Dear Forum members, I am trying to analyse data on elephant movement patterns to test whether they prefer waterpoints with a particular spatial configuration. I have used the Multidistance Cluster Analysis tool to identify the scale(s) of waterpoint clustering, and then performed a Cluster and Outlier analysis (Anselin's Local Morans I) on the water spatial distribution to identify areas with clusters vs dispersed waterpoints. All fine there!
But now I need to test whether elephant entry points into the riverine area that contains these waterpoints is clustered/favoured towards areas with clustered waterpoints, whether they enter as close as they can to any source of surface water, or whether water actually doesn't determine where they enter at all. I have GPS co-ordinates of these entry points over a 3-year period, for both elephant bulls and elephant cows. I would like to therefore also see whether these movement patterns differ between genders, and also whether it makes a difference whether the waterpoint is natural or artificially provided.
I have been looking into Geographically Weighted Regression to do all of this, but my sample size might be slight small (about 150 points), and there are so many variables that I'm a little confused. Also, I don't want to waste lots of time learning GWR if I'm actually barking up the wrong tree here!
I'd appreciate any advice on whether I should follow GWR or if there is some other method (or series of methods) I should be using.
Looking forward to hearing from you!
Angela
But now I need to test whether elephant entry points into the riverine area that contains these waterpoints is clustered/favoured towards areas with clustered waterpoints, whether they enter as close as they can to any source of surface water, or whether water actually doesn't determine where they enter at all. I have GPS co-ordinates of these entry points over a 3-year period, for both elephant bulls and elephant cows. I would like to therefore also see whether these movement patterns differ between genders, and also whether it makes a difference whether the waterpoint is natural or artificially provided.
I have been looking into Geographically Weighted Regression to do all of this, but my sample size might be slight small (about 150 points), and there are so many variables that I'm a little confused. Also, I don't want to waste lots of time learning GWR if I'm actually barking up the wrong tree here!
I'd appreciate any advice on whether I should follow GWR or if there is some other method (or series of methods) I should be using.
Looking forward to hearing from you!
Angela