on the ecoengine servers, (not to your local computer).
  2. It may take a couple of minutes for the system to  prepare the maps. When done,  the screen will indicate “data ready”  with a long list of available maps.
  3. View the data maps using “geobrowse
  4. If the data is licenced for re-distribution then the link “download” will be active.

Step 3: Start simulation. Run the PSS for your site  

[necessary when logged in as scientist and above]

  1. Click “start simulation” in the left hand nav bar under “Simulation”
  2. Click the “Start” button
  3. Click “Yes” to accept that by running the model you will delete the results of any previous run of this simulation.
  1. If the screen does not indicate that the model is running and provide a progress bar with updates, go back to the main screen and click 'start simulation' again to refresh this screen.
  1. The model will take some time to run.
  1. If this is a region/tile in the Andes run time will be about 10 minutes.
  2. If this is an area that has never been run before it will take a whole day, so you may close the  browser and computer overnight as the programme is runnning remotely on the ecoengine servers.
  3. The display should refresh every 10 seconds, if not then click “refresh”  to review the status of the simulation.

Step 4, 5 and 6: View results  

[necessary when logged in as scientist and above]

  1. To see results
  1. Click on the type of results you want to see back on the original screen
  1. Step 5 “Maps”
  2. Step 6 “Stats” (will be timeseries)
  3. Step 7 "Narrative" (will be a narrative of the model outcome)
  1. Can look at Results before the run finishes – can see them as they become available by clicking “Results – maps” and then refreshing that window with the refresh button
  2. To see maps from the maplist
  1. Click on “geobrowse”
  2. To get data click “download” – will be in zipped GIS Arc ASCII grid format
  1. Looking at Maps in Geobrowse
  1. Can Look at the maps in 6 ways (buttons below the map)
  2. Button 1: Google earth (will overlay data on google earth – must have google earth on your computer)
  3. Button 2: Google maps (overlay on google maps)
  4. Button 3:  Makes the mapped results “full size”
  5. Button 4: Change the color scheme
  6. Button 5: Shows Frequency Distribution
  7. Button 6: Will pop the map out into a separate window so can compare side by side with other maps
  8. Can change the scale to log scale for results that have a very large range of values
  1. Click “shifted log scale”
  2. Click “accept”
  1. Other
  1. Units are usually found at the top of the map screen, citations at the bottom of the map screen
  2. Can see the Google maps in 3 “styles” (click box next to Style label in the upper right of the map)
  1. Satellite
  2. Map – with cities, roads, etc
  3. Hybrid – labels in satellite version
  1. The model will first generate  map data for each timestep
  2. The last maps to be generated will be averages or totals for the year
  3. Description of map variables will come up next to each result with units links to glossary.  Hold the mouse over the concatenated name to see the full anme in a 'tool-tip'
  4. In the Google maps or Earth visualisations, one can zoom into a particular city by typing in the location or entering lat and lon
  5. When looking at a map can always click the “go back” button (or right click white space in window and choose back from context menu that appears) to be able to select other maps from the map list

If you have problems see troubleshooting 


Glossary (definitions of terms used by the PSS)

 




TIPS AND TROUBLESHOOTING


1. Use Mozilla Firefox - we cannot ensure that all features will work properly on other browsers 

2. If a page stops loading for more than a few seconds press F5 to reload 

3. If a model simulation is running and the progress bar is not being updated then click the 'start simulation' link from the main page again to refresh the simulation screen 

4.  System errors are automatically reported and will be fixed, please report any missing data where the system advises this and please use the 'Beta, report problem' from the main page to make us aware of any difficulties you are having with the system or changes that you would like to see made. 

5. AguaAndes/WaterWorld is a web application not a web site - do not bookmark pages and links as they may not work outside your current session.  If you want to keep a page use the permalink feature.

6.  Note that only one user can be logged into a given account at any one time.  If two users use the same account then there actions will conflict and lead to a very frustrating user experience for both.  All users should set up their own accounts.

7. It is only possible to run one simulation at a time.  If you define a new area or change to another run as a simulation is running then you will stop that simulation from running.


FULL USER GUIDE

VIDEO:

  Logging in and choosing the interface: Scientist and Licensed User, creating accounts (0h38m24s)

TEXT:

written by Mark Mulligan and Leonardo Saenz 

Your browser: Accessing the AguaAndes/WaterWorld Policy Support System  

AguaAndes/WaterWorld PSS is supported to run under Firefox and Chrome  web browsers only. You may be able to access AguaAndes PSS with other web browsers but success it is not guaranteed.

1. Start Firefox or Chrome. Firefox is freely available from http://www.mozilla-europe.org/en/firefox/ 

2. To get access to the System visit http://www1.policysupport.org/links/aguaandes

3. Allow pop ups from Agua Andes PSS if your browser requests this.

5. Enable Javascript in the unlikely event that it is not enabled in your browser.

6. Use Alt-Tab to navigate between the different pop-up windows as you interact with the system.

If you have problems see troubleshooting 

Before logging in 

An initial screen will allow you to log in to the system. Logging in allows us to store your simulations and provide you with full access to the system and simulation results.

To create an account use the Create account link.  If you have an account already provide your username and password in the fields stated. 

 

Creating an Account 

In order to sign up to Agua Andes PSS you need to hold a user account. To create an account, please follow the following steps.

  1. Click  the "Create an Account” link at the bottom of the control panel.

 

This option will open up the "Create a New Logging Account" screen in which you are requested to submit the following details.

 


2. The system then will send an email to the account you have entered. Please click the link you are provided with by email to authorize the activation of your account. This action will open up AguAAndes PSS and log you in.  If the email does not arrive within a few minutes check your spam folder and if it is not there contact us and we will confirm your account manually. 

Access levels 

You also need to indicate the kind of AguAAndes PSS user you are: there are two kinds of users: Policy Analyst and Scientist. Other user levels are licensed. In addition, you should select the language for your session (currently English or Spanish).

I am a:  

Though the same tools are used, the interface changes for different users so the level of detail shown is for example much greater when signed in as scientist compared with policy analyst.

  

Once logged in you are ready to use the PSS!

