Area Solar Radiation for Large DEM

1344
8
08-17-2022 08:58 AM
Jaslam
by
New Contributor II

I need a solar radiation map (Whole Year) for the entire United States, which I was thinking of creating with the solar radiation tool in ArcGIS Pro and a national-level DEM layer. Unfortunately, after 10 days of processing, it is only at 3%. Is there a simple way to obtain a Map of the Area Solar Radiation?

0 Kudos
8 Replies
Robert_LeClair
Esri Notable Contributor

Whoa - 10 days processing and 3% complete.  My suspicion is that the spatial resolution of the national level DEM is of a high spatial resolution like 10m or similar.  If you can get access to a DEM with low spatial resolution like 90m, then the processing time would be reduced. 

In the help/usage for the Area Solar Radiation tool it does state "Calculating insolation can be very time consuming, where the calculations for a large digital elevation model (DEM) can take several hours, and for a very large DEM, even days."  You're definitely seeing the days component.

0 Kudos
Jaslam
by
New Contributor II

Exactly, Its 30m resolution.

0 Kudos
Robert_LeClair
Esri Notable Contributor

For a nation wide extent, 30m is likely the culprit of the tool running slowly.  If you can find 90m spatial resolution DEMs, then I'd suggest running the GP tool against that.  What is the result?

You can download lower spatial resolution data here for the globe.  Would be worth a download and testing.

DanPatterson
MVP Esteemed Contributor

Area Solar Radiation (Spatial Analyst)—ArcGIS Pro | Documentation

30m is too fine.  10km perhaps.  Assuming you have projected data as well (?)

You need to split up the US into chunks since you can't just use one latitude as representative of the continental parts, let alone the remaining states.

The help topic has some pertinent recommendations for the whole process.

Also

How solar radiation is calculated—ArcGIS Pro | Documentation


... sort of retired...
Jaslam
by
New Contributor II

Thanks @DanPatterson 

Since the projection can affect accuracy, it is not advised for analysis, so I didn't do it.

yeah! You are perfectly true because I cannot use one latitude as a representative.

I ran the process for the state of California to assess the processing time, and it took 24 hours to process 2% 

Is there a way to make the procedure go more quickly? Can an approach, such as model construction, be helpful? I saw a post describing it, but because I'm new to this area, I found it challenging to understand.

Solved: Re: Solar Radiation Tools - - Esri Community

 

Any Thoughts!!! 

0 Kudos
DanPatterson
MVP Esteemed Contributor

Project your raster.

see How the Cell Size Projection Method environment setting works—ArcGIS Pro | Documentation

as indicated, use a larger cell size, the projection is going to be the least of the problems when it comes to accuracy, since you don't have anything to compare calculations to except for perhaps spot measures collected at few locations.

A model isn't going to speed things up, it is the number of calculations required.


... sort of retired...
0 Kudos
Robert_LeClair
Esri Notable Contributor

Another approach to consider is the results from NREL where they have Solar Resource Maps and Data for Global Horizontal Irradiance and Direct Normal Irradiance for the US.  These maps show Annual Averages and Monthly Averages.  Would this be of any help?

Jaslam
by
New Contributor II

Hello @Robert_LeClair 

I think it may be too coarse, but worth exploring. 

An Another useful link which I found is - https://globalsolaratlas.info/download/usa

Thanks Much

Jaslam