Comparison of global gridded precipitation products over a mountainous region of Africa

Authors

  • T. Dinku,

    Corresponding author
    1. International Research Institute for Climate and Society, The Earth Institute at Columbia University, 61 Route 9W, Palisades, NY 10964-8000, USA
    • International Research Institute for Climate and Society, The Earth Institute at Columbia University, 61 Route 9W, Palisades, NY 10964-8000, USA.
    Search for more papers by this author
  • S. J. Connor,

    1. International Research Institute for Climate and Society, The Earth Institute at Columbia University, 61 Route 9W, Palisades, NY 10964-8000, USA
    Search for more papers by this author
  • P. Ceccato,

    1. International Research Institute for Climate and Society, The Earth Institute at Columbia University, 61 Route 9W, Palisades, NY 10964-8000, USA
    Search for more papers by this author
  • C. F. Ropelewski

    1. International Research Institute for Climate and Society, The Earth Institute at Columbia University, 61 Route 9W, Palisades, NY 10964-8000, USA
    Search for more papers by this author

Abstract

Five gridded monthly precipitation products are evaluated using a gauge network over complex topography in Africa. The global gridded products considered are produced by the Global Precipitation Climatology Center (GPCC), NOAA Climate Prediction Center (NOAA–CPC), and the Climate Research Unit at the University of East Anglia (UEA–CRU). Three different products from GPCC are available at multiple spatial resolutions: 0.5, 1 and 2.5° ; the NOAA–CPC product has a spatial resolution of 2.5° while that of UEA–CRU is 0.5° . Comparisons of the GPCC and UEA–CRU products are carried out at spatial resolutions of 0.5, 1 and 2.5° , while NOAA–CPC is compared with the other products only at 2.5° resolution. There is very strong agreement between the gridded global products and the reference raingauge data. Average correlation coefficients are about 0.95, 0.92, and 0.90 at 2.5, 1.0 and 0.5° spatial resolutions, respectively. Both systematic and random errors are reasonably low. The performance of these products is highest during the wettest season (Jun–Aug), and relatively poor during the dry season (Dec–Feb). The seasonal differences are more prominent at high resolution. These results are very encouraging, particularly, when considering the complex terrain of the validation site. Copyright © 2008 Royal Meteorological Society

Ancillary