 



AguAAndes/WaterWorld functionality (Scientist interface) 

This section describes the functionality of AguAAndes from the perspective of the ‘scientist’ interface.

VIDEO:

Step 1 Define area (0m22s)

Step 2 Prepare data (2m52s)

Step 3 Start simulation (5m35s)

    Step 5 Results maps list (6m42s)

    The map view window (7m26s)

    View map by region (eg watershed, protected area) (7m47s)

    Viewing in Google maps/Earth (8m50s)

    Query points on a map and viewing the component inputs of a map (9m30s)

    Examining water balance (10m37s)

    Examining water quality (11m5s)

    Examining runoff (11m54s)

Step 4 Climate change scenarios (13m4s)

Step 6 and 7 Results maps and timeseries (23m57s)

Step 4 Policy Land use change scenario (24m46s)

TEXT:

Setting up a simulation

There are a series of instructions under the option "Set-up" of the control panel, which will allow you to set up a model simulation. These instructions are: define area and prepare data.


Land use change (documentation by Ruth Swetnam)
see here
Climate change (documentation by Arnout van Soesbergen)  

see here

EXAMPLE POLICY EXERCISE: Land use change

(documentation by Ruth Swetnam and Mark Mulligan) 

In order to run scenarios of change on your chosen landscape Steps 1 – 3 must already have been completed. If you have previously defined your study area, go to Manage Simulation, choose “Present your runs as a list”, Submit choice and then “Choose this baseline” to set your chosen area. Note that the map in the main screen may not change but the banner which indicates which simulation you are working with should, it is this which correctly shows which data is interacting with the model system.

If this banner does not update automatically, press the ( refresh) button

Once your baseline area is set, then run Step 2 “Prepare Data”. A window appears showing message “Maps in simterra database”, if you click on the + sign the full listing of inputs in the workspace should be listed. If the datasets are already in your database the tick boxes will be green, if there are any red crosses you need to click on the button “Copy data to your workspace” at the bottom of the list. If some maps are not available you need to click on ‘Build missing map tiles’ before you can continue.

The following options appear at the bottom of the screen. You have the option to list all the baseline workspace data which gives you the options to view map statistics and analyse maps. If all data is already available in your workspace you can click ‘Close window’ and continue

In the main window click Step 3:Start simulation - the model must have been run before a land use scenario is presented. This will open the Simulation window. Click on START and click yes to run the baseline simulation. If this tile has not been run previously, this process may take some hours. However, once this run has been completed it can be re-run very quickly (~ 6 mins). When all timesteps have been completed the bar will be green at 100%.

Once steps 1 to 3 have been run, you are then in a position to implement your POLICY EXERCISES by clicking on Step 4: policy exercises on the Control panel. This takes you to the Intervention Tool menu. If these steps are missed out, you will be reminded to re-run them when you try to implement scenarios of change later on.

Choose Land Use and Cover Change

Clicking on the Land Use and Cover Change option will open up the scenario window. If the data has not been prepared and copied over and the simulation ran, the system will give you a warning message here and will not take you any further until Stages 1 to 3 have been completed afresh. Once these have been done, the following window will appear:

There are two main approaches to constructing a land use change scenario, the first involves changing the amount of forest in the landscape either by afforesting or deforesting by a percentage area. The second approach allows different combinations of cover between tree / herb / bare as derived from the original MODIS vegetation input datasets. Each of these can in turn be tackled in two different ways: the first is to use pre-defined rules, the second allows you the user to create your own tailored rules. These will be discussed in turn with examples. It is however worth noting that by far the simplest approach is to use the predefined rules.

Using Pre-defined Rules:

In the first section, under the Forest to Pasture banner, click on the +symbol next to “Use a pre-defined rule” which opens the definitions boxes for input.

Firstly, give your scenario a new name which reflects its purpose (for example sitename_def50 for a deforestation scenario of 50%). As shown, -values will deforest the area, +values will reforest the area. These rules are implemented in turn from the top of the list to the bottom. One is not necessarily a subset of the other, so the logic may run as follows:

Example 1

Land outside protected areas (%)  -100 ------>  This alone would deforest all pixels outside any existing protected areas by 100% by converting all forest to herb.

Example 2

Land outside protected areas (%)  -100 AND

Land inside protected areas (%) +15   --------->  These two statements would mean that all pixels outside protected areas would be deforested by conversion of forest to herb AND then all pixels inside the existing protected areas would then be reforested by 15%. As these two zones of the map are mutually exclusive, both rules will be successfully implemented.

Example 3

Land outside protected areas (%) -50 AND

Land at High Elevations (%) +20        --------->  This would first of all deforest all land outside protected areas by 50% and then afforest any land inside the high elevation areas by 20%. So in this case  the second change could actually reafforest land outside protected areas on higher elevations.

Example 4

Land inside protected areas (%) +20 AND

Land near existing deforested areas = -30 --------->  In this case all land inside protected areas would be afforested by 20% BUT when the second rule is then applied land near to existing deforested areas (within 5 cells) would then be deforested by 30%..

It can be seen that some care needs to be taken to construct a scenario that is meaningful but the application of a number of rules would give considerable flexibility. In general, it is probably better to use a simple combination of rules which are easily understood. The simplest change of all would be to use the the bottom line for “All land” and enter one amount to either deforest or reforest over the whole landscape. Once your changes have been entered, click the Check and Submit button which will start the construction of the “new” landscape. A progress bar appears in a new window as the baseline maps are copied and the land use rules defined are then applied. When completed the following information appears:

At this point you have the option of comparing the input maps of the baseline as previously run and the scenario that you have just constructed. Clicking on the “Show baseline and scenario” will display the new inputs for the hydrological model so that you can examine the output of your land use decision rules on the landscape appearance. In the example below the scenario of change constructed,  deforested all lowland areas in Colombia by 50%, leaving high forest values on just the mountains which clearly stand out in the scenario map on the Right hand side.

