标题：Assessing the skills of inter-sectoral impact model intercomparison project climate models for precipitation simulation in the Gongola Basin of Nigeria
摘要：AbstractThe study quantitatively assesses the ability of five Global Circulation Models (GCMs) in the fast track of Inter-Sectoral Impact Model Intercomparison Project (ISI-MIP) to reproduce the observed precipitation climatology over the Gongola river basin of Nigeria for the period 1982−2004. The recent occurrences of recurrent flooding episodes in the basin is alarming. Hence, the models' present-day precipitation is evaluated relative to Global Precipitation Climatology Centre (GPCC) observational datasets based on spatial analysis, statistical measures and climate indices at annual, monthly and daily cycles to identify the most appropriate GCM for impact model in the basin. The results show that climate models replicate the annual precipitation pattern well, both spatially and in magnitude with varying margins. Moreover, the GCMs captured the orographic pattern in the Jos plateau and the general decreasing precipitation trend towards the basin's northeast. amongst the GCMs, IPSL-CM5A-LR better captured the rainy season in the basin extents from April to October and May to October respectively over the Jos plateau and other regions, with maximum rainfall occurring in August, exhibiting a unimodal pattern. The HADGEM2-ES however, better represented the most occurring rainfall intensity in the basin (5 to 50 mm hr−1) in most regions. The degree of pattern correspondence was found highest for IPSL-CM5A-LR with a correlation coefficient of 0.73. Only HADGEM2-ES was able to capture the spatial variability of maximum consecutive dry days over the study domain, increasing from 150 days around the Jos plateau to 200 days over Uba plain. In any case, the HADGEM2-ES appeared to be the most promising model for simulating the extreme conditions over the Gongola basin and can therefore be selected for the application of hydrological impact model for adaption strategy.
