Multi-temporal and Multi-Scale Remote Sensing Techniques to Assess the Risk of Crop Production in Soil Salinization Scenario

As the world population reaches eight billion people with indications of further growth, food security becomes one of the most important topics. At the same time, agriculture is facing a loss of arable land, reducing production capacities. Soil salinization represents a growing threat to coastal agriculture, as the combination of sea level rise and prolonged drought conditions. Identifying and mapping areas prone to this type of risk might help apply precise soil meliorative techniques and choose the right crop to grow in these soils. Nowadays, remote sensing techniques can provide valuable information for this purpose, thanks to frequent and low-cost data at different spatial scales. In this research, the Structure from motion (SfM) technique paired with Unmanned Aerial Vehicles (UAV) was used to assess the fitness of two different crops: soybean (Glycine max) and maize (Zea mays) in different salt-affected fields, in the Po river delta, North-Eastern Italy. Multi-temporal SfM surveys, using a multi-spectral camera, were conducted in July and August 2022 to map the consequences of high soil salinity (due to significant drought, low discharge, and consequent saltwater intrusion along the reaches of the Po river delta) on the vegetative status of crops through vegetation indices like the Normalized Difference Vegetation Index (NDVI). Moreover, to measure the salinity level, geolocated soil samples were taken from each field, and the amount of salt was determined using electrical conductivity using XS Instruments COND 80 electrical conductivity meter (Giorgio Bormac s.r.l, Carpi, Italy) at a sensitivity of 1 µS. Salinity values measured in the field were used to create salinity maps through spatial interpolation in GIS software. The latter allowed the salinity maps to be compared with orthomosaics of NDVI values obtained from SfM surveys. Furthermore, multi-spectral images from open-source satellites made it possible to broaden the scale of investigation in both spatial and temporal terms and to compare different data acquisition techniques. Results show a clear relationship between high-ground salinity measurements and low NDVI values, highlighting how remote sensing techniques could provide helpful information for monitoring the progressive effects of soil salinity on crops. It can be observed that soybean is quite sensitive to salinity, perishing after a long exposure even to medium-low salinity levels (1.5 dS/m – 2 dS/m). At the same time, maize seems more tolerant, with plants also surviving high salinity levels (more than 5 dS/m). Other than indicating salinity stress to which plants are exposed, these maps could also apply salt-reducing techniques, such as flushing, more precisely, thus obtaining optimal results while saving water.Acknowledgments: this study was carried out within the Agritech National Research Center and received funding from the European Union Next-GenerationEU (PIANO NAZIONALE DI RIPRESA E RESILIENZA (PNRR) – MISSIONE 4 COMPONENTE 2, INVESTIMENTO 1.4 – D.D. 1032 17/06/2022, CN00000022).