One of my spatial problems is examining the spatial distribution of mitigated wetlands in the Willamette Valley to examine the quality of location chosen for restoration. The data set I used to test the hot spot tool is a point file of wetland mitigation sites (i.e. sites that have been restored or created based on intentional disturbance elsewhere).
The mitigation data look clustered when examined visually, and average nearest neighbor confirms this hypothesis.
It seems intuitive that wetlands would be clustered towards streams so I ran average nearest neighbor on the valley’s streams to examine spatial distribution. This showed that the streams are less clustered than mitigated wetlands, indicating there other factors that explain locations of mitigated wetland sites.
Categorical data is largely unusable in the spatial statistics toolbox. However, I wanted to examine the spatial distribution of mitigated wetlands compared to historic vegetation cover. In order to work around the categorical data, I first created a layer that only contained historic wetland vegetation; I then ran the “near” tool to calculate distance between the mitigated wetlands and the historic wetland polygons. Lastly, I ran the hot spot analysis on this distance.
Red indicates increased distance from a historic wetland. The results show that since most of the valley was once floodplain wetlands, most sites are situated on historic wetlands; an area near Portland, however, shows a hot spot of mitigated wetlands that are located further from historic wetland vegetation.
Next, I used the “near” tool again to calculate distance between mitigated wetlands and the closest stream within my study area (a subset of the valley). I then ran a hot spot analysis. Results show significant clusters of mitigated sites that are closer to water sources than others (blue indicates closer). Edge effect could come into play in this scenario since both layers were clipped to my specific study area.
Lastly, I wanted to look at the mitigated sites in relation to a LiDAR flood inundation data set of the Willamette main stem using water depth. There were only a handful of sites situated within the two year flood plain, so the hot spot analysis turned out as expected: no significant hot or cold spots
I think it was discussed in class, but if you were interested in the spatial relationship with specific veg layers, you could recode them as presence/absence data (1/0) to get around the issue of categorical data. Your method for grouping the historic data together seems appropriate, but you do lose some of the resolution.
Presence/absence could surely be another way around this problem. I would be interested to see what a hot-spot analysis of a binary field would look like; I’ll have to try it out.
Excellent analysis!
For the Average Nearest Neighbor analysis it would be interesting to compare the clustering for mitigated points, to the historical points. Be very sure that the study area is identical for both of these analysis (you can create a polygon study area and use the AREA of that polygon in each analysis).
I really like the hot spot analysis using near streams to see the spatial pattern of mitigation. Couple comments:
1) Edge effects are less severe with hot spot analysis than some statistics. Because the data has been clipped, the tool will have less information to use in computing a result, but there aren’t any undercount type of biases.
2) Hot Spot Analysis was not intended to use with binary data, unfortunately. You can, however, overlay your study area with a fishnet grid and count the incidents within each polygon grid cell… then run Hot Spot Analysis on the counts.
Hope this helps!
Very best wishes,
Lauren