this is the first question I have asked on Stack Overflow. However, I am a student and have been using this website for several years without needing to ask a question. There is a wealth of information on here and I appreciate the people who take the time to answer questions. If I need to make any changes to the question or format of the question I will be more than happy to.
I am researching habitat use by a wildlife species. I conducted field studies on GPS collared animals and took vegetative measurements in the field and landscape measurements in GIS.
Currently, I need to classify each plot (unique.id) into a “forest type” (i.e., Douglas-fir low- elevation forest, ponderosa pine woodland, etc) based on the attributes of the plot. The "forest type" is arbitrary and created by me. I am not using to R classify for me, just to provide visual aids and summary statistics on each plot.
To aid in this, I would like to display a histogram of the tree diameter distributions by tree species for each plot. In the same image window, I would like to display a few other variables from that plot such as canopy cover (canopy), stand age (age), species of the tree that age was taken from (agespecies), elevation (Elev), aspect (Aspect), and stem density (density). Due to the large number of plots, I would be nice to print them all to a pdf or other format for review outside of R.
I am not looking for R to classify the plots for me, just to provide some summary and visual information for each plot to assist my in classifying it.
So far, I have been using the “histogram” function in the “lattice” library, but am open to using something different. I have been able to write code to build a diameter histogram and loop it for each plot. I have also been able to add a subset if I am just running one plot at a time, but I don’t know how to loop the subset. I also am unsure of how to add multiple subsets (canopy, age, agespecies, Elev, Aspect, density) to the histogram.
Finally, most plots do not contain every possible tree species. Is there is a way to order the histograms by which species has the highest number of counts and/or not show that histograms that are empty?
I have pasted my code so far and the structure of the data below. The data are in two separate files, “dbh” and “masterplot”
Data:
> str(dbh)
'data.frame': 80719 obs. of 7 variables:
$ unique.id: Factor w/ 1165 levels "CalvA1","CalvA10",..: 1 1 1 1 1 1 1 1 1 1 ...
$ species : Factor w/ 14 levels "abla","abpr",..: 1 2 3 3 4 4 5 5 5 7 ...
$ dbh : num 7.8 1.1 3.3 3.8 4.1 3.4 6.1 4.2 3.2 3.8 ...
str(masterplot)
'data.frame': 1170 obs. of 41 variables:
$ unique.id : Factor w/ 1165 levels "CalvA1","CalvA10",..: 1 2 3 4 5 6 7 8 9 10 ...
$ canopy : num 16 19 28 25 1 3 23 14 7 18 ...
$ age : num 147 72 167 64 153 144 192 154 173 44 ...
$ agespecies : Factor w/ 14 levels "abla","alru",..: 6 11 7 11 7 6 11 6 11 6 ...
$ Elev : num 1597 1850 1638 1540 1695 ...
$ Aspect : num 238.6 246.1 165.5 242.1 24.4 ...
$ density : num 8700 6600 6800 7800 14600 5600 13900 4600 3900 4000 ...
Code:
lathist.fx=function(x){
windows()
histogram(~dbh[unique.id==x] | species, breaks=c(0,4,11,50),data=dbh,)}
for (i in dbh$unique.id)
lathist.fx(i)
I think the subsets will look something like this…
sub=masterplot$age[masterplot$unique.id=="LeftA36"]
