ConservationInternational / ShinyCam

Basic Analytics for Camera Trap Data
6 stars 1 forks source link

Species alert tab #24

Open arodriguezca opened 7 years ago

arodriguezca commented 7 years ago

Problem: There is a need to be aware when species are observed in unusual areas. In ecology, this is usually addressed with polygons that determine the habitat boundaries. These boundaries data are available as shapefiles here. It would be nice to compare the actual observations against the shapefile boundaries, and when an animal goes beyond its normal limits, statistically address the possible cause (perhaps we have new information about habitat range or perhaps we made a mistake in animal identification). Moreover, we may be able to analyze movement patterns of the species across seasons and years.

Desired Solution: A new tab fully dedicated for this. It will show the map with the boundaries from the shapefiles along with the observations per specie. In addition, if there are animals out of those boundaries, it will highlight those observations and give a good guess about the possible cause of it (according to the statistical evidence). Also it will have the option to show the movement patterns for that specie, across the selected time frame (season or year). Just to be clear, we have to do this as close as the functionality/interaction experience as "Interactive Map" tab.

arodriguezca commented 7 years ago

Today @nicoleeel,@kramea and me worked on plotting the species boundaries (polygons) in the map. So far it was implemented, but looks messy because they all have the same colors and there is no legend. There is work to be done in presenting this nicely.

Besides that, we used the shapefiles for terrestrial mammals, and they are pretty big (and slow to load). I suggest to have a preprocessing R script to select the ones that match with the species we have in our data. Then, it will be easier to run locally the Shiny App. @pabloQuispe is working on that, but he is not done yet.

Future work: I will to finish the colors and legend. If @pabloQuispe is successful, I'll add it in a commit as well.