What data are analyzed/processed?
All data are processed/cleaned, but I still need to compare estimated water levels to wetland water level data.
What statistical analyses still need to be done?
I still need to create predictive models, and from those, calculate foraging area for a given water level.
What else needs to be done?
I still need to add vegetation data, if I have time.
This project is exploring peripheral Great Salt Lake wetland water levels and available American white pelican foraging habitat. Essentially I'll be creating a water level model using Great Salt Lake bathymetry data and estimating water levels around the lake. From those water level estimates, I'm going to try to calculate how much wetland area is submerged at a given water level and is therefore available for pelicans to feed in.
Data used:
USGS water level data from the Saline, UT water level gauge (1966 to present) [1]
Piezometer data from a handful of Great Salt Lake wetlands (2012 - 2015) [2]
A piezometer is a device which measures the pressure (i.e. depth of water above the device) of groundwater at a specific point.
Great Salt Lake bathymetry raster [3]
USFWS Utah wetland delineation shapefile [4]
First, I cleaned my water level data using R, by collapsing daily water level measurements to weekly means to make data easier to work with (for this project, I just need water levels of some sort to model with, not high resolution data water levels). I had to covert the USGS water level data from feet to meters, but the piezometer data were already in metric units. I also calculated the minimum and maximum water level measurements from the USGS water level dataset for use in ArcGIS Pro:
Min: 1276.823
Max: 1283.574
Neat visualization of weekly mean water levels over years at the water level gauge I used (some years are missing weeks near the end).
A horrendous mess of lines, but this is the wetland water level data I have available from 2012 to 2015, collapsed to weekly means.
Then, in ArcGIS Pro, I imported the Great Salt Lake bathymetry raster and created a hillshade layer from it using the Hillshade tool. Along with the bathymetry layer, I also imported the USFWS wetland boundaries shapefile. With the bathymetry layer, I reclassed the symbology using the Reclassify tool to the min and max GSL water level weekly means for easy visualization.
A screenshot of reclassed Great Salt Lake surface elevations based on the bathymetry raster I downloaded.
Grey is either below the water's surface or above any elevation the water level has reached within the span of the dataset.
This isn't actually from the area I'm assessing, but I thought it was nice looking.
I haven't run into many issues yet, although the bathymetry model isn't as high resolution as I'd like. One issue that may be a bother later is that based on the pelican GPS telemetry data I have from other projects, pelicans don't strictly use wetlands as delineated by the wetland shapefile. I may need to spend some time manually delineating some of their foraging areas based on water levels if it seems like they're just milling around in multiple wetland areas.
I also will need to find a wetland vegetation layer if I have time to explore that (although with how involved this project is, I might not be able to get to it!).