Spatial correlates of COVID-19 first wave across continental Portugal.

The first case of COVID-19 in continental Portugal was documented on the 2nd of March 2020 and about seven months later more than 75 thousand infections had been reported. Although several factors correlate significantly with the spatial incidence of COVID-19 worldwide, the drivers of spatial incidence of this virus remain poorly known and need further exploration. In this study, we analyse the spatiotemporal patterns of COVID-19 incidence in the at the municipality level and test for significant relationships between these patterns and environmental, socioeconomic, demographic and human mobility factors to identify the mains drivers of COVID-19 incidence across time and space. We used a generalized liner mixed model, which accounts for zero inflated cases and spatial autocorrelation to identify significant relationships between the spatiotemporal incidence and the considered set of driving factors. Some of these relationships were particularly consistent across time, including the 'percentage of employment in services'; 'average time of commuting using individual transportation'; 'percentage of employment in the agricultural sector'; and 'average family size'. Comparing the preventive measures in Portugal (e.g., restrictions on mobility and crowd around) with the model results clearly show that COVID-19 incidence fluctuates as those measures are imposed or relieved. This shows that our model can be a useful tool to help decision-makers in defining prevention and/or mitigation policies.

A comprehensive approach to understanding flood risk drivers at the municipal level.

During the period 1998-2017, floods were responsible for 11% of the loss of life and 23% of the economic loss caused by climate-related and geophysical-related disasters worldwide. An integrated and effective definition of flood risk management strategies therefore still requires synthesized and comprehensive knowledge about the driving forces of flood risk. In this study, 278 Portuguese municipalities are analyzed and classified according to flood hazard, exposure, and vulnerability. After evaluating the three components that describe risk, an index of the flood risk is calculated and a cluster analysis is further performed to understand the role of the risk drivers (hazard, exposure, and vulnerability) in each municipality. The proposed approach therefore provides flood risk indexes on a municipal basis, which are built upon different sources of both cell-by-cell data and an aggregation of municipal-level data that has been statistically validated. Municipalities both in the NW part of the country and along the valleys of major rivers demonstrate a significant superimposition of high levels of exposure and hazard, while vulnerability presents a disperse pattern throughout the country. The results obtained using this approach should contribute to the diversification of flood risk management strategies. This is still lacking in the majority of the national-level flood risk governance processes, namely those strategies that focus on the contingency of daily activities and those aiming at a long-term reduction of the exposure, vulnerability, and hazard components that shape flood disasters.