Once you are happy with your landscape scenario, the hydrological model can then be rerun. So click Run scenario in the previous window or go back to Step 3: Start simulation in the main control panel. Check that the  banner lists the name of the scenario landscape, which in this case was deforlow (if this is not displayed, remember to refresh with the  button

Then re-run the simulation by clicking the Start Button and click yes. This will then trigger the system to re-run the hydrological model on the new “scenario” landscape. The results of which will then appear in the alternatives listing for this baseline landscape, with all the same output maps as previously created for the Baseline.

With the Forest to Pasture / Pasture to Forest approach it is also possible to construct a new scenario landscape by defining your own rules . Again, - values indicate deforestation, + values indicate reforestation. In this part of the system, you can limit these changes to particular areas of the map (such as particular Ecoregions, or elevations above 1000m, or to specific protected areas by entering the identifying number of the site). The drop down arrow  next to the box of the “where” statement allows the site to be specified and then a particular condition applied.

After constructing the new landscape the same process as listed above for the predefined rules is then repeated to re-run the hydrological model.

The second major approach to constructing land use change scenarios is the use specific values for the percentage of Tree / Herb and Bare land in each pixel.  So rather than simply reforesting or deforesting specific cover percentages can be applied.

The logic of the rules works in the same basic way as described previously for the predefined rules (i.e. top to bottom and additively). However, here actual percentages are entered.

Similarly, specific changes can be made to particular locations by changing these variables  under the “Define your own rule” section

You may apply these rules for all objects in a map (for example all protected areas) by choosing for example where Protected areas > 0 when defining the rule.  To apply to a single protected area, IBA or other feature you must find the unique ID number for that feature.  When you choose the map from the list after where the map will open in a small window.  By clicking the Google Maps icon underneath that map you can open a Google maps interface for the map.  By moving the map so that the white crosshairs in the centre of the map screen lie over the feature of interest you can then click Query to obtain its unique ID which you can then enter on the web form after this value: on the form.

With all four of these approaches the general method is the same. First you construct your “new” landscape (called a alternative) to your chosen baseline. Then you Prepare the Data, Re-run the simulation and at that point you can also compare the results of your land use simulation with the baseline.

USING YOUR OWN EXISTING LANDSCAPE MAPS

In order to make the model as flexible to existing inputs as possible, two additional routes to including map data are also now available to the user. At the bottom of the Land Use Change scenario window are two options to allow you to import your own maps.

The first allows you copy in a map  through an FTP site or other web storage site.

IMPORTANT - your map must be carefully specified and must have specific characteristics. Click on the + symbol next to the “maps must have these characteristics”. This will detail the details which need to match in the ascii file.

The maps must be in ARCASCII format and zipped to a .zip or gz file. The header of the ascii file must match that of the baseline tile you are working from. So the values shown in the grey box above will vary with the study site. ALL decimal places must match exactly.

The maps to be uploaded also need to have the correct filenames for the system to recognise them. By clicking on the + symbol next to the “maps must have these names” link, a full file listing is provided. So if you are using your own climate data / elevation data etc, they need to be given the correct input names.

If you do not have an available ftp server you can upload the same datasets by copying them onto the Waterworld server with the “upload your own maps” button. Note that you must have the Adobe flash player browser plugin installed to use the upload functionality (http://get.adobe.com/flashplayer/)

The same restrictions of file headers and names also applies here but you can click on the Select button of each functional type given above and a File Browser will open for you to locate the appropriate zip file. These must be selected and uploaded, one each for tree, herb and bare.

Finally, at the bottom of the Land use change scenario window there is a button which allows you to list all your alternative data, where the standard download, geobrowse and stats functions are still available to you.

EXAMPLE SCENARIO: Climate change (documentation by Arnout van Soesbergen)

Defining scenarios of climate change

To run scenarios of climate change, steps 1 to 3 must have been completed first. To return to a baseline simulation, go to ‘Manage simulations’ and decide on your own simulations or community simulations on 1 km or 1ha resolution and present runs as list. Click ‘submit choice’ and an overview of available runs will be given along with information on the run. Climate change scenarios can only be run if the run_status is completed for the baseline run. If this is not the case you will need to run this first by clicking ‘Choose this baseline’ and follow steps 2 and 3.

If the baseline run has been completed click on ‘Choose this baseline’ and close the pop up window. You are now ready to define a scenario in Step 4: Policy exercises. You can check at any time the details of your simulation (run, scenario, policy option and parameter set) by looking at the banner at the top of the main page. You may have to refresh the window by pressing the refresh button.

Click on Step 4: Policy exercises under the simulation tab and the intervention tool menu will open:

Choose Climate Change and Submit choice. This will take you to the scenario window:

If the baseline run is not completed, you will receive a warning here to run the baseline before attempting to do a scenario simulation:

If the baseline run is completed the following climate scenario window will open:

There are four options here to implement climate change scenarios.

Option 1:

The first option is to choose an existing IPCC downscaled GCM scenario. To use this option you need to choose the assessment, the scenario, the institute that executed the downscaling, the GCM and the time period of interest. Currently you can only choose output from the IPCC AR4 assessment for the A1B, A2A, A2 and B1 SRES emission scenarios. The downscaling institutes are CI (Conservation International), CIAT (Institute of Tropical Agriculture) or KCL. The number of GCM and the time period you can choose from depends on the scenario and downscaler. Currently the following combinations are available:

scenario

downscaler

nr of GCM

time periods

A2A

CIAT

17 models

2050s

A2

CI

20 models

2041-2060

2081-2100

B1

CI

21 models

2041-2060

2081-2100

A1B

CI

24 models

2041-2060

2081-2100

A1B

KCL

5 models

2041-2060

2081-2100

A2A

KCL

5 models

2041-2060

2081-2100

For the CIAT and CI downscaled data it is possible to use the individual GCM or a mean of all available models as well as the mean of all models + or - one standard deviation. For the KCL data, only the mean of 5 models can be used. 

Once you have chosen a scenario click on Submit:

The system will then set up the chosen scenario and copy the baseline over. This may take a few minutes. 

