Agriculture along the coast depends on optimal water resource management, especially in delta areas. Indeed, freshwater aquifers close to the sea are at risk of saltwater intrusion, with disastrous consequences for crops. This is a complicated process that is the result of multiple factors, both natural and man-made. Droughts are responsible for particularly severe saltwater intrusion events. Indeed, lack of rainfall leads to reduced river flow, which favors the flow of marine water inland. Climate change is aggravating this condition in diverse parts of the globe. Therefore, it is crucial to investigate the process of saltwater intrusion in river deltas deeply. Although this phenomenon has already been addressed in the literature in some areas of the world, there is still much to be done to assess the effects of saltwater intrusion on crops at the sub-regional level. In this task, multi-temporal remote sensing opens up new horizons of knowledge. New Earth observation (EO) technologies make it possible to monitor the evolution of the process over several years of observation and vast areas. The open-source big data offered by international space programmes are an excellent starting point for understanding this trend. This study aims to examine the effects of saltwater intrusion on agricultural greening in the Po Delta (Italy), an important European food production area. The main economic activity in the area is agriculture, made possible by centuries of land reclamation and co-existing with wetlands of considerable ecological importance, now threatened by the salinization of the water. In fact, during dry summers, the Po River's flow rate decreases significantly, favoring the intrusion of saltwater for tens of kilometers inland, affecting irrigation systems and causing severe impacts on production. This study's analyses are based on the correlation between Po river water salinity at 47 sampling sites and NDVI values using Landsat 5. The research results could provide a low-cost, multi-temporal tool based on remote sensing to quantify/map the effects of saltwater intrusion on agriculture at the delta scale, helping stakeholders adopt a more efficient/sustainable use of freshwater near the sea.

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). The authors also thank the Start-up funding from Inner Mongolia University (21800-5223728).

How to cite: Luo, J., Straffelini, E., and Tarolli, P.: Multitemporal remote sensing to investigate saltwater intrusion impact on agricultural greening in the Po River Delta (Italy), EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1868, https://doi.org/10.5194/egusphere-egu23-1868, 2023.

Monitoring crop physiological responses to drought is crucial to understand progressive impacts on food production and identify resilient and sustainable irrigation practices. Although many climate research experiments provide valuable data, long-term measurements of soil properties and plant characteristics at the field scale are not always affordable. We combined drone-based multispectral remote sensing with measurements of soil properties over multiple pilot farms, where soil sampling was performed for each plot during the drone survey. Our goal was to determine if drone-based indices capture drought stress responses of different crops (maize and sugar beet) and whether responses are affected by soil physical and chemical characteristics (e.g., texture, density, porosity, moisture, pH, electrical conductivity, organic carbon and total nitrogen contents, availability of micro and macro nutrients). 
Significant relationships were found between vegetation indices and soil features for different crop types. Differences found at the field scale were related mostly to organic carbon content and resulted in heterogeneous responses to irrigation practices. Our spatial variability analysis pointed out an overall homogeneous response for areas submitted to severe and moderate drought having similar soil properties, independently of the crop type. More investigation is needed to address the possible effect of local practices (e.g., fertilization, amendment, tillage) at the field scale. The feasibility of carrying out systematic drone flights coincidentally or close to-ground campaigns will reveal the consistency of the observed spatial patterns in the long run.

How to cite: Sofia, G., Sinatra, M., Tarolli, P., and Zaccone, C.: Enhancement of drought monitoring by means of soil sampling and drone-based multispectral sensing, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-5230, https://doi.org/10.5194/egusphere-egu23-5230, 2023.

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).

How to cite: Cucchiaro, S., Nikolic, N., Straffelini, E., Masin, R., and Tarolli, P.: Multi-temporal and multi-scale remote sensing techniques to assess the risk of crop production in soil salinization scenario, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-5552, https://doi.org/10.5194/egusphere-egu23-5552, 2023.

