Climate Change and Variability Effects on Malaria Prevalence in the Limpopo Province, South Africa

Abstract

As the world strives towards the elimination of the highest death causing vector-borne disease, namely malaria, research is needed in order to contribute towards this goal in South Africa. This study contributes towards the goal of eliminating malaria. The transmission of malaria is a complex biological process, which is associated with not only climate conditions but non-climate conditions like human behaviour, socio-economic, political situation and other environmental factors like ecology, hydrology and associated conditions. Research has shown that in malaria prone areas climatological elements have an effect on the propagation of malaria. However, the extent of the effect and the most influential climatological element vary from area to area when the local level is considered. This study sought to understand the strength of the relationship between malaria and climatological elements at the local level in Vhembe District, Limpopo Province, South Africa. The study analysed the climate elements rainfall, temperature and Relative Humidity, in terms of their effect and influence on malaria in Vhembe District. Weather data from the network of weather stations in Vhembe District were obtained for the period 1981-2010, from the South African Weather Service in line with the 30-year period recommended by the World Meteorological Organisation (WMO). Data on malaria cases were obtained from the Regional Malaria Control Institute in Limpopo Province. The malaria data were only available in suitable form for a period from 1998-2010. The relationship between malaria and climate was, therefore, only analysed for the period 1998-2010. Climate normals for the area were obtained using the 1981-2010 data, per local municipality and then at the aggregated level of Vhembe District. Analysis of the data included the computation of anomalies (departures from the long-term mean), Mann-Kendall trend analysis, Pearson correlation coefficient (r) analysis, coefficient of determination (R2 ) in assessing the variation in malaria cases as influenced by climate, lag-time analysis by cross-correlation to determine the lag between climate variable alteration and the onset of malaria, and mapping of malaria cases per given land demarcation unit in a Geographic Information System (GIS). The results indicated that the climate of Vhembe District was quite variable in the analysis period, with very few significant trends established for the area. The association between malaria and climate varied for the four local municipalities in Vhembe District which are Musina, Makhado, Thulamela and Mutale. The most influential climatological element on malaria was determined as the monthly minimum temperature if the existing Relative Humidity conditions are maintained. Statistical analysis showed that there was a statistically significant relationship between malaria and annual Relative Humidity (r = 0.709 at a = 0.05 level of significance) for Vhembe District. It becomes important that the minimum temperatures in the district are monitored more closely since this can help in early detection and surveillance of malaria in the malaria prone areas. The only significant annual long term trends for the 30-year period 1981-2010 were a decreasing air temperature for Thulamela (Z-statistic = 1.67 at a = 0.1 level of significance), an increasing maximum temperature for Musina (Z-statistic = 1.93 at a = 0.1 level of significance) and Vhembe (Z-statistic = 1. 71 at a = 0.1 level of significance), increasing minimum temperatures for Makhado (Z-statistic = 2.03 at a = 0.05 level of significance); increasing minimum temperatures for Vhembe (Z-statistic = 3.28 at a = 0.01 level of significance) and decreasing Relative Humidity for Mutale Local Municipality (Z-statistic = 1.71 at a = 0.1 level of significance). In terms of short term trends, i.e. between 1998 and 2010, the climate for the municipalities indicated mostly decreasing trends including in the number of malaria cases. The GIS-derived spatial distribution maps showed that there was a pocket nature in the distribution of malaria in Vhembe District, and that altitude somewhat influenced the spatial distribution. Most of the malaria cases were reported in areas below 700 meters above sea level (MASL), with very few and erratic cases between 701 and 1300 MASL. Based on the results, a Malaria-Climate Management Framework (CliMM) was developed. The aim of the CliMM framework is to help in identifying malaria prone villages in Vhembe District. After identification, the CliMM framework then leads into the steps that may be taken to manage the particular area/village towards minimisation and possibly elimination of malaria. With the proposed CliMM framework, climate considerations can be brought into the management of malaria in Vhembe District.