Option 2:

The second option is to use the new CIAS CLIMGEN scenarios (Community Integrated Assessment System). These scenarios can be imported either in NETCDF format or as ARCASCII format. To access the upload screen for these scenarios, click on the next to the CIAS CLIMGEN banner and do the same for either NETCDF or ARCASII. When NETCDF format is chosen, the following options are available: NETCDF format multiple variables per file or NETCDF format one variable per file. If multiple variables are chosen, the following data upload screen will be shown:

Data should be zipped using winzip (.zip) or gzip (.gz) and have to be accessible on an FTP or HTTP server. If a username and password are necessary provide these details, if not leave these to default (Anonymous and login name). You can also provide some details on this screen on the GCM used, name of the experiment and you can name your scenario.

NetCDF files can also be uploaded as one variable per file which enables the option of uploading separate NetCDF files for cloud cover, precipitation, tmin, tmax, vapour pressure and number of wet days. The following data upload screen will be shown if this option is chosen:

To upload data in ARCASCII format click on the next to ARCASCII format as described above and the following data upload screen will be shown:

To use the ARCASCII format both baseline and future files for precipitation and mean temperature have to be uploaded through FTP or HTTP.  The data should consist of one zipped (.zip or .gz) file containing 12 global monthly grids for baseline precipitation, baseline temperature, future precipitation and future temperature. Full url links for these four files have to be given in the correct boxes. Finally, the scenario can be named and data on the GCM, experiment and time period can be given.

Option 3:

The third option for climate change scenarios is to copy your own scenario. This option is a bit similar to option 2 in the sense that files for precipitation and temperature for a climate change scenario have to be uploaded through FTP or HTTP. To use this option click on the next to ‘..or copy your own scenario’ and the following data upload screen will open:

The data has to be provided as .zip or .gz with one file for precipitation containing 12 global monthly grids and one file for temperature containing 12 global monthly grids using the following filenames: wcprec.zip for the files wcprec1.asc to wcprec12.asc (for precipitation) and wctmean.zip for the files wctmean1.asc to wctmean12.asc (for temperature). Note that temperature data has to be in degrees * 10.

The full url to the zipped files must be given and if the FTP or HTTP server require a username and/or password these must be provided in the boxes. If not leave these options to default

IMPORTANT: The ascii files must be carefully specified and must have the following characteristics:


Option 4:

The fourth option is to develop your own simple scenario. Click on the next to ‘...or develop your own simple scenario’ and the following input screen will open:

Here you can develop your own simple scenario by applying seasonal uniform temperature (in degrees Celsius) and precipitation (as percentage) changes. Give the scenario a name and click ‘Check and Submit’

Running the model and looking at results

Once a scenario is defined (option 4) or all information is provided (options 1 to 3) click on ‘Check and Submit’. The system will now start building the scenario. A progress bar will appear in a new window and the baseline will be copied over. This could take a few minutes depending on the kind of scenario defined. Errors may occur particularly for options 2 and 3 when the system needs to download information from FTP or HTTP servers. These errors could relate to access errors -check username/password for servers- or specification of the files -make sure they exactly match the correct file names and specifications with the exact decimal places-.

When the system has finished building the scenario, the following screen will appear:

 

Here you have the option to compare the input maps of the baseline and the scenario that you just created, to run the model for the climate change scenario or to have a look at the new input data. By clicking on ‘Show baseline and scenario’ you can compare input maps of the baseline and the created scenario. This will open a list of monthly maps that have changed under the scenario. By clicking on the sign next to the variable and month of interest, statistics, maps and histograms of the baseline and the scenario will be shown. For example a change of mean monthly precipitation for January under a simple scenario of - 10% uniform change:

The histograms can be downloaded as excel files by clicking on the symbol under the histograms. Also, the maps can be shown in a separate window by clicking or as an overlay in Google Earth by clicking  or as an overlay in Google Maps by clicking

Another option to look at the output is to click the analyse buttonthis will open up a window where you can plot output of different maps against each other.  

Results can also be viewed as stats in Step 6 from the main menu where time series of the main results can be compared. In the example below, the difference between input precipitation for the baseline and the scenario is plotted:

Finally, results can be given as a narrative in Step 7 where clicking on the highlighted ‘show’ will fill in the numbers in the narrative or clicking on Show all will fill in all the values in the narrative.

Mobile use of data offline (in the field)

Having viewed results in Google Earth, these can be saved (from the Google Earth link called ‘offline’) for offline viewing (for example in the field) using Google Earth mobile on Android smartphones and tablet computers in combination with the KMLZ to Earth app .  You can use the Wifi file transfer or AirDroid apps to upload the kmz files to your device.

The climate change scenario is now defined and can be run by clicking Run scenario in the previous window (the run will start straight away) or by clicking Step 3: Run Simulation in the main window. Do check that the banner displays the correct details (remember to refresh using the  button:

If the run is started by clicking on  ‘Step 3: start simulation’ , click yes when asked to overwrite the existing maps. The run will take about 6 minutes to complete.

Once the run is finished, you can either look at all results by clicking ‘View all results’ in the model run window or look only at the map results by clicking on ‘step 5: Results: maps’ This will open up a window with a list of output maps for the run with the options to download the maps, to geobrowse, compare to baseline, show stats or analyse. By default a list of the 4 main system state maps will be displayed. All output maps are shown after clicking on the + sign next to All maps

When clicking on , a window will open with three options. 1) the actual values for the scenario run and the baseline run,  2) the difference between the scenario and the baseline and 3) the % change from the baseline. The first option will display the baseline and scenario output map next to each other along with statistics and histograms for both maps. Options 2 and 3 will show single difference maps. By default the map for option 2, the difference between the scenario and the baseline will be shown. Other options can be viewed by clicking the + next to the description. The example below is for the change in annual total water balance using the simple change scenario described above:

Manage simulations (Scientist) 

Under manage simulations you can change the current simulation to a previous (baseline or alternative) simulation, view the number of alternatives (scenarios and policy options) associated with a particular baseline or delete previous simulations that you no longer need.  Use this option to ensure that you have no more than three simulations stored at a time.

Step 5: Results : maps (Scientist) 

Here you can visualise, compare and download results of the system in map form.

Step 6: Results : stats (Scientist) 

Here you can visualise, compare and download results of the system in timeseries form.

Step 7: Results: narrative (Scientist) 

Here the system provides a narrative on the outcomes of a particular baseline or scenario simulation.