Nowadays, agricultural cultivation in steep slope areas represents an additional challenge against global climate change. Due to increased rainfall intensity and their complex geomorphology, steep-slope vineyards are also strongly affected by erosion processes. Finding solutions to mitigate this problem is a priority to ensure sustainable production. A widely used conservation approach against erosion involves maintaining an organic soil cover through herbs, mulching, or reusing crop residues. Conversely, tillage and weed removal can accelerate soil erosion in steep slope areas and influence their micro-topography by altering soil surface roughness and sediment connectivity. The latter, useful to identify the surface portions that are most connected and prone to more significant erosion, has often been neglected in past soil erosion studies but needs to be considered to effectively define the sediment surface contributing to estimates of erosion processes. Remote Sensing techniques (e.g., digital photogrammetry with Structure from Motion-SfM, Light Detection and Ranging-LiDAR technology) have recently provided new opportunities for surveying erosion processes, especially by exploiting Unmanned Aircraft Systems (UAS) that can also mount multispectral cameras as payloads, which, integrated with topographic data, can provide helpful information for analyzing the status of crops. This research carries out a multi-temporal analysis of sediment connectivity following management changes in the row and vineyard inter-row cover to assess the effects on erosion processes using Digital Elevation Models (DEMs) provided by LiDAR and SfM surveys with UAS. All maps were generated with a centimetric resolution to capture the micro-topographic features. The study vineyard (1943 m²) is located in a steep slope area (ca. 18%) in the municipality of Betanzos (43° 15' 56.20" N; 8° 12' 1.32" W; Coruña, Spain) under a temperate oceanic climate and managed according to organic farming practices, without any irrigation system.

In 2021, all rows and inter-row areas had covered with resident vegetation. In 2022, three treatments were carried out in the inter-row areas (i.e., mixed seeding cover, massive straw mulching, and cover by resident vegetation), while all the rows were covered with jute agro-textile. Eight topographic surveys (i.e., SfM with RGB and multi-spectral camera and LiDAR by UAS) were carried out in the study area from March 2021 to November 2022. The multi-temporal DEMs were used to derive multi-temporal maps of sediment connectivity indices (IC), Differences of IC (DoIC), and DEM of Differences (DoDs) to calculate vegetation volumes and estimate net soil loss and deposition rates over time. Subsequently, the geomorphological information was correlated with the multi-spectral surveys by making orthomosaics and calculating vegetation indices (e.g., NDVI, NDRE) that allowed the assessment of the land cover condition as a result of changes in cultural management of vineyard rows and inter-rows. The preliminary results show how the information obtained from the extensive database created (i.e., DEMs, DoDs, CI and DoIC maps, vegetation indices) is very useful in assessing the effectiveness of the conservation cultivation approaches used, identifying the portions of soil potentially more prone to erosive processes to provide a useful planning tool for stakeholders for sustainable vineyard management.

How to cite: López-Vicente, M., Cucchiaro, S., and Tarolli, P.: Multi-temporal analysis to asses different conservative cultural management on erosion processes in steep-slope agriculture through remote sensing techniques, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-5765, https://doi.org/10.5194/egusphere-egu23-5765, 2023.

The territory of north-eastern Italy, crossed by the Po River, which flows eastward into the Adriatic Sea, is home to flourishing agricultural production. The area is among Europe's most important rural regions and is crucial for food production. However, the sector is facing the impact of climate change. Among the most worrying phenomena is an increase in the frequency of more severe and longer drought periods, leading to progressively arid climatic conditions. The summer of 2022 was one of the most critical times on record, with the combination of extreme temperatures and severe water shortages. The effects severely impacted agriculture, with crop loss, irrigation problems, and saltwater intrusion into the Po River delta. Emerging multi-temporal satellite remote sensing technologies and the application of big data-based algorithms allow in-depth knowledge of phenomena occurring on Earth and the subsequent research of mitigation solutions. Specifically, monitoring the impacts of extreme drought in the region can be useful in understanding which areas are most at risk in the short term, while the use of future climate models can guide more resilient agricultural management in the future. This research first proposes the application of multi-temporal MODIS satellite indices to assess the agricultural drought that affected north-eastern Italy in the summer months of 2022 and secondly analyses the possible traces of climatic aridification. In addition, we present a study on the relationship between agricultural lands and current & future climates, carried out using high-resolution climate zone maps (RCP8.5 scenario). The aim is to understand the potential future climate in the currently cultivated fields. Mapping present and future critical areas and knowing which farming systems are most at risk due to climate change can be valuable information for managing agricultural assets under the threat of climate change. 

Acknowledgements - 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).

How to cite: Straffelini, E. and Tarolli, P.: The importance of a multi-temporal approach to assess climate change impacts in Northern Italian agriculture, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-7043, https://doi.org/10.5194/egusphere-egu23-7043, 2023.

