Hydrology has reported that the elevation dataset we are using, which contains the mean elevation for each grid square, is too coarse, resulting in bad data for the Melton Ratio.
The issue is that the Melton Ratio is the difference in elevation betweeen the highest and lowest grid cells of the watershed, divided by the square root of the area of the watershed. Our coarse mean-per-grid-cell elevation data is smoothing away peaks and valleys, resulting in too-small values for the elevation difference.
Options to fix this include:
Using a less coarse elevation dataset. I expect this would be a lot of trouble, since the watershed API operates by correlating three datasets with identical grids.
Creating minimum and maximum elevation datasets to use in parallel to our mean-elevation datasets. I expect this to be a much easier route, since we can slot them easily into the grid-navigating code we already have.
Hydrology has reported that the elevation dataset we are using, which contains the mean elevation for each grid square, is too coarse, resulting in bad data for the Melton Ratio.
The issue is that the Melton Ratio is the difference in elevation betweeen the highest and lowest grid cells of the watershed, divided by the square root of the area of the watershed. Our coarse mean-per-grid-cell elevation data is smoothing away peaks and valleys, resulting in too-small values for the elevation difference.
Options to fix this include:
watershedAPI operates by correlating three datasets with identical grids.