The impact of climate variability and land cover change on land surface conditions in North-eastern Nigeria

Recent droughts in the West African Sahel have been the most catastrophic since modern records began posing a threat to the economy and security of the region. Two contending views have evolved in the scientific community to explain the causes of re-occurring droughts in the West African Sahel Region. These themes are “the regional land-atmosphere feedback” mechanism and “ocean-atmosphere interaction”. This study is specific to a part of West Africa the North-eastern part of Nigeria and attempts to examine the impact of climate variability and land cover change on land surface conditions of fluxes in energy and momentum in the past (1980-2000) and in the future (2046-2065) based on the IPCC A2 emission scenario. The level of recovery of the region from previous droughts in the 1980s was evaluated using the Standardised Precipitation Index (SPI) and linear regression to identify trends in monthly rainfall and number of rainy days in a month using monthly rainfall time series data between 1980 and 2006. The study further applies a range of stochastic linear models (ARIMA) to predict monthly rainfall time series over a 24 month period, a Cellular Automata –Markov model to project land cover for the year 2046, and a more dynamic land surface scheme the Joint UK Land Environment Simulator (JULES) for simulating past (1980-2000) and future (2046-2065) land surface conditions of soil moisture, soil temperature, surface runoff, Gross Primary Productivity (GPP), latent and sensible heat fluxes. GIS techniques are used to assemble data on soil texture and fractional land cover types used as boundary conditions required by JULES in some of the simulations. As part of the model evaluation process the JULES land surface model output of surface soil moisture is evaluated with an European Remote Sensing (ERS) satellite product. The sensitivity of the model to input data is examined through changes in scale and non-linearity in the calculation of soil hydraulic parameters. Results suggest that despite a recovery in rainfall in the 1990s from the previous droughts there is no significant recovery in monthly rainfall in the months following the onset of the wet season. The JULES model is more sensitive to scale than non-linearity in the calculation of soil hydraulic parameters. A strong correlation between the model’s near surface soil moisture and the ERS satellite near surface soil moisture product in areas where the satellite is believed to perform well, the RMSE and the similarities in the pattern of anomalies between the model and ERS satellite surface soil moisture is an indication of the ability of the model to successfully simulate land surface conditions in the study area. Simulations into the future (2046-2065) using the IPCC A2 emission scenario suggest a significant change in the land surface conditions due to changes in climatic conditions rather than changes in land cover fraction, despite a projected change in land cover based on previous trends from a predominantly broadleaf trees to a dominance of C4 grass (mostly croplands).