A Holistic Irrigation Advisory Policy Scheme by the Hellenic Agricultural Organization: An Example of a Successful Implementation in Crete, Greece DOI Open Access
Nektarios N. Kourgialas

Water, Journal Year: 2024, Volume and Issue: 16(19), P. 2769 - 2769

Published: Sept. 28, 2024

The aim of this communication article is to present a successful irrigation advisory scheme on the island Crete (Greece) provided by Hellenic Agricultural Organization (ELGO DIMITRA), which well adapted different needs farmers and water management agencies. motivation create stems from need save resources while ensuring optimal production in region like where droughts seem occur more frequently recent years. This scheme/approach has three levels implementation (components) depending spatial level end-users’ needs. first concerns weekly bulletins main agricultural areas with informing local managers about crop second an innovative digital web-based platform for precise determination Crete’s crops at parcel as adaptation strategies context climate change. In platform, important features such real-time meteorological information, data cultivation type parcels, validated algorithms calculating needs, accurate soil texture map derived satellite images, appropriate agronomic practices conserve based geomorphology farm are considered. third proposed approach includes open-source Internet Things (IoT) intelligent system individual scheduling. IoT moisture atmospheric sensors installed field, corresponding laboratory hydraulic characterization service. third-level provides specialized information automated optimization use. All above approaches have been implemented evaluated end-users very high degree satisfaction terms effectiveness usability.

Language: Английский

Sustainable Fertilization of Organic Sweet Cherry to Improve Physiology, Quality, Yield, and Soil Properties DOI Creative Commons
Liliana Gaeta, Luigi Tarricone, Alessandro Persiani

et al.

Agronomy, Journal Year: 2025, Volume and Issue: 15(1), P. 135 - 135

Published: Jan. 8, 2025

Sustainable fertilization techniques are essential in Mediterranean farming systems, where the depletion of organic matter, influencing soil water and nutrient availability, is becoming an increasing concern. In this context, fertilizers offer effective strategy to restore fertility while reducing environmental impacts. This research aimed evaluate effects different on quality tree performance a sweet cherry (Prunus avium L.) orchard. study was conducted two growing seasons (2021–2022) orchard Southern Italy, comparing four treatments: (i) compost, (ii) compost combined with tea, (iii) mixed manure, (iv) unfertilized control. The results indicated that applied both as foliar spray, significantly improved status, particularly under stress conditions, reflected by more negative stem potential values. Moreover, treatment enhanced photosynthetic performance, yield, fruit quality, achieving highest ratio soluble solids content/total acidity. findings suggest combination could be sustainable valuable option for orchards. However, further studies necessary understand benefits other orchards well long-term soils.

Language: Английский

Citations

1
Prediction of Winter Wheat Parameters with Planet SuperDove Imagery and Explainable Artificial Intelligence DOI Creative Commons
Gabriele De Carolis, Vincenzo Giannico, Leonardo Costanza

et al.

Agronomy, Journal Year: 2025, Volume and Issue: 15(1), P. 241 - 241

Published: Jan. 19, 2025

This study investigated the application of high-resolution satellite imagery from SuperDove satellites combined with machine learning algorithms to estimate spatiotemporal variability some winter wheat parameters, including relative leaf chlorophyll content (RCC), water (RWC), and aboveground dry matter (DM). The research was carried out within an experimental field in Southern Italy during 2024 growing season. Different (ML) were trained compared using spectral band data calculated vegetation indices (VIs) as predictors. Model performance assessed R2 RMSE. ML models tested random forest (RF), support vector regressor (SVR), extreme gradient boosting (XGB). RF outperformed other prediction RCC when VIs predictors (R2 = 0.81) RWC DM bands 0.71 0.87, respectively). explainability SHAP method. A analysis highlighted that GNDVI, Cl1, NDRE most important for predicting RCC, while yellow red prediction, nir prediction. best model found each target used its seasonal trend produce a map. approach highlights potential integrating remote monitoring wheat, which can sustainable farming practices.

Language: Английский

Citations

1
High-Resolution Mapping of Maize in Mountainous Terrain Using Machine Learning and Multi-Source Remote Sensing Data DOI Creative Commons
Luying Liu, Jingyi Yang, F Yin

et al.

Land, Journal Year: 2025, Volume and Issue: 14(2), P. 299 - 299

Published: Jan. 31, 2025

In recent years, machine learning methods have garnered significant attention in the field of crop recognition, playing a crucial role obtaining spatial distribution information and understanding dynamic changes planting areas. However, research smaller plots within mountainous regions remains relatively limited. This study focuses on Shangzhou District Shangluo City, Shaanxi Province, utilizing dataset high-resolution remote sensing images (GF-1, ZY1-02D, ZY-3) collected over seven months 2021 to calculate normalized difference vegetation index (NDVI) construct time series. By integrating survey results with series Google Earth for visual interpretation, NDVI curve maize was analyzed. The Random Forest (RF) classification algorithm employed comparative analyses accuracy were conducted using Support Vector Machine (SVM), Gaussian Naive Bayes (GNB), Artificial Neural Network (ANN). demonstrate that random forest achieved highest accuracy, an overall 94.88% Kappa coefficient 0.94, both surpassing those other yielding satisfactory results. confirms feasibility precise extraction southern China, providing valuable scientific support optimizing land resource use enhancing agricultural productivity.

