Accuracy of Giovanni and Marksim rainfall data for use in the agricultural adaptation to climate change
Date
2009
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Agricultural adaptation to climate change requires accurate, unbiased, and reliable climate data. Availability of observed climatic data is limited because of inadequate weather stations. Rainfall simulation models are important tools for simulation of rainfall in areas with limited or no observed data. Various weather generators have been developed that can produce time series of synthetic climate data. Verification of the applicability of the synthetic data is essential in order to determine their accuracy and reliability for use in areas different from those that were used during models development. The Marksim and Giovanni weather generators were compared against 10 years of observed data (1998-2007) for their performance in simulating rainfall in four stations within the northern bimodal areas of Tanzania. The observed and synthetic data were analyzed using climatic dialog of the INSTAT program. Results indicated that during the long rain season (masika) Giovanni predicted well the rainfall amounts, rainy days, and maximum dry spells compared to Marksim model. The Marksim model estimated seasonal lengths much better than the Giovanni model during masika. During short rain season (vuli), Giovanni was much better than Marksim. All the two models had better predictions during masika compared to vuli. The Giovanni model estimated probabilities of occurrence of rainfall much better (RMSE = 0.23, MAE = 0.18, and d =0.75) than Marksim (RMSE = 0.28, MAE = 0.23, R2 = 0.30 and d =0.63). The Marksim model over-predicted the probabilities of occurrence of dry spells greater than seven days (MBE = 0.17) compared to the Giovanni model (MBE = 0.01). In general the Giovanni model was more accurate than the Marksim model in most of the observed weather variables. The web based Giovanni model is best suited to the bimodal areas provided the web and remote sensing data representing the area are available. The Marksim model gives more accurate synthetic data if climate normals are used as input variables. Even without the climate normals the Marksim model can still be used to generate synthetic data in bimodal climatic areas where there are no observed data.
Description
item.page.type
IDRC-Related Report
item.page.format
Keywords
BIMODAL RAINFALL, CLIMATE NORMALS, DRY SPELLS, GIOVANNI, INSTAT, MARKSIM, CLIMATE DATA, AGROMETEOROLOGY