Redefine land use and cover according to your own map

 

Use this function to use your own, or the in-built, land cover and use classification to assign land cover and land use characteristics used for the alternative simulation. Choose the land use classification to use from the drop-down list (for your own map, upload this as an AOI first and refresh this page) and then complete the table for the codes in this map that apply to any of the listed land use and cover types used by costingnature. Leave the - in the 'codes' column to leave these land cover/use maps unchanged. Where you set codes, also set the values to be used for each land cover/use variable in areas where the cover class occurs. Use a - i n the 'Values' column to indicate that these values should be taken from the in-built baseline variable values for areas of the AOI in which the class occurs. You may set more than one map code to separate values for a given variable by separating the codes and corresponding values by commas. Be sure that cropland+pasture<=100 and tree+herb+bare=100 and that at least one land use has values >0 for each code. A given pixel may have more than one land use eg a combination of natural and protected or cropland and pasture.

For example here class 1027 is forest, 800 is cropland, 1133 is urban in my uploaded land cover class AOI- I set the land use and cover for each class as below:

 


AguAAndes/WaterWorld functionality (Hyperuser interface) 

WaterWorld - Step 1 Define area (brief) (0h55m42s)

Demo Step 2: Prepare data (brief) (0h57m16s)

    Mask by Watersheds of points (0h58m15s)

    WW Documentation and Help (1h0m5s)

Step 3: Start Simulation - write maps option - default is annual outputs for quicker runtime (1h3m22s)

    Manage simulations (1h4m28s)

Step 5: WW Results list: monthly maps view and download (1h5m50s)

    examining water balance. View by function (1h8m41s)

    POIs ROIs and ZOIs (1h11m10s)

    Other options - Analyse map - scatterplot and stats (1h12m54s)

Step 4: Policy Exercises - Climate Change (1h14m53s)

Step 4: Policy Exercises - Show baseline and scenario, List alternative workspace data and compare to baseline (1h17m9s)

Step 4: Policy Exercises - Land Cover and Land Use Change (1h20m57s)

Step 4: Policy Exercises - Land and Water Management, bench terraces, scenario by uploading your own maps (1h27m27s)

Step 4: Policy Exercises - Replace input maps (1h28m7s)

Step 4: Policy Exercises - extractive Scenarios for mining concessions and Oil and Gas. (1h29m15s)

Step 4: Policy Exercises - Population growth and urbanisation scenarios (1h29m54s)

Step 5: Results of climate change scenarios: changes from baseline (1h32m29s)

Free version and licensed features (1h36m15s)



AguAAndes/WaterWorld functionality (Megauser interface) 

Step 1 Define area (0m40s)

Step 2 Prepare data (1m58s)

    Masking an area by a region of interest (ROI) (2m28s)

    Investigating input maps - show workspace data (6m33s)

Step 3 Start simulation (7m31s)

    Write monthly output maps option (7m44s)

    Selecting previous runs and managing simulations (8m45s)

Step 5 Results maps list (9m29s)

    Examining the water balance output map (11m8s)

    Query input components for a map at a point (12m40s)

    View by function (14m8s)

    Define Regions of Interest (ROIs ) (14m37s)

    Change image (scales and colours) (15m20s)

    Example - Runoff - downloading maps and visulaising in Google Earth (17m33s)

Step 4 Policy exercises (19m17s)

    Setting up Extractives (mining and oil and gas) scenario (20m3s)

Step 5 Examining the results of scenario (26m30s)


Ecoengine Functions by step (links to video demonstrations)

This section describes the generic functionality of the ecoengine framework and how to use it.

Step 1: Define area

Step 1 Define area (0m40s)

Step 2: Prepare Data

Use alternative SIMTERRA input maps (0h0m40s)

Masking data (1h23m56s)

Masking an area by a region of interest (ROI) (2m28s)

Mask by Watersheds of points (0h58m15s)

Investigating input maps - show workspace data (6m33s)

Examining the input data, for example Landsat 2000 Treecover (1h27m14s)

Set your own land use data for baseline using % values (0h4m49s)

Set your own land use data for baseline using system values (-) (0h11m46s)

Step 3: Start simulation

Write monthly output maps option (7m44s)

Manage Simulations

Selecting previous runs and managing simulations (8m45s)

Deleting old simulations to free up space (1h20m55s)

Looking at a previously saved run. Looking at monthly impacts (1h55m31s)

Step 4: Policy exercises

Climate Change scenarios(1h14m53s)

Land Cover and Land Use Change (1h20m57s)

Run land use change model (including multiple deforestation monitoring products and management effectiveness) (0h6m55s)

Land and Water Management options (1h53m18s)

Land and Water Management, bench terraces, scenario by uploading your own maps (1h27m27s)

Extractive Scenarios for mining concessions and Oil and Gas. (1h29m15s)

Population growth and urbanisation scenarios (1h29m54s)

The population growth scenario tool uses a gridded projected rate of growth from IPCC SRES B2 scenario to 2025[1] and allows the user to increase or decrease this rate by a factor eg 1.2 or 0.8 across the study area and/or add a given percent change per year for a set number of years.  To apply the IPCC projection as is (for 50 years) leave the values as their defaults.  To add or subtract a given % change per year, change the 0 %/yr to another number.  The projection can be applied across the entire study area (the default) or to parts of the study area (based on a where rule).  Additional in-migration to the area can also be specified on the basis of a percent of the study area population per year and is distributed according to accessibility.  This can simulate new migration patterns along new roads, for example.  A rural to urban migration could be simulated by simulating in-migration, where urban areas >0, by a given percentage per year.

