The Effects of Meteorological Factors on Dengue Cases in Malaysia

Abstract

Dengue is a vector-borne disease affected by meteorological factors and is commonly recorded from ground stations. Data from ground station have limited spatial representation and accuracy, which can be overcome using satellite-based Earth Observation (EO) recordings instead. EO-based meteorological recordings can help to provide a better understanding of the correlations between meteorological variables and dengue cases. This paper aimed to first validate the satellite-based (EO) data of temperature, wind speed, and rainfall using ground station data. Subsequently, we aimed to determine if the spatially matched EO data correlated with dengue fever cases from 2011 to 2019 in Malaysia. EO data were spatially matched with the data from four ground stations located at states and districts in the central (Selangor, Petaling) and east coast (Kelantan, Kota Baharu) geographical regions of Peninsular Malaysia. Spearman’s rank-order correlation coefficient (ρ) was performed to examine the correlation between EO and ground station data. A cross-correlation analysis with an eight-week lag period was performed to examine the magnitude of correlation between EO data and dengue case across the three time periods (2011–2019, 2015–2019, 2011–2014). The highest correlation between the ground-based stations and corresponding EO data were reported for temperature (mean ρ = 0.779), followed by rainfall (mean ρ = 0.687) and wind speed (mean ρ = 0.639). Overall, positive correlations were observed between weekly dengue cases and rainfall for Selangor and Petaling across all time periods with significant correlations being observed for the period from 2011 to 2019 and 2015 to 2019. In addition, positive significant correlations were also observed between weekly dengue cases and temperature for Kelantan and Kota Baharu across all time periods, while negative significant correlations between weekly dengue cases and temperature were observed in Selangor and Petaling across all time periods. Overall negative correlations were observed between weekly dengue cases and wind speed in all areas from 2011 to 2019 and 2015 to 2019, with significant correlations being observed for the period from 2015 to 2019. EO-derived meteorological variables explained 48.2% of the variation in dengue cases in Selangor. Moderate to strong correlations were observed between meteorological variables recorded from EO data derived from satellites and ground stations, thereby justifying the use of EO data as a viable alternative to ground stations for recording meteorological variables. Both rainfall and temperature were found to be positively correlated with weekly dengue cases; however, wind speed was negatively correlated with dengue cases.

The practicality of Malaysia dengue outbreak forecasting model as an early warning system

“…Other rain gauges also showed similar findings. Thus, indicating the existence of localized rainfall in both areas, which often occurs in Malaysia ( Muhammad et al., 2020 ; Singh et al., 2022 ). Next, we extracted the moderate correlation and respective distances as presented in Table 7 for both areas.…”

“…But the rainfall data (secondary data) obtained from the Meteorological Department based on their weather stations showed insignificant results. This is because Malaysia's rainfall pattern is very localized ( Muhammad et al., 2020 ; Singh et al., 2022 ), meaning there are rains in certain areas but no rain just a few blocks away. The localized rain can jeopardize rainfall data quality (obtained from Meteorological Department).…”

“…The localized rain can jeopardize rainfall data quality (obtained from Meteorological Department). But rainfall is an important indicator in dengue outbreak forecasting model ( Hii et al., 2012 ; Liu, Yin, et al., 2021 ; Ong et al., 2018 ; Patil & Pandya, 2021 ; Singh et al., 2022 ) because rainfall influence the Aedes mosquito population (the vector of dengue virus) to breed. Therefore, a practical solution is by placing several rain gauges (mobile weather stations) in the dengue hot spot areas, thereby enhancing the quality of the rainfall data, and potentially increasing the accuracy of the dengue outbreak forecasting model.…”

Review of Importance of Weather and Environmental Variables in Agent-Based Arbovirus Models

“…Infectious diseases such as dengue, malaria, chikungunya, Zika, and yellow fever, to mention a few, are emerging as a worldwide challenge in public health. They occur rapidly and spread to large areas in a relatively short space of time, thus leading to an increase in mortality rates globally as well as in Sub-Saharan Africa (SSA) [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ].…”

“…Arthropods are abundant in tropical and subtropical regions, resulting in the high proportion of arboviruses in these regions [ 6 ]. The global distribution of viruses has caused some arboviruses to be endemic in specific regions of the world.…”