Soil piping is a land degradation process relatively common in semiarid environments which is related to rilling and gullying in badland areas. This process is a result from a complex combination of different factors as lithology (e.g., swelling clays), topography (e.g., hydraulic gradient) and climate (e.g., strong seasonal contrasts). Better understanding of piping erosion is needed since it can negatively impact land productivity and agricultural sustainability and may affect soil nutrients load and carbon cycles. Piping studies have been frequently focused on the qualitative and quantitative implications of chemical and physicochemical factors affecting the initiation of piping processes together with a qualitative analysis of the hydrological and geomorphological related processes. However, less attention has been given to the study of these processes from a quantitative point of view. High-resolution topography surveying has improved the spatial and temporal scales at which is possible to investigate the landscape through the analysis of landform attributes and the computation of topographic changes. Within this background, the aim of this work is to infer in the key geomorphic piping processes in terms of contributions to shaping the landscape by the application of multi-temporal topographic surveys through SfM-photogrammetry and TLS. To this end we analyse a 7-year dataset of seasonal and annual high-resolution topographic surveys of a badlands landscape dominated by soil piping processes in Valpalmas (NE Spain). We examine the magnitude and distribution of geomorphic processes at multiple temporal scales and its relation with landform morphometric attributes and meteorological variables.

This research project was supported by the MANMOUNT (PID2019-105983RB-100/AEI/ 10.13039/501100011033) project funded by the MICINN-FEDER, the PRX21/00375 project funded by the Ministry of Universities of Spain from the “Salvador de Madariaga” programme, the Spanish Ministry of Science, Innovation and Universities (project EQC2018-004169-P) and by a grant from the Priority Research Area “Anthropocene” under the Strategic Programme Excellence Initiative at the Jagiellonian University. Manel Llena has a “Juan de la Cierva Formación” postdoctoral contract (FJC2020-043890-I/AEI/ 10.13039/501100011033) from the Spanish Ministry of Science and Innovation.

How to cite: Llena, M., Revuelto, J., Gómez-Gutiérrez, Á., López-Moreno, J. I., Errea, M. P., Alonso-González, E., Bernatek-Jakiel, A., and Nadal-Romero, E.: High-resolution topographic surveys as a quantitative method for a better understanding of soil piping processes in badlands landscapes: Valpalmas (NE Spain), EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-7057, https://doi.org/10.5194/egusphere-egu23-7057, 2023.

Updated crop type information is essential for agricultural monitoring and irrigation management applications. However, despite their importance, crop-type datasets, especially recent in-season data are still unavailable in most of developing countries where food security is a major concern. The lack of information on crop types is due to the limitations of traditional field surveys, which are generally infeasible, expensive, and inconsistent over larger spatial scales. Although remote sensing and machine learning approaches can provide cost-effective solutions for automatic crop-type mapping, huge amounts of high-quality training data are required in order to develop such applications. Therefore, semi-supervised or unsupervised learning methods become a potential solution to overcome the issue of lacking labeled training data. 

In this work, we explored semi-supervised and self-supervised techniques to map crop types in data-poor regions such as Saudi Arabia. The Self-supervised Transformer with Energy-based Graph Optimization (STEGO) method is a transformer-based segmentation technique that has the capability of both discovering and segmenting objects without the need for labeled data or human intervention. We evaluated the capability of STEGO to classify crop fields using Sentinel2 images. These images consisted of the derived maximum normalized difference vegetation index (NDVI), the standard deviation of NDVI, and the green chlorophyll vegetation index (GCVI).  The STEGO approach uses a novel contrastive loss that helps to distill pre-trained unsupervised visual features into semantic clusters, which is reported to outperform other unsupervised clustering methods. We set k=5, where k is the number of classes that reflects 4 crop types and a background class. 

Preliminary results of STEGO applied to small agricultural regions in Aljouf which is located in the north of Saudi Arabia captured the difference between crops when analyzed visually. Furthermore, assigning a crop-type label to each cluster class can be a challenging task. As for now, a brute force approach is followed to find the best assignment, and that is the assignment that provides the best results. As well as referring to previous knowledge of major crops grown in the region of interest, for this region, the major crops were wheat, olives, and tomato, in addition to “other” and “background” classes, which make up the 5 classes. Further work includes quantifying the accuracy of the clustering performance using the mean intersection over union metric (mIoU) and examining the effects of regional and national upscaling on the performance. 

How to cite: Alwahas, A., Johansen, K., and McCabe, M.: Crop Type Mapping Using Self-supervised Transformer with Energy-based Graph Optimization in Data-Poor Regions, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-7923, https://doi.org/10.5194/egusphere-egu23-7923, 2023.

Soil water content (SWC) estimation is a crucial issue of agricultural production, and its mapping is an important task. We aimed to study the efficacy of UAS-based thermal (TH) and multispectral (MS) cameras in SWC mapping.

