I am planning to work with the Solar Radiation modeling tools over a large area from Virginia to North Carolina. One of the required inputs to the Area and Point tools is "Latitude". The tool sets this value automatically based on the input raster. However, it seems that if the study area is as large as mine then using one value is not appropriate.
Will I need to break the study region up into sections and mosaic them back together? If so, it seems that the sections would need to overlap so that the edges are modeled accurately.
Any advice on how to handle solar radiation modeling across a large study area would be very helpful.
I am planning to work with the Solar Radiation modeling tools over a large area from Virginia to North Carolina. One of the required inputs to the Area and Point tools is "Latitude". The tool sets this value automatically based on the input raster. However, it seems that if the study area is as large as mine then using one value is not appropriate.
Will I need to break the study region up into sections and mosaic them back together? If so, it seems that the sections would need to overlap so that the edges are modeled accurately.
Any advice on how to handle solar radiation modeling across a large study area would be very helpful.
The analysis is designed only for local landscape scales, so it is generally acceptable to use one latitude value for the whole DEM. With larger datasets, such as for states, countries, or continents, the insolation results will differ significantly at different latitudes (greater than 1 degree). To analyze broader geographic regions, it is necessary to divide the study area into zones with different latitudes.
In your case the Latitude varies several degrees (from 39.466133 to 33.842074). If you look at a tool like the Clear Day Solar Radiation calculator (www.ksre.ksu.edu/irrigate/Software/CDRCalc.xls�??) you can derive that the effect of the Latitude on clear day solar radiation will be about 2% in the summer, but rises to 28% in winter time! Taking a central Latitude will still give results that are 12-13% off at the Northern and Southern edges (in the winter).
In this case splitting the area up in horizontal slices (<1 degree Latitude) and scripting the analysis (Python) in which you provide for each "slice" the central Latitude of that slice would probable provide the best result.
It seems that splitting the area up will result in edge effects. It may make sense for the edges of each split to overlap by some amount. What do you think?
Sounds to me that that's the way to go. You should have a look to what distance of the edge the distortion is present and let that define the amount of overlap.