Change input maps (1h54m10s), Replace input maps (1h28m7s)

Show baseline and scenario, List alternative workspace data and compare to baseline (1h17m9s)

Run land use change model - deforestation (0h0m0s)

Run progressive deforestation scenarios (0h7m38s)

Compare alternatives (compares baseline and multiple alternatives for a given map) 

Set your own land use data for scenario through Step 4: policy exercise (0h13m57s)

Step 5: Results maps

Results list (1h40m8s)

The map view window (7m26s)

Examining water balance (10m37s)

Examining water quality (11m5s)

Examining runoff (11m54s)

Viewing in Google maps/Earth (8m50s)

Query input components for a map at a point (12m40s)

Examining change from the baseline (2h2m57s)

Examining monthly maps, view and download (1h5m50s)

Citation information (1h39m5s)

Permalink for export of (google) maps (1h48m29s)

Analyse map - scatterplot and stats (1h49m39s)

POIs ROIs and ZOIs (1h11m10s)

Regions of interest (copy/paste points of interest and show on output maps) (0h18m30s)

Create ZOI with bounding box (0h0m0s)

Step 6: Results stats

Results time series (1h51m41s)

Step 7: Results narrative

Results narrative (1h52m31s)

Questions and Answers

Running the model over multiple tiles and combining using ArcMap and ArcHydro or SagaGIS (1h32m36s)

Run times in new areas (1h38m43s)

Free version and licensed features (1h36m15s)



GUIDANCE FOR PRE- and POST- PROCESSING OF DATA

Conducting Hydrological Analysis Over Multiple Tiles (by Louis Parker)

Hydrological analysis in WaterWorld is conducted on single tiles at 1ha or 1km spatial resolution. It is possible however to conduct analysis over multiple 1 ha or 1km resolution tiles using the hydrology functions in the ArcGIS ArcHydro extension. This may be necessary when the catchment or area of interest is located over multiple tiles in the PSS, for example, a river basin may cross numerous tiles and the emphasis may be on understanding flow over the entire catchment.

Downloading and installing Arc Hydro

Arc Hydro can be downloaded from the ESRI website. Please follow this link:

http://support.esri.com/en/downloads/datamodto ss l/detail/15

Download the Arc Hydro edition which is compatiblewith your ArcGIS.

If you are using ArcGIS 10 then please follow this link to download Arc Hydro Tools 2.0:

http://blogs.esri.com/esri/arcgis/2011/10/12/arc-hydro-tools-version-2-0-are-now-available/

 

Arc Hydro Functionality

To understand the range of options available please download the ‘Arc Hydro Data Model and Tools’ which can be acquired from the ESRI website:

http://resources.arcgis.com/content/hydro/surface-water/about

 

Preliminary requirements

Please note that the user must have selected the area of interest, prepared the data, and run the baseline simulation (please see steps 1-3 of the System Documentation). Water World enables a maximum of 4 tiles (simulations) to be stored at any one time.

Step 1

Navigate to the ‘Results: Maps’ section in Water World. Open the ‘Water Balance’ map; please note that this will be used later to create a weighted run-off grid in ArcGIS.

Download the map as an ‘ARCASCII’ file and open this in ArcGIS. ‘Export Data’ and save as a grid file and set the cell size, this must be the same across all the tiles downloaded from WaterWorld. Repeat this process for all the tiles that cover the area of interest (i.e. the river basin).

Step 2

A Digital Elevation Model (DEM) will be used in ARCGIS to create a runoff map from the water balance data. A 90m resolution DEM developed by the NASA Shuttle Radar Topographic Mission (SRTM) which has been hole-filled by the CGIAR can be downloaded from the King’s College London geodata portal mirror. Please follow this link to download the aforementioned CSI-SRTM version 4.1: http://geodata.policysupport.org/srtm

Open the DEM data in Google Earth. Move to your area of interest (i.e. the river basin), click the centre of the tile to open the information on the DEM. Click the link to visualise the tile and access 1 degree tiles.

Then download the 1 degree tiles of interest as an ARCASCII and open in your arc map project.

Step 3

a)      Merge the DEM

To create a runoff map the DEM raster grids will need to be merged. Information on the steps to complete this are available from the ESRI support page (link provided below). Repeat the merging process for the water balance tiles acquired WaterWorld.

http://support.esri.com/en/knowledgebase/techarticles/detail/20988

Once the DEM’s have been merged you are now ready to begin creating a runoff map which spans multiple tiles.

b)      Identify Sinks

Expand the Spatial Analyst tools box, locate the Hydrology tool box.

Flow across the DEM occurs from the steepest downslope direction (ESRI,2008), it is however possible that there are cells which are lower than all of the the surrounding cells (sinks) and will therefore trap water, these areas can be real or due to an DEM errors (ESRI, 2008). It is therefore necessary to identify these areas which are known as sinks.

 Locate the Sink tool (found in the hydrology tool box) and enter the DEM. This will produce a map of the sinks, if there are areas which the user knows are natural sinks then these locations can be identified and separated, those sinks that are due to errors in the DEM must be filled.

c)      Fill the DEM

The DEM is filled to ensure that sinks (the depressions caused by errors in the DEM) do not interrupt the transfer of water.

Use the FILL function found in the hydrology tool box, enter the DEM. Save the output DEM, for example DEM_Fill. This will be the DEM which is used for the rest of the analysis.

d)      Flow Direction

Now that the DEM is filled, it is necessary to create a flow direction map.

Select the Flow Direction function in the Hydrology tool box. Enter the filled DEM (DEM_Fill) as the ‘input surface raster’ and save the output file to your working directory.

The flow direction grid should have a total of 8 values, these are 1, 2, 4, 8, 16, 32, 64, and 128. These 8 values represent the different flow directions, they are the 8 adjacent cells into which the water can flow (Berkeley, 2012).

 

e)      Flow Accumulation

The Flow Accumulation function enables the creation of a runoff map from the water balance data. It produces a map displaying the amount of water that would flow into each cell (ESRI, 2008b). The flow accumulation of a cell is the number of cells that are flowing into it. Areas which have high flow accumulation are the stream channels.

To create a Flow Accumulation map open the Hydrology tool kit

-          select Flow Accumulation

-          Select the flow direction map

-          Save the map to your working directory

-          Select your Water Balance map which was acquired from WaterWorld as the ‘input weight raster’

-          Press ok

The flow accumulation map should resemble a runoff map, with stream channels of higher flow noticeable.

 

It is important to note that the flow accumulation function in Arc Map enables a weighted flow accumulation map to be produced. We have used the water balance grid (Precipitation- evapotranspiration) acquired from WaterWorld to produce a weighted Flow Accumulation grid. If no weight is selected, each cell is given a default value of one. Using the Water Balance map as the weight ensures that each cell has a value (weight) that is appropriate for runoff generation.

