HeatherSavoy-USDA
commented
1 year ago This issue was discussed at the Mar 2 GeoCDL meeting. The consensus was to create an additional input parameter dedicated to specifying depth/height that is flexible enough to handle the different ways datasets store data in this third physical dimension, e.g.
- a few predefined intervals of potentially unequal length but consistent across horizontal space (e.g. SoilGrids #74)
- interval definitions that are not consistent across horizontal space (e.g. gNATSGO)
- evenly discretized similar to x-y coordinates
When we get to more soils or atmospheric datasets, data variables can be a function of depth/height. We might consider an input parameter and some dataset properties/methods for subsetting that dimension.