Language: Английский

Citations

0
Predicting Olive Tree Chlorophyll Fluorescence Using Explainable AI with Sentinel-2 Imagery in Mediterranean Environment DOI Creative Commons
Leonardo Costanza, Beatriz Lorente, Francisco Pedrero Salcedo

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(5), P. 2746 - 2746

Published: March 4, 2025

Chlorophyll fluorescence is a useful indicator of plant’s physiological status, particularly under stress conditions. Remote sensing an increasingly adopted technology in modern agriculture, allowing the acquisition crop information (e.g., chlorophyll fluorescence) without direct contact, reducing fieldwork. The objective this study to improve monitoring olive tree (Fv′/Fm′) via remote Mediterranean environment, where frequency factors, such as drought, increasing. An advanced approach combining explainable artificial intelligence and multispectral Sentinel-2 satellite data was developed predict fluorescence. Field measurements were conducted southeastern Italy on two groves: one irrigated other rainfed reflectance bands vegetation indices used predictors different machine learning algorithms tested compared. Random Forest showed highest predictive accuracy, when predictors. Using spectral preserves more per observation, enabling models detect variations that VIs might miss. Additionally, raw minimizes potential bias could arise from selecting specific indices. SHapley Additive exPlanations (SHAP) analysis performed explain model. using Key regions associated with Fv′/Fm′, red-edge NIR, identified. results highlight integrating grove management, providing tool for early detection targeted interventions.

Language: Английский

Citations

0
Coupling Different Machine Learning and Meta-Heuristic Optimization Techniques to Generate the Snow Avalanche Susceptibility Map in the French Alps DOI Open Access
Enes Can Kayhan, Ömer Ekmekcioğlu

Water, Journal Year: 2024, Volume and Issue: 16(22), P. 3247 - 3247

Published: Nov. 12, 2024

The focus of this study is to introduce a hybrid predictive framework encompassing different meta-heuristic optimization and machine learning techniques identify the regions susceptible snow avalanches. To accomplish aim, present research sought acquire best-performed model among nine scenarios three meta-heuristics, namely particle swarm (PSO), gravitational search algorithm (GSA), Cuckoo Search (CS), ML approaches, i.e., support vector classification (SVC), stochastic gradient boosting (SGB), k-nearest neighbors (KNN), pertaining families. According diligent analysis performed with regard blinded testing set, PSO-SGB illustrated most satisfactory performance an accuracy 0.815, while precision recall were found be 0.824 0.821, respectively. F1-score predictions was area under receiver operating curve (AUC) obtained 0.9. Despite attaining similar success via CS-SGB model, time-efficiency underscored PSO-SGB, as corresponding process consumed considerably less computational time compared its counterpart. SHapley Additive exPlanations (SHAP) implementation further informed that slope, elevation, wind speed are contributing attributes detecting avalanche susceptibility in French Alps.

Language: Английский

Citations

1
Temporal Vine Water Status Modeling Through Machine Learning Ensemble Technique and Sentinel-2 Multispectral Images Under Semi-Arid Conditions DOI Creative Commons
Vincenzo Giannico, Simone Pietro Garofalo, Luca Brillante

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(24), P. 4784 - 4784

Published: Dec. 22, 2024

New challenges will be experienced by the agriculture sector in near future, especially due to effects of climate change. For example, rising temperatures could result increased evapotranspiration demand, causing difficulties management irrigation practices. Generally, an important predictor plant water status taken into account for monitoring and is stem potential. However, it requires a huge amount time-consuming fieldwork, particularly when adequate data necessary fully investigate spatial temporal variability large areas under monitoring. In this study, integration machine learning satellite remote sensing (Sentinel-2) was investigated obtain model able predict potential viticulture using multispectral imagery. Vine were acquired within Montepulciano vineyard south Italy (Puglia region), semi-arid conditions; over two years during seasons. Different algorithms (lasso, ridge, elastic net, random forest) compared vegetation indices spectral bands as predictors independent analyses. The results show that possible remotely estimate vine with forest from (R2 = 0.72). Integrating techniques help farmers technicians manage plan irrigation, avoiding or reducing fieldwork.

Language: Английский

Citations

1
A Holistic Irrigation Advisory Policy Scheme by the Hellenic Agricultural Organization: An Example of a Successful Implementation in Crete, Greece DOI Open Access
Nektarios N. Kourgialas

Water, Journal Year: 2024, Volume and Issue: 16(19), P. 2769 - 2769

Published: Sept. 28, 2024

The aim of this communication article is to present a successful irrigation advisory scheme on the island Crete (Greece) provided by Hellenic Agricultural Organization (ELGO DIMITRA), which well adapted different needs farmers and water management agencies. motivation create stems from need save resources while ensuring optimal production in region like where droughts seem occur more frequently recent years. This scheme/approach has three levels implementation (components) depending spatial level end-users’ needs. first concerns weekly bulletins main agricultural areas with informing local managers about crop second an innovative digital web-based platform for precise determination Crete’s crops at parcel as adaptation strategies context climate change. In platform, important features such real-time meteorological information, data cultivation type parcels, validated algorithms calculating needs, accurate soil texture map derived satellite images, appropriate agronomic practices conserve based geomorphology farm are considered. third proposed approach includes open-source Internet Things (IoT) intelligent system individual scheduling. IoT moisture atmospheric sensors installed field, corresponding laboratory hydraulic characterization service. third-level provides specialized information automated optimization use. All above approaches have been implemented evaluated end-users very high degree satisfaction terms effectiveness usability.

Language: Английский

Citations

0