How to predict floods and droughts to minimize crop damage

The current drought that affects eastern Argentina, Uruguay and southern Brazil, and which began in 2019, is classified as one of the five major drought events since 1950. Its persistence caused its impacts to be manifested in different aspects of the climate, such as the state of soil moisture, the flow of rivers and vegetation.

This phenomenon has generated a lot of tension in the agricultural sector in the region, since the performance of the production of the different crops is very sensitive to changes in climatic conditions.

Impact on agriculture

Agricultural activities in Argentina are very vulnerable to atmospheric and climatic phenomena. Phenomena such as droughts and floods are the main causes of crop yield loss with a strong economic impact.

The Agricultural Risk Office (ORA) of the Ministry of Agriculture, Livestock and Fisheries of Argentina specializes in the evaluation of agricultural risk and, in particular, agroclimatic risk. To carry out this task, it has generated a series of tools that allow real-time monitoring of soil moisture conditions for specific crops.

These tools combine information on current weather conditions with information on crop development (times for sowing, flowering, among others) to understand whether water conditions can affect the normal development of crops. In this way, they allow producers to plan agronomic tasks and the ORA to estimate early the decrease in crop yields in the campaign and probable losses.

Evolution of the water reserve in the soil for the town of Paraná, Entre Ríos (Argentina). The horizontal axis indicates the data registration dates and the vertical axis indicates the water content in the soil (mm).

Agricultural Risk Office

Climate variability in Argentina

The climatic variability experienced by Argentina, and especially its central-eastern region, has been extensively studied in recent decades, both to understand the causes that produce this variability and the ability to forecast it months in advance.

In particular, it is known that the phenomenon of El Niño – Southern Oscillation (ENSO), which produces changes in the temperature of the Tropical Pacific Ocean, affects the occurrence of rainfall in southeastern South America.

During El Niño events, where the Tropical Pacific Ocean is warmer than its typical conditions, rainfall tends to be more abundant in the east of Argentina. The opposite tends to occur during La Niña-related events, like the one currently occurring, where rainfall is usually scant.

Advances in research have allowed the development of numerical weather and climate forecast models that predict the development of ENSO phenomena with moderate success up to 6 months before its occurrence.

These numerical tools are also used to estimate weather conditions months in advance, although with uneven performance depending on the region of interest. In the case of southeastern South America, forecasts of the weather conditions for the following quarter can be obtained with good performance.

These forecasts are probabilistic in nature. They do not provide a specific value of expected rainfall or temperature, but rather the probability that these variables are very different from their typical values.

Forecasts for the agricultural sector

From the strategic development project Application of Climate Forecasts to Agricultural Risk Management, a group of researchers from the Faculty of Economic Sciences and the Faculty of Exact and Natural Sciences from the University of Buenos Aires we have worked together with the ORA to extend the application of climate forecasts to monitoring tools for soil conditions.

Within the framework of this collaboration, we generate 30-day forecasts of the hydric conditions of soybean cultivated soil in different stations in Argentina.

To develop this tool, we use 30-day forecasts of climate variables that influence soil moisture conditions, such as maximum and minimum temperature, rainfall, wind, and solar radiation intensity.

Perspective of the water reserve in the soil for the locality of Paraná, Entre Ríos, based on the forecasts initiated on December 1, 2021. The horizontal axis indicates the dates from the sowing of the considered crop to the estimated completion of its cycle . The vertical axis indicates the water content in the soil (mm).

Author provided

Forecasts often have errors, since it is not possible to represent the weather perfectly. Therefore, we apply mathematical techniques that correct systematic errors to improve their performance.

Then, these forecasts are implemented in the tools developed by the ORA to obtain a 30-day perspective of soil conditions in different locations in Argentina. As more than one forecast is used in the development of the outlook, it is presented in such a way that it shows the uncertainty associated with future conditions, which allows estimating the probability that the water conditions are between normal or atypical values.

Research shows that forecasts like the ones we developed under this project tend to perform well during the first 10 days of the chosen period and its quality degrades over time. However, under certain climatic conditions, forecasts may perform well. up to the first 20 days.

Having good predictions on these time scales can allow farmers to take early action in case of extreme events. They would also allow the ORA to develop quantitative assessments of agroclimatic risks related to future climatic conditions in real time.

Marisol OsmanPostdoctoral Researcher at Karlsruhe Institute of Technology, Researcher at Center for Research on the Sea and the Atmosphere (CONICET/UBA), Buenos Aires’ University

This article was originally published on The Conversation. read the original.

Leave a Comment