Open AccessModelling the impacts of climate variability on crop yields in Nigeria: performance evaluation of RegCM3‐GLAM system
Author(s) -
Matthew Olaniran J.,
Abiodun Babatunde J.,
Salami Ayobami T.
Publication year - 2015
Publication title -
meteorological applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.672
H-Index - 59
eISSN - 1469-8080
pISSN - 1350-4827
DOI - 10.1002/met.1443
Subject(s) - environmental science , transpiration , climate change , climate model , climatology , correlation coefficient , extinction (optical mineralogy) , sensitivity (control systems) , crop , dry season , atmospheric sciences , statistics , mathematics , geography , ecology , geology , forestry , paleontology , botany , photosynthesis , cartography , electronic engineering , engineering , biology
This study evaluates the capability of a Climate‐Crop Modelling System ( RegCM3‐GLAM ) in simulating the regional climate and crop yields (maize, rice, cowpea, and groundnut) over Nigeria. Daily climatic data obtained from a Regional Climate Model ( RegCM3 ) simulation was used as the input data in the General Large Area Model ( GLAM ) to simulate the crop yields for 11 years (1999–2009), and a series of sensitivity experiments were performed to test and optimize the GLAM parameters over the region. The results show that RegCM3 gives a realistic simulation of the Nigerian climate. The correlation coefficients obtained between the observed and simulated climatic variables are between 0.72 and 0.96 at p < 0.01. However, the model slightly underestimates rainfall and maximum temperature in the wet season (April to October) and overestimates rainfall and maximum temperature in the dry season (November to March). GLAM also gives a realistic simulation of the mean and spatial distribution of crop yields in Nigeria. The root mean square errors of the simulations are generally less than 36% of the observed yields. The performance evaluation of the model varies with ecological zones. The model shows the best performance in simulating maize and the worst performance in simulating cowpea over the Savannas. Sensitivity experiments reveal that simulated crop yield is sensitive to model parameters (harvest index, extinction coefficient, optimum temperature, and transpiration efficiency), with both extinction coefficient and transpiration efficiency showing more significant impact. It has been concluded that the performance of GLAM over the country can be further improved by enhancing the quality of meteorological input data.