It’s not all about drought: what “drought impacts” monitoring reveals about semiarid northeast Brazil

Abstract

Drought impacts monitoring is conducted on the ground in much of Brazil by local observers at monthly and municipality scale. The monitoring was established to contribute towards the Brazilian Drought Monitor (https://monitordesecas.ana.gov.br/, established in 2014) with a multichoice questionnaire principally aimed at ground truthing the monthly map of drought condition, but with additional questions, including an open question, about drought impacts.Our research focussed on Ceará State in drought-prone semiarid northeast Brazil. In Ceará, over 3600 questionnaires were completed by agricultural extension officers since February 2019 based on their visits to a wide area of the municipalities. These local reports about drought impacts have been under-exploited by the Drought Monitor yet represent a rich resource of impact information. We aimed to understand what drought impacts were reported and if there were differences between these local reports and the Drought Monitor maps.We manually coded all the reports to deductively identify impacts and other useful information. Despite some spatial and temporal gaps, the data reveal: a catalogue of the most significant impacts experienced on the ground per municipality per month, impact drivers (including non-climatic drivers), and areas of greater/lesser vulnerability (i.e. where more/less impacts were reported despite matching drought condition).Analysis shows that impacts still occur, and are often normalised during non-drought periods. The impact drivers are either non-extreme hydrometeorological conditions or socially constructed vulnerability such as a lack of water infrastructure or poverty. The normalisation of “impacts” includes, in particular: a level of crop losses that is considered usual (up to 50% losses are acceptable) and consistently low reservoir levels (around 10% of capacity) around which the agricultural and domestic systems are adapted. The frequent non-correspondence of Drought Monitor drought severity and experienced drought severity suggests the Drought Monitor, and other Drought Monitors around the world based on traditional hydroclimatic indices, are not optimal for triggering emergency response, which they are often (mis)used for, though are relevant for triggering discussion and action on drought preparation, as is usually their principal aim.