Hello, I am doing my MSc thesis research about Eurasian nightjar in Slovenia. I want to analyse the habitat around playback points. I will take a buffer of 400 m and each point and compare the different types of habitats (meadow, shrubs, woodland,...). The problem is my points are sometimes less than 800 m apart so I get a zone where the two buffers overlap. Can I arbitrary decide that I will exclude the points which buffers overlap more than 10 % (of the total surface area of the 400 m buffer) and include ones that overlap less than this amount? There will still be a problem of psuedoreplication but can I ignore it? Should I take less, 5 % per example?
The buffers look like this: http://oi67.tinypic.com/1678m6g.jpg
Best regards,
Ivan
Hello Daniel, nightjars can be heard as far as ~500 m away from the census point. And their singing territory is around 10 ha. I think I loose some important habitat types if I take only 300 m of buffer. Either way I could try making the analysis with 300 m and 400 m buffers. Thanks for the answer.
Hi Ivan,
I wasn't able to see your points and buffers, and am wondering if the point locations are set for some reason? Couldn't you use a spatially stratified random point generator to generate your sample locations, and build in a rule that the points must be 800 m away from each other, therefore you could use them all?
Alternatively, if you do remove points where buffers overlap by more than 10%, you could build in a routine to randomly select only one of the two points to remove. It sounded like you were proposing to exclude both points, which doesn't seem necessary.
Lara
Hi Lara,
the census points were already set for a monitoring before I started my research. Otherwise I could place them differently in order to avoid this issue.
I was thinking of keeping the point with presence in the case one is with presence and one without. In the case both are presence/absence I would randomly exclude one. Thank you for the answer.
It depends on the number of samples that you have. If you can have enough occurrences (e.g. > 30) while avoiding overlapping buffers, it's probably better to suppress any overlap. But if you don't have enough sample points, then you have no choice...
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology