My data collection has been going well. I’m trying to see if moisture and sunlight create zonation in forest ground cover.I did an initial batch of 6 replicates of 20 samples, with 9 squares that I recorded what kind of plants were in. I picked six random points and used random numbers to go 1-20 paces north or south, and 1-20 paces east or west. The 9 squares turned out to not be very useful for what I was doing since I’m not measuring patchiness I’d be better off taking more less detailed quadrats.
I wound up starting over since I changed how I was measuring a variable and switched to doing transects. I’m using tree type as a proxy for moisture, since where I live summer tends to be where it’s limited and dry summers have killed some trees in my study area.
I’ve originally planned to use tree type as a proxy for light, with coniferous, deciduous and no tree cover above being my three treatment groups. I got some feedback and got myself (actually made it out of a broken telescope and a few other things) a thing that’s basically a vertically pointing periscope that only lets you what’s directly overhead. I can now estimate canopy cover directly.
Switching from doing random sampling to transects has made me aware that ground cover types are “patchier” than I’d thought. I also noticed plants growing in less than ideal spots near healthier plants, such as sickly Oregon grape competing with ivy. There seems to be pretty clear zonation with tree types as I get closer to the river and ground cover increases, so I suspect that using trees as a proxy for moisture is somewhat valid.
Tree type can by used a proxy for moisture though it depends on how fine you want to measure moisture as most trees have a range of moisture they can tolerate. I have had students do soil measurements in various ways.