 

References

ESRI (2008) ArcGIS 9.2 Desktop Help, Flow Direction [online] Available from: http://webhelp.esri.com/arcgisdesktop/9.2/index.cfm?TopicName=Flow_Direction [Accessed on 15th of July 2012]

ESRI (2008b) ArcGIS 9.2 Desktop Help, Flow Accumulation [online] Available from: http://webhelp.esri.com/arcgisdesktop/9.2/index.cfm?TopicName=Flow_Accumulation [Accessed on 15th of July 2012]

Berkeley (2012) Hydrological Analysis with ArcGIS [online] Available from: http://gif.berkeley.edu/CE/Summer2007/GISanalysis_Exercise_August22_UCCE_Hydro.pdf [Accessed on 15th of July 2012]


Creating a Zone of Interest (ZOI) map from a  shapefile using SAGAGIS (by Mark Mulligan)

- Download and install the free and open source SAGAGIS GIS (if you do not already have it).  You can also install SAGAGIS before QGIS and then use the SAGAGIS functions from QGIS

- Download any of the available input or output maps in ARCASCII format to use as a template and unzip it.

- Use your file manager to pick up the ARCASCII file and drop it into the SAGAGIS window.

- Ensure that your shapefile is in the Geographical Coordinate System (GCS, latitude longitude) and uses a WGS84 DATUM (Use Modules > Projection - Proj.4 > Coordinate Transformation (Shapes) to reproject if necessary)

- Use your file manager to pick up the file with .shp extension in your polygon shapefile and drop it into the SAGAGIS window

- Check that the grid and the shapefile overlap by double clicking the name of the grid under the  tab and choosing to display in a new window.  Then go back to the  tab and double click the name of the shapefile and choose to display in the window named as your grid is.  Does the shapefile overlay the grid?  If not then you may need to run another tile in the PSS or check that the shapefile is really in The geographic coordinate system (GCS) and WGS84 datum.

- In SAGAGIS from choose the Grid-gridding collection and double click the Shapes to Grid module.

- Under Shapes choose the name of your loaded shapefile

- Under attribute choose the field that will be used (choose any since we will convert the grid to zero (outside)  and 1 (inside) the area of interest

- Under Preferred target grid type choose integer (1 byte)

- Under Target grid choose the grid that you have loaded (to do this choose grid here and you will be asked to define the grid system in the next step)

- Click Okay. In the next window, you will be asked to choose the Grid system, so, choose the only one available, which is the system of the grid you loaded and to which the shapefile must be converted.

- Under Number of Values choose [create], and then click Okay.

- Now go to the tab and double click in the new grid with the [Count] tag, which will be named the same as your shapefile.  Choose New to display in new map.  You should see your shapefile shape as a grid with values of 1 where the area of interest is, and 0 elsewhere (it may be clipped to the edge of the tile if it extends beyond it).  This will now be used to set your ZOI in the PSS. If you see a blank map then either your shapefile is not a polygon or it is in a different coordinate system to the GCS, WGS84 of the grid.  Fix it and try again.

If your shapefile has small polygons which are lost on conversion to grid cells then you need to download the latest version of SAGAGIS (at least 2.1.1 here) and when using Shapes to Grid choose output values as data/no-data, polygon should be set to cell and fit to cells  should be ticked.  Note that you will not need to install SAGAGIS, just download the zip file, extract it and run the file called SAGA_GUI.exe.

- You now need to save the map as a zipped ARCASCII to include it in the PSS. Go back to  and choose Import/Export - Grids and then double-click Export ESRI ARC/Info Grid.  

- Under grid system choose your only available grid system.  

- Under grid choose the name of the grid that you created with the the [Count] tag at the end.  

- Under file choose a folder and file. Make sure to include the .asc extension. Click Okay and the file will be written.

- Under ASCII Precision choose 0.  

- In your file manager (e.g. Windows Explorer) find the .asc file and zip it, giving the .zip file the same name as the .asc file.  This can now be used to set a ZOI using upload ZOI in the PSS.


Creating a PSS grid from a  shapefile using SAGAGIS (by Mark Mulligan)

- Download and install the free and open source SAGAGIS GIS (if you do not already have it).  You can also install SAGAGIS before QGIS and then use the SAGAGIS functions from QGIS

- Download any of the available input or output maps in ARCASCII format to use as a template and unzip it.

- Use your file manager to pick up the ARCASCII file and drop it into the SAGAGIS window.

- Ensure that your shapefile is in the Geographical Coordinate System (GCS, latitude longitude) and uses a WGS84 DATUM (Use Modules > Projection - Proj.4 > Coordinate Transformation (Shapes) to reproject if necessary)

- Use your file manager to pick up the file with .shp extension in your polygon shapefile and drop it into the SAGAGIS window

- Check that the grid and the shapefile overlap by double clicking the name of the grid under the  tab and choosing to display in a new window.  Then go back to the  tab and double click the name of the shapefile and choose to display in the window named as your grid is.  Does the shapefile overlay the grid?  If not then you may need to run another tile in the PSS or check that the shapefile is really in The geographic coordinate system (GCS) and WGS84 datum.

-Right click on your shapefile and choose Attributes>Show and ensure that the shapefile has a numeric field that contains the information you want to write to the grid (text fields cannot be written to grids).  To convert a text field of eg polygon names to a numeric field from choose Table-Tools-Enumerate Table field.  Choose your shapefile, the name of the attribute that you want to enumerate and choose the shapefile again as the output file.  Click Okay.  A new field called ENUM_ID will be added to your shapefile

- In SAGAGIS from choose the Grid-gridding collection and double click the Shapes to Grid module.

- Under Shapes choose the name of your loaded shapefile

- Under attribute choose the field that will be used to write data to the grid (the enum_id field if you had to create one)

- Under output values choose attribute

- Under Preferred target grid type choose the correct data type (floating point for scalar quantities and Integer for integers)

- Under Target grid choose the grid that you have loaded (to do this choose grid here and you will be asked to define the grid system in the next step)

- Click Okay. In the next window, you will be asked to choose the Grid system, so, choose the only one available, which is the system of the grid you loaded and to which the shapefile must be converted.

