INVESTIGADORES
LOVINO Miguel Angel
congresos y reuniones científicas
Título:
Flash drought hotspots around the world
Autor/es:
MIGUEL A. LOVINO; MARÍA JOSEFINA PIERRESTEGUI; OMAR V. MÜLLER; ERNESTO H. BERBERY; GABRIELA V. MÜLLER
Lugar:
Houston
Reunión:
Congreso; 102nd annual meeting AMS; 2022
Institución organizadora:
AMS
Resumen:
Flash droughts (FD) are characterized by rapid drying of the soil, causing severe impacts worldwide. In agricultural regions, they can negatively impact crop yields and damage natural ecosystems. The goal of this study is to identify the regions in the world prone to flash droughts. To this end, we propose a novel definition of flash drought that: a) reflects soil moisture stress, b) is simple and applicable on a global scale, and c) avoids current shortcomings of indices applied over very dry/wet climates. The definition is based on root-zone soil moisture (SM) along with soil properties associated with the soil type. Here, we employ ERA5 1-m root-zone SM pentad data for the 1950-2020 period complemented with the ERA5 field capacity and wilting point volumetric soil moisture values associated with each soil class in each grid point. Drought impacts agriculture when plant water requirements are not met. A drought event must reflect the soil moisture stress on the land system. In this study, the flash drought definition is based on the Soil Water Deficit Index (SWDI) to assess both soil moisture and plant water stress. SWDI is based on two characteristics, the field capacity and the wilting point. SWDI equals zero when there is no water deficit (the soil is at the field capacity). Negative SWDI values indicate a soil water deficit, which becomes absolute (wilting point) when SWDI is less than -10. At this point, the SM is below the lower limit of available water for plants.We define a flash drought when the SWDI decays from -3 to -5 in a period of 20 days (4 pentads) to ensure the rapid depletion of root zone soil moisture. We also require that the SM depletion lasts at least 15 days (3 pentads) to exclude those potential short events that deplete SM rapidly down within ten days but then recover up suddenly. The lower threshold (SWDI = -5) is selected to ensure severe drought conditions, while the upper threshold (SWDI = -3) is chosen to take into account that moderate drought conditions are previously established. The range is selected after rigorous comparative analysis with previously established definitions of SM percentiles. Using our FD definition, we detect FD hotspots by counting the total number of FD events during the whole year over the 1950?2020 period. To determine the spatial pattern of changes in the occurrence of FD, we estimate linear trends by least squares in time series of the number of FD events per year. Trends that are statistically significant at the 5% significance level are detected by the Mann?Kendall test.We find that several regions around the world that are most beneficial for agriculture are also the hotspots for flash drought occurrence. This is the case of croplands over central and eastern EEUU, southeastern South America, India, southern China, central Europe, southwestern Russia, and the transition belt between the Sahel and the tropical forests in Central Africa. Over most of these regions, more than 35 FD events were recorded throughout the study period, implying a frequency of more than one FD event every two years. Our spatial patterns of FD frequency are consistent with previous regional studies in the US, China, and India that used different FD definitions.An evaluation of trends reveals that flash droughts increased in most of the world between 1950 and 2020. The most significant upward trends occurred in regions previously characterized as hotspots for FD occurrence. The mean occurrence of FD increased by 100% in Central Europe, 82% in southern Brazil, 45% in southern China, and 42% in the eastern United States between 1951-1970 and 2001-2020 periods.