Skip to main content
ARS Home » Research » Publications at this Location » Publication #109876
Title: MODELS OF SOLAR IRRADIANCE BASED ON DAILY HUMIDITY, PRECIPITATION, AND TEMPERATURE DATA FOR VARIOUS SPATIAL SCALES AND GLOBAL-CHANGE APPLICATIONS

Author
item WINSLOW, JEROME - UNIVERSITY OF WYOMING
item Hunt Jr, Earle
item PIPER, STEPHEN - SCRIPPS INSTITUTION

Submitted to: Ecological Applications
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/17/2001
Publication Date: N/A
Citation: N/A

Interpretive Summary: Incoming solar radiation affects plant photosynthesis and transpiration; however incoming solar radiation is often not measured. Alternatives to direct measurements of solar radiation are to use estimates of radiation from satellites or to use global climate models; both are not appropriate for many applications. K. L. Bristow and G. S. Campbell devised an empirical algorithm to predict incoming solar radiation from daily precipitation, maximum temperature, and minimum temperature. The predictions of incoming solar radiation for maritime and tropical regions are much worse than predictions for regions with continental climates. Our goal was to develop a model which can be used around the world. We found one fraction of radiation transmitted through the atmosphere is approximately equal to one minus the average daytime relative humidity. From these data we developed two models, RH-RAD which uses average relative humidity and VP-RAD which estimates atmospheric vapor pressure from nighttime minimum temperature. We tested the predictions globally and found that for 72% of the vegetated earth's surface, predicted at observations of 1 megajoule per meter-squared per day. Whereas, the two models still need some improvements, they are better than other models based on the Bristow and Campbell algorithm.

Technical Abstract: The total daily solar irradiance (Rs) received at the earth's surface is a critical component of ecosystem carbon, water, and energy processes. K. L. Bristow and G. S. Campbell devised an empirical algorithm to predict irradiance from daily precipitation, maximum temperature, and minimum temperature. However, their equation requires site-specific coefficients tuned with available data. When a single set of coefficients is used globally, the Bristow-Campbell equation's performance is degraded in tropical and moist maritime areas. We found for seven sites located around the world, daily transmittance is approximately equal to one minus the daily-average relative humidity. From these data we developed two models, RH-RAD based on relative humidity and VP-RAD based on nighttime minimum temperature. These models predict daily transmittance for estimation of daily solar irradiance or photosynthetically active radiation (PAR). Both models had good agreement with the site data used for model development. We compared VP-RAD predictions globally with PAR derived from satellite measurements. For 72% of 13072 1-degree grid cells of the vegetated land surface, VP-RAD and satellite-derived PAR converged within 1.0 megajoule per meter-squared per day. Whereas, the comparisons with satellite-derived PAR show that VP-RAD still needs improvement in tropical regions, VP-RAD is a substantial improvement compared to related models that predict atmospheric transmittance from meteorological data.