Predicting malaria cases using remotely sensed environmental variables in Nkomazi, South Africa

  • Abiodun Morakinyo Adeola | abiodun.adeola@weathersa.co.za South African Weather Service, Pretoria; UP Institute for Sustainable Malaria Control, School for Health Systems and Public Health, University of Pretoria, Pretoria, South Africa. http://orcid.org/0000-0002-6105-7110
  • Joel Ondego Botai South African Weather Service, Pretoria; Department of Geography, Geoinformatics and Meteorology, University of Pretoria, Hatfield; School of Agricultural Earth and Environmental Sciences, University of KwaZulu-Natal, South Africa.
  • Jane Mukarugwiza Olwoch Department of Geography, Geoinformatics and Meteorology, University of Pretoria, Hatfield, South Africa; Southern African Science Service Centre for Climate Change and Adaptive Land Use (SASSCAL), Windhoek, Namibia.
  • Hannes C.J. de W. Rautenbach South African Weather Service, Pretoria; UP Institute for Sustainable Malaria Control, School for Health Systems and Public Health, University of Pretoria, Pretoria; Faculty of Natural Sciences, Akademia, Centurion, South Africa.
  • Omolola Mayowa Adisa Department of Geography, Geoinformatics and Meteorology, University of Pretoria, Hatfield, South Africa.
  • Christiaan de Jager UP Institute for Sustainable Malaria Control, School for Health Systems and Public Health, University of Pretoria, Pretoria, South Africa.
  • Christina M. Botai South African Weather Service, Pretoria, South Africa.
  • Mabuza Aaron Malaria Control Programme, Mpumalanga Department of Health, Nelspruit, South Africa.

Abstract

There has been a conspicuous increase in malaria cases since 2016/2017 over the three malaria-endemic provinces of South Africa. This increase has been linked to climatic and environmental factors. In the absence of adequate traditional environmental/climatic data covering ideal spatial and temporal extent for a reliable warning system, remotely sensed data are useful for the investigation of the relationship with, and the prediction of, malaria cases. Monthly environmental variables such as the normalised difference vegetation index (NDVI), the enhanced vegetation index (EVI), the normalised difference water index (NDWI), the land surface temperature for night (LSTN) and day (LSTD), and rainfall were derived and evaluated using seasonal autoregressive integrated moving average (SARIMA) models with different lag periods. Predictions were made for the last 56 months of the time series and were compared to the observed malaria cases from January 2013 to August 2017. All these factors were found to be statistically significant in predicting malaria transmission at a 2-months lag period except for LSTD which impact the number of malaria cases negatively. Rainfall showed the highest association at the two-month lag time (r=0.74; P<0.001), followed by EVI (r=0.69; P<0.001), NDVI (r=0.65; P<0.001), NDWI (r=0.63; P<0.001) and LSTN (r=0.60; P<0.001). SARIMA without environmental variables had an adjusted R2 of 0.41, while SARIMA with total monthly rainfall, EVI, NDVI, NDWI and LSTN were able to explain about 65% of the variation in malaria cases. The prediction indicated a general increase in malaria cases, predicting about 711 against 648 observed malaria cases. The development of a predictive early warning system is imperative for effective malaria control, prevention of outbreaks and its subsequent elimination in the region.

Downloads

Download data is not yet available.
Published
2019-05-14
Section
Original Articles
Keywords:
Malaria, Environmental, Climatic, Remote sensing, SARIMA, Prediction, South Africa
Statistics
Abstract views: 507

PDF: 289
APPENDIX: 49
HTML: 11
Share it

PlumX Metrics

PlumX Metrics provide insights into the ways people interact with individual pieces of research output (articles, conference proceedings, book chapters, and many more) in the online environment. Examples include, when research is mentioned in the news or is tweeted about. Collectively known as PlumX Metrics, these metrics are divided into five categories to help make sense of the huge amounts of data involved and to enable analysis by comparing like with like.

How to Cite
Adeola, A., Botai, J., Mukarugwiza Olwoch, J., de W. Rautenbach, H., Adisa, O., de Jager, C., Botai, C., & Aaron, M. (2019). Predicting malaria cases using remotely sensed environmental variables in Nkomazi, South Africa. Geospatial Health, 14(1). https://doi.org/10.4081/gh.2019.676