Calculating daily, weekly, monthly home range values, please help

[Adam McLain]

Hello All,

Relatively new GIS'er here. My lab has Arc 10.2, but we did not purchase the full package, only the basic version. I also have the R packages "adehabitat" and "move", and a copy of QGIS which I've only begun playing around with today. What I am working with is a large data set with four individual aye-ayes, all observed over a multi-year period. I would like to generate daily, weekly, and monthly average homerange values for these animals, meters squared would be fine (or pretty much any other unit of measure). I have spent the last couple of weeks playing around with the R packages with very little success. Can someone please recommend the best way to generate the data that I want? This is part of a post-doctoral research project, and I'm mostly trained in molecular work, not GIS, so any very basic help anyone could give me would be very helpful. Where do I start, what's the best way to go, etc.? Is kernel density superior to the maximum convex polygon method or vise-versa? Which one would work best for a dataset like mine... any help is very appreciated!!

Thank you,

Adam

[GIS in Ecology]

Hi Adam,

 

First, sorry for taking so long to moderate your post (I was away from my computer for a few days).

 

In terms of your actual questions about estimating home ranges, in general maaximum convex polygons (MCPs) will help you identify every area used by your animals, but it will give you no information about how the home range itself is used. Thus, if your animals use some areas of their home ranges more than others, you won't be able to tell this from an MCP. In contrast, a kernel density estimate (KDE) will identify how the home range is used, and in particular, will help identify core areas of the range (i.e. the areas which are used most often). In addition, because MCPs simplly draw a line around the outside of all the points where an individual was recorded, it will be much mroe open to bias from outlier locations which don't represent part of the usual home range. As a result, it is usually worth doing both MCPs and KDEs as this will give you the most information about the home ranges occupied by individual animals, with the MCP telling you everwhere that was used while the KDE showing you the usual home range (usually the 95% percentage volume contour of the KDE as this will exclude any outliers) and the core home range (usualy the 50% percentage volume contour of the KDE).

 

As long as you havee a sufficient number of locations, you can calculate both your MCP and KDE for any time period you want. In general, though, you would want to be working with a minimum of 30 - 50 locations, and this might limit th shortest time periods you can analyse.

 

In terms of actually calculating the MCPs and the KDEs, there are a number of possibilities. As you hae mentioned, you can use the adehabitat tool sett for R. However, if yuou are new to GIS, you are likely to struggle with this because of basic issues, such as sorting out projections and coordinate systems for your data, and making sure that everything uses the same ones.  In terms of other options, doing it in ArcGIS is probaly easiest and has the most readily available step by step instructions available (such as my book 'An Introduction To Using GIS In Marine Biology: Supplementary Workbook Four'. However, if your lab doess not have a licence for the spatial analyst extension, this might not be an option (it is work checking if you do have the licence for the spatial analyst extension or not).

 

In this case, you can use QGIS to do home range analyses. I haven't used it myself, but it looks like there is a specific plug-in for estimating home ranges in QGIS called Animove (see http://www.faunalia.eu/en/animove.html#animove-for-qgis for more details). You can find a video which demonstrates how to do this here: http://www.youtube.com/watch?v=98ORgGX8vSM.

 

There is also another plug in, called Home Range Analysis which allows QGIS to interact with R and Adehabitat to generate home ranges (see http://hub.qgis.org/projects/quantum-gis/wiki/HomeRange_plugin for details). It can be tricky getting all the different software packages talking to each other properly. However, if you are wanting to compare your home ranges to those created by others in R using Adehabitat, this might be the weay forward. This is because Adehabitat uses different algorhythms to generate the KDEs than the Animove pluig in (and indeed different from the ones used by ArcGIS).

 

If you are a complete novice GIS user, the place where you are most likely to become unstuck is the issue of projections/coordinate systems. You will need t ensure that all your data are in a suitable projection/coordinate system, and in the same projection/coordinate system before you start. You will have to use one where the map unit is metres (i.e. a projected one and not a geographic one which uses decimal degrees as the map units). In addition, if you have recordced your data using GPS/satellite telemetry based on the WGS 1984 datum, you may need to transform it into a different datum if you choose to use an existing local, national or regional projection/coodinate system (an example of this here in the UK is the British National Grid, which is based on the OSBG1936 datum). If you want to find out more about what projection/coordinate system are, you can look at the FAQs on the GIS In Ecology webpage (http://www.gisinecology.com/faqs__frequently_asked_question.htm) and you can find out more abotu selecting a projection/coordinate system in another post on this  forum (https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&frm=1&source=web&cd=4&cad=rja&uact=8&ved=0CEUQFjAD&url=https%3A%2F%2Fgroups.google.com%2Fd%2Fmsg%2Fgis-in-ecology-forum%2F9l67V6nMRoA%2FHGl0xqu1uisJ&ei=8rpfU-qrJ4SbPYmpgQg&usg=AFQjCNEMC87g5fHsV_RUEEOAusMXVZg9ZQ).

 

I would say that before you even try a home range analysis, it is important that you get your head round the issues of projections and coordinate systems. If you do, it will make it much less likely that you will have problems that you cannot find a solution to. To help with this, even if you aer using the QGIS plug ins or adehabitat, I would suggest that it might be worth checking out the supplementary workbook I mentioned above. Even though it is for marine biologists and uses ArcGIS as its example software to illustrate how to do the exercises, you will need to go through the same basic processes regardless of the software you use. You can find out more about this workbook here http://www.gisinecology.com/marine_supplementary_book4.htm.

 

So hopefully this provides you with enough information to allow you to select a suitable software package and to get started. If you have any further questions (or if any of the above is unclear), or if you run into any further problems, feel free to post again on this thread.

 

Good luck with your project. 

 

All the best,

 

Colin