| Title |
Uncertainty among rainfall products. Modelling household nutrition in Southern Africa. |
? |
| Author |
Luetkemeier, R., Stein, L., Drees, L. & S. Liehr |
? |
| Abstract |
Good quality data on spatial and temporal precipitation patterns are a prerequisite for a range of applications like short-term weather forecasts, medium-term humanitarian assistance programmes and long-term climate modelling. In Sub-Saharan Africa, however, the terrestrial climate station networks are frequently insufficient. Comprehensive datasets of climatic variables are sparse as in the Cuvelai-Basin in Angola and Namibia. Reliable estimates of rainfall are thus difficult to obtain but of high importance for monitoring food security conditions in African food systems that primarily depend on rain-fed subsistence agriculture. This paper analysed six commonly used monthly rainfall products (CHIRPS 2.0, GPCC v7, TAMSAT, ARC 2.0, CRU-TS 3.23, TRMM 3b43) with respect to their spatial and temporal quality in comparison with sparsely available climate station records. For the purpose of estimating the degree of uncertainty that propagates through a modelling stage, the products were used as input data for an exemplary crop growth model to obtain nutritional scores of an average household’s requirements for dietary energy, proteins, lipids and carbohydrates. The results show that the rainfall products predict consistent spatial and temporal rainfall patterns in accordance with the climate station records. However, large differences exist when it comes to absolute values of monthly and annual rainfall. This uncertainty is mitigated by crop model parameters like soil water properties but results in uncertain estimates of nutritional scores.
The study results emphasise the importance of considering the differences between rainfall products, especially when ground measurements are scarce. Ensemble modelling can account for this uncertainty and prevent misinterpretations. |
? |
| Citation |
Luetkemeier, R., Stein, L., Drees, L. & Liehr, S. (2017): Uncertainty among rainfall products. Modelling household nutrition in Southern Africa. In: South African Journal of Science (subm.) |
? |
