This report presents the outcome of a research undertaken as part of the BNRCC project. The project aimed to study how global warming could induce climate change in Nigeria, and understand the potential impacts on the agricultural and human health in the country. The understanding would guide the policy makers in implementing suitable climate change mitigation and adaptation options in the country. Trends in the past climate over Nigeria were investigated by analyzing the historical climate records; generation of the future climate change information by downscaling two future climate projections (called A2 and B1) from 9 Global Climate Models; and assessment of the impacts of the climate change on agriculture and human health by using the climate change information in crop-yield and disease-epidemic models, respectively. The following is the step by step summary of the major findings, problems encountered, and recommendations for future works.

Scientific Findings:

• The historical record between 1961 and 2000 shows a significant positive trend in temperature over the country. The trend, which is statistically significant (90% confidence level), is about 0.01oC per year for maximum temperature and 0.02oC per year for minimum temperature. This indicates that, over the period, the maximum temperature has increased by 0.4oC, the minimum temperature by 0.8oC. Also, over the period, the number of cold nights and cold days has decreased, while the number of hot days and nights has increased, over the period.

• No significant trend was found in rainfall over the country between 1961 and 2000.

• The simulation of the past (baseline, 1961-2000) climate, downscaled from the 9 CGMs, reproduced the Nigerian climate very well. The simulation replicated all the essential climatic features in the spatial and temporal distributions of temperature and rainfall over the county. The highest error in the simulated temperature is less than 2oC, and that of rainfall is about 2 mm/day. The simulation also shows a positive trend in maximum and minimum temperatures and no trend in rainfall over Nigeria. Hence, level of the agreement between the simulation and observation showed that the statistical downscaling of the 9 GCMs simulations is reliable for studying the climate change over Nigeria.

• The future climate projections, downscaled from the 9 CGMs for B2 and A1 scenarios, suggest a future increase in temperature over the entire country. The increase is higher with A1 scenario than with B1 scenario. While B1 scenario projects a consistent temperature increase of about 0.02oC per year from 2000 till 2100, A2 projects a temperature increase of about 0.04oC per year from now till 2050, and about 0.08oC from 2050 till 2100. Both scenarios suggest a higher temperature increase over inland station than over the coastal stations. For instance, the highest increase in maximum temperature (about 4.5oC in 2008-2010 for A2 scenario) is expected over the northeast regions (Short grass savanna) and the lowest increase (about 3.5oC in 2008-2010 for A2 scenario) over the southwest (Rainforest). This temperature distribution is consistent with those given over Nigeria in the IPCC report.

• The future projections show no specific trend in future rainfall over the entire country, but the expected changes in rainfall show some spatial variation. Both A2 and B1 scenarios suggest a wetter climate in south (at least 0.2 mm/day south of 8oN), but a drier climate to the north, in the future. In the projection, the south zones would experience moister climate under B1 scenarios than under A2 scenarios, while the northeast zone would be much drier with A2 scenario than with B1 scenario. The highest rainfall variability in all the zones occurs in July, and the variability varies from about ±0.5 mm/day in B1 scenario to about ±2 mm/day in A2 scenario.

• The scenarios suggest a longer rainfall season (i.e. earlier onset and later cessation) in future over Mangrove, Rain forest and Tall Grass savanna, but shorter rainfall season (i.e. early cessation) over Short grass savanna. The expected maximum increase in the rainfall season is about 2 weeks, and the maximum decrease is about a week, for both scenarios.

• More rainy days, more days with of extreme rainfall, and more flooding are projected over the zones, over the northeast region, where A2 scenario suggest less of these extreme events over the Sahel zone.

• Heat waves are projected to occur more frequently over the entire country in the future. Over the Short grass savanna zone, Number of days with heat wave is anticipated to increase by 85% for B1 scenario (and by 95% for A2 scenario) in 2046-2065. These are projected to double for both scenarios in 2080-2100.

• The future climate changes could have devastating negative impacts on agriculture in Nigeria, because the changes decrease crop productivity (i.e. maize yields) over the entire country. The greatest impact occurs northeast, where a drier and hotter future climate is projected. And A2 scenario would produce a worse impact on the crop yield than B2 scenario does. For example, in 2081-2100, B2 scenario projects a 10% decrease in maize in the south, 20% in the northeast; but A2 scenario projects higher values: 30% and 50% decrease, respectively.

• In addition, the climate change could also have a severe effect in the health sector, because the changes enhance disease (e.g. Malaria) epidemics over the entire country. The highest increase in the malaria epidemics is projected to occur in the south, where a hotter and wetter climate is expected. The increase malaria epidemics with A2 scenario than with B1 scenario. For instance, with B1 scenario, more than 40% increase in malaria epidemics is expected over Mangrove in 2081-2100, but with A2 scenario, it is about 100%.  In A2 scenario the number of months with malaria epidemics would increase by 3 - 4 months over the Mangrove zone.

Problem Encountered:

• The network of 40 meteorological stations in Nigeria provides only provides a fair coverage over the country; the coverage worst in the northern part the country. In addition, the station data required rigorous quality control before we could use them for the downscaling or trend analysis. In some station, missing data limits the application of the data for the trends analysis, so we had to combine the data with the CRU observation dataset.

• It was very difficult to obtained observed crop yield data over Nigeria to validate the baseline simulation from the crop model. Many stations have no crop yield record, and in even the few station records, the data were over short periods with a lot of missing data. Hence, it was difficult to compare the long-term mean and the yearly variability with those from crop yield model. To circumvent this problem we combined the available station data with zonal crop yield data reported in the literature, and based our model validation on zones, rather than stations.

• Accessing heath related records in Nigeria for this study was very difficult. Despite all our efforts and contacts, we could only obtain malaria occurrence data from 3 stations (Akure, Ibadan and Lagos). And the data only cover 10 years, and have many data gaps. There is virtually no literature on malaria epidemics in Nigeria from which we could glean more data. So we had to base the validation and fine-tuning of our malaria model on the data from 3 stations. However, this would not have any significant effect on the model results, because the model does not use the observed data as inputs.

• For the purpose of flood risk modeling, the downscaled data is limited in spatial coverage as well as the simulation of the maximum values under the climate change. This problem is hard to circumvent, though increase in the frequency of high flooding events can still be simulated and used as a proxy for the maximum floods.

Recommendation:

• Use a dynamic model to downscale estimate of changes in extreme (maximum) rainfall.

• Reforestation over Nigeria remains a viable climate change mitigation option. However, before embarking on such expensive project, we recommend a series of climate modeling experiment using reforestation scenarios with the aim of quantifying how much the reforestation could reduce the projected temperature over Nigeria.

• Look at developing scenarios of near-term climate change, based on the existing trends and knowledge of how the climate may change in the next 20 years. This should be developed as a scenarios rather than using CGMs to model the future climate.