We are working in a solar energy project to locate potential sites for solar panel farms. We are using the Landsat 7 panchromatic band to derive radiance with GRASS using i.landsat.toar (W/m2*sr*µm units). Now, we have access to some ground data from sensors measuring global horizontal irradiance (GHI) which output is in W/m2 units. I would like to compare these datasets for validation.

Is there a way to convert radiance in W/m2*sr*µm to irradiance in W/m2?

I have read in [1] that radiance = irradiance / π, but I don't completely understand, also, this doesn't take the spectral part (µm) into account.

[1] http://earth.esa.int/landtraining07/D1LB3-Su.pdf

closed as too broad by PolyGeo Aug 24 '16 at 18:55

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    You might be best off consulting the operating characteristics of the sensors to find out the spectrum they detect and the weights they apply to the wavelengths within that spectrum so you can integrate over the wavelengths. You also would need information about the orientations of the sensors so that you can integrate over the spherical angles. – whuber Jun 17 '15 at 16:58
  • 2
    I think that you may have misread your reference. [1] does not state that radiance = irradiance / pi, but it does state that radiance = radiant emittance / pi. Going from radiance (from GRASS) to GHI requires implementing a radiative transfer model - such as the one delineated on page 37 of [1]. – Mikkel Lydholm Rasmussen Jun 17 '15 at 17:17
  • 2
    With Landsat bands you are measuring reflected light, which of course is not the same as the amount of radiation illuminating your targets (solar panels). Your total incoming solar radiation budget is going to be a function of the atmosphere (i.e., clouds, haze, etc.), not what is reflected back from the ground. How big of a search area are you considering? If it's local, then aside from obscuring topography and/or trees, then the atmospheric differences between sites may be negligible. If it's regional or global, then some thought needs to go into cloudy days and atmospheric conditions. – aaryno Jun 17 '15 at 17:28
  • 2
    I just re-read your comment about GHI... To answer your question more specifically, @whuber is correct -- you need to get the operating characteristics of your ground sensors. This would be in addition to measuring the surface reflectance (as a function of wavelength) which, hopefully, is constant through time. The π factor comes from trying to account for the hemispherical scatter of light after it hits your target. – aaryno Jun 17 '15 at 17:35
  • 2
    If you still haven't, check out JRC's Photovoltaic Geographical Information System - Interactive Maps and their available for download Satellite based solar radiation data. Recently I derived global irradiation on inclined surfaces time series maps using various inputs and, of course, r.sun.daily. – Nikos Alexandris Sep 18 '15 at 19:21