- Now go to the tab and double click in the new grid with the [FIELD] tag, where FIELD is the field that you chose to transfer.  The gridwill be named the same as your shapefile.  Choose New to display in new map.  You should see your shapefile shape as a grid with.  If you see a blank map then either your shapefile is not a polygon or it is in a different coordinate system to the GCS, WGS84 of the grid.  Fix it and try again. If you see only a few features then you had selected those in an earlier step - go back to your shapefile map and choose  to draw a box around all the features to select them all and then repeat the process.

If your shapefile has small polygons which are lost on conversion to grid cells then you need to download the latest version of SAGAGIS (at least 2.1.1 here) and when using Shapes to Grid choose output values as data/no-data, polygon should be set to cell and fit to cells  should be ticked.  Note that you will not need to install SAGAGIS, just download the zip file, extract it and run the file called SAGA_GUI.exe.

- You now need to save the map as a zipped ARCASCII to include it in the PSS. Go back to  and choose Import/Export - Grids and then double-click Export ESRI ARC/Info Grid.  

- Under grid system choose your only available grid system.  

- Under grid choose the name of the grid that you created with the the [FIELD] tag at the end.  

- Under file choose a folder and file. Make sure to include the .asc extension. Click Okay and the file will be written.

- Under ASCII Precision choose 0 if your data is integer.  

- In your file manager (e.g. Windows Explorer) find the .asc file and zip it, giving the .zip file the same name as the .asc file.  This can now be used to set a ZOI using upload ZOI in the PSS.


Combining tiles and clipping them to a shapefile in SAGAGIS (by Mark Mulligan)

If your study area covers more than one PSS tile you may need to combine tiles and then clip them to an area of interest such as a watershed

Download and install the free and open source SAGAGIS GIS (if you do not already have it).

Download any of the available input or output maps in ARCASCII format and unzip them.

Use your file manager to pick up the ARCASCII file and drop it into the SAGAGIS window.

Ensure that your shapefile is in the Geographical Coordinate System (GCS, latitude longitude) and uses a WGS84 DATUM

Use your file manager to pick up the file with .shp extension in your polygon shapefile and drop it into the SAGAGIS window

In SAGAGIS from choose the Grid-Merging.  Choose the grids to merge.  Under “target-grid” give a name to the grid that you wish to create from the merged grids.  Click Okay.  The grids will load and merge.  Now go to the tab and double click the new grid and choose New to view it and check that it is Ok.  

To clip the grid, from the  tab choose Shapes-Grid, Clip grid with Polygon.  Choose the grid system of your merged grid (it will be the one with the most rows and columns.  Under Input choose the name of the merged grid.  Under Polygons choose the name of the shapefile.  Click Okay and your merged grid will be cut to the polygon shapefile.

You must now export this merged, clipped grid.  Under  choose the Import/Export - Grids and double-click Export ESRI Arc/Info Grid.  Choose the grid system of your merged grid.  If you are unsure go to the  tab and right click the grids that you no longer need and choose close to close them and remove their grid systems from view.  Choose the grid to be exported.  Under File choose a location to save the exported ARCASCII file and click Okay.  The merged, clipped grid will be saved.


Ls -Cutting and resampling your gridded data so that it is compatible with the PSS and ready for upload in SAGAGIS (by Mark Mulligan)

If your gridded data are not the same cellsize or extent as those in the tile that you are running in the PSS, so long as your data are in the same coordinate system (GCS WGS84) and cover the tile area, you can use SAGAGIS to resample them so they can be uploaded.  

- Download and install the free and open source SAGAGIS GIS (if you do not already have it).

- Download any one of the PSS input or output maps that you will use as a template map and unzip the ARCASCII

- Use your file manager to pick up the ARCASCII file and drop it into the SAGAGIS window.

-Open your own gridded dataset in SAGAGIS by drag and drop or using the import functions

- From the  tab choose Grid-Tools, Resampling

- In the dialogue that appears, for GRID, choose the grid system and grid for  your data grid that you want to resample

- For PRESERVE DATA TYPE, uncheck the box

- For TARGET GRID, choose GRID (which means we will name a grid later). Click OK and this window closes.

- A choose grid window then opens.  Here choose the grid system for the template grid downloaded from the PSS.  For grid choose create as we will build a new grid for the output resampled data. Click Okay.  If your grids do not overlap at all then SAGAGIS will not ask any more questions and will not be able to produce useful output (you may see a blue grid but it will be of no use).  

If your data do overlap,  a new window opens in which you need to specify the resampling method. For continuous data, f there are no no-data areas, then use BICUBIC SPLINE INTERPOLATION.  If there are some no-data areas if you are resampling to a grid cell size that is very close to the original size, then choose NEAREST NEIGHBOUR to avoid the creation of false cell values in no-data areas.  If you are resampling to a radically different cell size, then use an interpolation method like BICUBIC SPLINE INTERPOLATION.  If the data are a classified map (discrete data) such as land use then choose Majority as we want to retain the same classes.  Click Okay.

- You will now see a resampled new map under the grid system associated with the map that you downloaded from the PSS.  It should be a square tile with the same footprint as the PSS tile that you downloaded.

If your original data are higher spatial resolution than the PSS template and this process leads to loss of features then instead of using interpolation choose Maximum value as the interpolation method.

You must now export this resampled, clipped grid.  Under  choose the Import/Export - Grids and double-click Export ESRI Arc/Info Grid.  Choose the grid system of your merged grid.  If you are unsure go to the  tab and right click the grids that you no longer need and choose close to close them and remove their grid systems from view.  Choose the grid to be exported.  Under File choose a location to save the exported ARCASCII file and click Okay.  The merged, clipped grid will be saved.


.


[1] (Center for International Earth Science Information Network (CIESIN) 2002. Country-level Population and Downscaled Projections based on the B2 Scenario 1990-2100 [digital version]. Palisades NY: CIESIN Columbia University. Available at http://www.ciesin.columbia.edu/datasets/downscaled.)