Identifying And Predicting Climate Stress For Agricultural Practices In The North West Region Of Syria

Abstract

Climate changes affect all sectors of life, especially the agricultural sector, where rising temperatures and drought have caused a decrease in many types of agricultural crops, which poses a threat to global food security. Here we examine potential changes in climate variables (precipitation and temperature) over the freed areas of northern Syria, with the aim of developing a new prediction system for multi-year agrometeorological risks (i.e. drought and extreme heat) over the freed areas of northern Syria. We first conducted interviews, highlighting that regional practitioners do adapt their practices depending on weather/climatic forecast, switching to crops that are more resilient to drought, or adapting their agricultural calendar, but that stressing the need for more reliable forecast of agrometeorological risks. Using ERA5 data between 1979 and 2021, we indeed found an increase in drought risks, which is strongly related to an increase in maximum temperature, enhancing evapotranspiration.  We thus test the benefit of neural network in providing reliable prediction for maximum temperature and drought indices.Preliminary results are promising with minimal errors in the mean, and in the variance of predicted data, as compared to the original data. Therefore, we implemented different sensors over the freed areas of northern Syria to monitor climate variations, and to set a live monitoring system, from which new and accurate prediction for climate stress will be provided on seasonal basis.