Author: Sam Zipper (email@example.com)
This repository is home to code for the High Resolution Mapping of EvapoTranspiration (HRMET; pronounced "hermit") model.
HRMET is introduced and described in the following publication:
Zipper, S.C. & S.P. Loheide II (2014). Using evapotranspiration to assess drought sensitivity on a subfield scale with HRMET, a high resolution surface energy balance model. Agricultural & Forest Meteorology 197: 91-102. DOI: 10.1016/j.agrformet.2014.06.009
Please cite this paper if you use HRMET or any derivative thereof.
HRMET.R = this is the HRMET code
Publications.txt = a text file tracking publications that use HRMET; if you publish something, let me know by submitting a push request! (or send an email)
HRMET_HowTo/* = Example HRMET files showing how to run HRMET over a grid with spatially variable inputs and estimate uncertainty in results; however, these are with a no-longer-maintained MATLAB version of the code (the version used in the original HRMET paper) and are not complete. Regardless, they may help you understand how to use the current, R version of the code at your own study site.
Key Assumptions of HRMET
-HRMET calculated the 1D surface energy balance. However, it is typically applied over fields to produce raster maps of ET. In order to do this, you simply have to define the relevant inputs at all locations you want to map ET, and then run HRMET at each grid point.
-Thus, assumptions of spatial homogeneity of inputs should be made with care. For example, in Zipper et al. (2014), we assume uniform meteorological conditions over our relatively small (~600 x 600 m) field. This assumption gets increasingly problematic as your spatial scale increases. HRMET was designed for precision-agriculture scale applications; however, the physical principles should work at larger scales, so long as the the input data is sufficiently high-resolution.
-HRMET does not work well in extremely short canopies or deserts (h approaching 0 m).
-HRMET does not work well when the canopy height (h) exceeds the height of temperature and wind speed measurements (Zair and Zu).
-The G_Tw coefficient (used in cloudiness estimation) takes a summer value by default; future versions should automatically select this based on DOY.
Bug Fix History
2017-06-07: Fix small bug in cloudiness fraction calculation (it was being inadvertently set to 0 under most conditions)