Search

Purpose: This research aims to develop a meta-machine learning model to optimize soil and nitrogen management for durum wheat in Italy. It addresses the challenges of increased food production on limited land amidst rising input costs, geopolitical changes, and climate change. The goal is to aid decision-makers in achieving maximum crop yield and income margins through effective agronomic strategies.The study developed a meta-machine learning model, integrating classification and regression models, and tested it at four sites in Marche and Basilicata, Italy, over several years. The model incorporated data from remote sensing, crop phenology, soil chemical properties, weather data, soil management, and nitrogen levels. A Random Forest model was used to classify crop phenology, while a Neural Network model predicted yield. Eleven nitrogen levels were compared across these sites.The Random Forest model achieved an accuracy of 0.98, kappa of 0.96, and recall of 0.98 for predicting crop phenology. The Neural Network model for yield prediction had an R squared of 0.90 and a Root Mean Square Error of 0.59 t ha-1. Key factors identified for model accuracy were temperature, precipitation, NDVI, and nitrogen input. Simulations of 30 soil management and fertilization combinations revealed that no-tillage management increased grain yield. The Marginal Fertilizer Yield Index determined optimal nitrogen application.The meta-machine learning model accurately predicted durum wheat yield and identified effective agronomic strategies, demonstrating the potential for broader application in field conditions. The model offers a promising approach to sustainable agriculture and climate change mitigation by utilising publicly available spatial datasets.Methods: This research aims to develop a meta-machine learning model to optimize soil and nitrogen management for durum wheat in Italy. It addresses the challenges of increased food production on limited land amidst rising input costs, geopolitical changes, and climate change. The goal is to aid decision-makers in achieving maximum crop yield and income margins through effective agronomic strategies.The study developed a meta-machine learning model, integrating classification and regression models, and tested it at four sites in Marche and Basilicata, Italy, over several years. The model incorporated data from remote sensing, crop phenology, soil chemical properties, weather data, soil management, and nitrogen levels. A Random Forest model was used to classify crop phenology, while a Neural Network model predicted yield. Eleven nitrogen levels were compared across these sites.The Random Forest model achieved an accuracy of 0.98, kappa of 0.96, and recall of 0.98 for predicting crop phenology. The Neural Network model for yield prediction had an R squared of 0.90 and a Root Mean Square Error of 0.59 t ha-1. Key factors identified for model accuracy were temperature, precipitation, NDVI, and nitrogen input. Simulations of 30 soil management and fertilization combinations revealed that no-tillage management increased grain yield. The Marginal Fertilizer Yield Index determined optimal nitrogen application.The meta-machine learning model accurately predicted durum wheat yield and identified effective agronomic strategies, demonstrating the potential for broader application in field conditions. The model offers a promising approach to sustainable agriculture and climate change mitigation by utilising publicly available spatial datasets.Results: This research aims to develop a meta-machine learning model to optimize soil and nitrogen management for durum wheat in Italy. It addresses the challenges of increased food production on limited land amidst rising input costs, geopolitical changes, and climate change. The goal is to aid decision-makers in achieving maximum crop yield and income margins through effective agronomic strategies.The study developed a meta-machine learning model, integrating classification and regression models, and tested it at four sites in Marche and Basilicata, Italy, over several years. The model incorporated data from remote sensing, crop phenology, soil chemical properties, weather data, soil management, and nitrogen levels. A Random Forest model was used to classify crop phenology, while a Neural Network model predicted yield. Eleven nitrogen levels were compared across these sites.The Random Forest model achieved an accuracy of 0.98, kappa of 0.96, and recall of 0.98 for predicting crop phenology. The Neural Network model for yield prediction had an R squared of 0.90 and a Root Mean Square Error of 0.59 t ha-1. Key factors identified for model accuracy were temperature, precipitation, NDVI, and nitrogen input. Simulations of 30 soil management and fertilization combinations revealed that no-tillage management increased grain yield. The Marginal Fertilizer Yield Index determined optimal nitrogen application.The meta-machine learning model accurately predicted durum wheat yield and identified effective agronomic strategies, demonstrating the potential for broader application in field conditions. The model offers a promising approach