Atmospheric transmissivity: distribution and empirical estimation around the central Andes

Authors

  • Guillermo A. Baigorria,

    Corresponding author
    1. Department of Production Systems and Natural Resources Management, International Potato Center, PO Box 1558, Lima 12, Peru
    2. Laboratory of Soil Science and Geology, Wageningen University, PO Box 37, 6700 AA Wageningen, The Netherlands
    • Department of Production Systems and Natural Resources Management, International Potato Center, PO Box 1558, Lima 12, Peru
    Search for more papers by this author
  • Esequiel B. Villegas,

    1. Dirección General de Investigación y Asuntos Ambientales, Servicio Nacional de Meteorología e Hidrología de Perú, PO Box 1308, Lima 11, Peru
    Search for more papers by this author
  • Irene Trebejo,

    1. Dirección General de Investigación y Asuntos Ambientales, Servicio Nacional de Meteorología e Hidrología de Perú, PO Box 1308, Lima 11, Peru
    Search for more papers by this author
  • Jose F. Carlos,

    1. Dirección General de Investigación y Asuntos Ambientales, Servicio Nacional de Meteorología e Hidrología de Perú, PO Box 1308, Lima 11, Peru
    Search for more papers by this author
  • Roberto Quiroz

    1. Department of Production Systems and Natural Resources Management, International Potato Center, PO Box 1558, Lima 12, Peru
    Search for more papers by this author

Abstract

This study of the distribution in space and time of atmospheric transmissivity τ takes into account the fact that, in complex terrain, many factors affect this variable; thus, it is not possible to use the generalizations that can be applied under more homogeneous conditions. Climatic controls, topography and even sea currents have important effects on clouds and aerosols affecting τ, simultaneously leading to differences in the distribution of incident solar radiation. Different models exist to estimate incoming solar radiation as a function of relative sunshine hours (observed sunshine hours/theoretical sunshine hours, n/N) or differences between maximum and minimum temperatures Δ T. We calibrated, validated and evaluated four of these empirical relations based on data from 15 weather stations in Peru. Models were calibrated using 66% of the daily historical record available for each weather station; the rest of the information was used for validation and comparison. The Ångström–Prescott model was used to estimate incoming solar radiation based on n/N, and gave the best performance of all the models tested. The other models (Bristow–Campbell, Hargreaves, and Garcia) estimated incoming solar radiation based on Δ T. Of all the models in this group, the Bristow–Campbell model performed best; it is also valuable because of the physical explanation involved. The empirical coefficients of all the models evaluated are presented here. Two empirical equations are proposed with which to estimate values of the coefficients bB and cB in the Bristow–Campbell model, as a function of Δ T and latitude, allowing the model to be applied to other study areas. Copyright © 2004 Royal Meteorological Society

Ancillary