The study area for the analysis is situated at NE-Hungary near the town of Tépe. the plot is a part of a larger area of intensive agriculture, where the arable crop at the year of surveys was maize. The experimental AOI was set to a maximum size of 200 x 200 meters due to the time-demand and limitations of the multi-sensor surveys. On the plot 3 major soil types are found meanwhile relative relief differences are also notable. Soil samples were collected at the time of surveys to measure the reference SWC rates in laboratory conditions using the gravimetric method.

The aerial mapping tasks were carried out using a DJI Matrice M210 payloads: 1) Micasense RedEdge-MX Dual, 2) Zenmuse XT2. High resolution DEM of the initial surface were mapped by a DJI Matrice M210 RTK v2 + a Zenmuse X7 lens. All imagery were processed in Pix4D Mapper. Machine Learning algorithms were then utilized to model the relationship between reflectance values, land surface temperature and the reference SWC values.

Our surveys were dedicated to a sensitivity analysis on the different settings of Pix4D regarding the downscaling to different pixel resolution of the multispectral data and spectral reflectance calibration too. We have analyzed the differences on the SWC modeling accuracies on the different soil types and relief conditions to develop a more robust estimation for precision drainage designs.

-----------

The research is supported by the NKFI K138079 project.

How to cite: Bertalan, L., Pataki, A., Nagy, L. A., Négyesi, G., and Szabó, S.: Multi-temporal UAS surveys for reconstructing soil water content of ploughland plots through multispectral and thermal infrared imagery, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-11235, https://doi.org/10.5194/egusphere-egu23-11235, 2023.

Climate change exacerbates food and water security, and it has been extended to hunger and malnutrition globally. In 2019, The Food and Agriculture Organization (FAO) announced North Korea was a food shortage country, and it has continued today. In South Korea, an agricultural drought has occurred irregularly since the 2010s by climate change and it impacts both crop quality and quantity negatively. Since these agricultural droughts and food shortages have increased gradually, in terms of climate change, agricultural risk must be addressed at the national level and long-term perspective. To address the agricultural risk quantitively in South and North Korea, rice productivity and Irrigation water demand are set as proxy variables of food and water variation in the agricultural domain. The overall methodology is divided into three-stage. 1) Classification of rice paddy area on the Korean Peninsula using remote sensing data based on machine learning algorithms, 2) Crop simulation using EPIC model on representative years and calibration/validation of the model, 3) multi-temporal crop simulation based on SSP scenario until the 2100s. The rice productivity and irrigation water demand matrix based on the scenario will be calculated and it will be used as a proxy variable in the water-food nexus. Consequently, the results will be extended to agricultural risk assessment on the national level-based Water-Food Nexus approach to address the Sustain Development Goals (SDGs) in Korean Peninsula

How to cite: Lee, S., Karthe, D., Lee, W.-K., and Lee, H.: Assessment of Long-Term and Spatio-temporal Changes between Rice Productivity and Irrigation Water Demand in South and North Korea based on Shared Socioeconomic Pathways (SSP) Scenario, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13314, https://doi.org/10.5194/egusphere-egu23-13314, 2023.

Strong interannual variability in precipitation amounts and distribution, as well as recurrent droughts, are cornerstones of African countries. These phenomena primarily impact rainfed crops, of which wheat is the most important and accounting for more than 80% of cultivated areas in Morocco. An early and consistent projection of pre-harvest grain production would help decision-makers anticipate management demands, detect yield gaps, and better understand wheat response to local climatic circumstances. How early a prediction is needed and the required depend on the nature of the stakeholder. In other words, early in-season forecasts are useful for producers so that they can adjust their inputs accordingly, whereas late-season forecasts are acceptable for other stakeholders, for example those interested in production monitoring.

In this work, we used satellite-derived phenology measures, climate, and soil data to generate in-season yield prediction models for rainfed and irrigated wheat in Morocco. The primary aims were to evaluate the predictive capabilities of the models as time progresses and the transferability of the models outside the area of their implementation. The findings demonstrated that the generated models' accuracy increases over time (i.e., when additional phenological measures are integrated into the models) and that Ensemble models and Random Forest models outperformed the conventional MLR models, including the regularised regression models (Lasso, Ridge, ElasticNet).

How to cite: Oulaid, B., Waine, T., Milne, A., El Alami, R., and Corstanje, R.: Combining remote sensing and data-driven models for early season wheat forecasts in Morocco: How transferable and early could a yield predictive model be?, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13599, https://doi.org/10.5194/egusphere-egu23-13599, 2023.