Specific humidity verification skill of ECMWF weather forecast model in the region of Vietnam

In the context of an accelerated warming climate, outbreak episodes of certain diseases such as dengue are expected to increase in the future, thus increasing health burden in areas such as Vietnam. Because meteorological variables exert a great influence on dengue transmission, having accurate weather forecast data is key to predict future dengue outbreaks and mitigate future damages. Previous studies assessed the skill of ECMWF ensemble model on precipitation and temperature over Vietnam, but they did not include specific humidity, which also plays an important role in vector-borne disease transmission. Hence, the goal of this study is to assess the skill of ECMWF ensemble model at predicting specific humidity in the area of Vietnam, focusing on the regions of North and South Vietnam (particularly the area nearby Hanoi city and Ho Chi Minh City or HCMC). We used ERA5 reanalysis between the dates 2001-2020 for the evaluation. ECMWF data was preprocessed using a bias correction technique (quantile mapping) using ERA5 as reference to correct possible biases. Early results showed that ECMWF model predictions show a little bias (around 0.5 g/kg) and that it is usually skilful up to 10 days in advance, specially during dry season. Nonetheless, there is a considerable decrease in skill during the rainy season (from May to October), which is the period when Dengue activity is higher. These results will provide more insight about the ECMWF skill in the medium-range forecast and it will be useful to stablish Dengue-modelling warning systems and reduce future damages caused by this disease.