to sustainable agriculture and climate change mitigation by utilising publicly available spatial datasets.Conclusions: This research aims to develop a meta-machine learning model to optimize soil and nitrogen management for durum wheat in Italy. It addresses the challenges of increased food production on limited land amidst rising input costs, geopolitical changes, and climate change. The goal is to aid decision-makers in achieving maximum crop yield and income margins through effective agronomic strategies.The study developed a meta-machine learning model, integrating classification and regression models, and tested it at four sites in Marche and Basilicata, Italy, over several years. The model incorporated data from remote sensing, crop phenology, soil chemical properties, weather data, soil management, and nitrogen levels. A Random Forest model was used to classify crop phenology, while a Neural Network model predicted yield. Eleven nitrogen levels were compared across these sites.The Random Forest model achieved an accuracy of 0.98, kappa of 0.96, and recall of 0.98 for predicting crop phenology. The Neural Network model for yield prediction had an R squared of 0.90 and a Root Mean Square Error of 0.59 t ha-1. Key factors identified for model accuracy were temperature, precipitation, NDVI, and nitrogen input. Simulations of 30 soil management and fertilization combinations revealed that no-tillage management increased grain yield. The Marginal Fertilizer Yield Index determined optimal nitrogen application.The meta-machine learning model accurately predicted durum wheat yield and identified effective agronomic strategies, demonstrating the potential for broader application in field conditions. The model offers a promising approach to sustainable agriculture and climate change mitigation by utilising publicly available spatial datasets.Graphical abstract: This research aims to develop a meta-machine learning model to optimize soil and nitrogen management for durum wheat in Italy. It addresses the challenges of increased food production on limited land amidst rising input costs, geopolitical changes, and climate change. The goal is to aid decision-makers in achieving maximum crop yield and income margins through effective agronomic strategies.The study developed a meta-machine learning model, integrating classification and regression models, and tested it at four sites in Marche and Basilicata, Italy, over several years. The model incorporated data from remote sensing, crop phenology, soil chemical properties, weather data, soil management, and nitrogen levels. A Random Forest model was used to classify crop phenology, while a Neural Network model predicted yield. Eleven nitrogen levels were compared across these sites.The Random Forest model achieved an accuracy of 0.98, kappa of 0.96, and recall of 0.98 for predicting crop phenology. The Neural Network model for yield prediction had an R squared of 0.90 and a Root Mean Square Error of 0.59 t ha-1. Key factors identified for model accuracy were temperature, precipitation, NDVI, and nitrogen input. Simulations of 30 soil management and fertilization combinations revealed that no-tillage management increased grain yield. The Marginal Fertilizer Yield Index determined optimal nitrogen application.The meta-machine learning model accurately predicted durum wheat yield and identified effective agronomic strategies, demonstrating the potential for broader application in field conditions. The model offers a promising approach to sustainable agriculture and climate change mitigation by utilising publicly available spatial datasets. [ABSTRACT FROM AUTHOR]

Biofuels play a key role in the decarbonization process of the transport sector contributing to climate change mitigation, energy security and rural development. However, the competition between energy and food in the use of resources, such as land, water and energy input, requires a robust and integrated methodological framework to promote a holistic sustainable resource management. Here, we propose a land suitability model for energy crops to analyze the dynamic interrelationships between land, water and energy, with the aim of promoting the efficient use of resources in the bioenergy context. The model was developed using multi-criteria decision analysis (MCDA) techniques in geographic information systems (GIS) integrated with the Water-Energy-Food nexus approach. The model was tested for the Basilicata region (southern Italy), comparing two energy crops: (i) rapeseed (Brassica napus L.) and (ii) cardoon (Cynara cardunculus L.). The application to the case study shows the importance of the land suitability analysis to identify the most suitable crop in a given territory, while the nexus spatial analysis for locating the suitable areas where a more efficient use of water, energy and land resources is reached. The results revealed also that the use of irrigation water would allow greater use efficiency of the other resources, such as land and energy, an aspect not to be underestimated in the energy, land and agricultural planning processes. The research work contributes to the debate on the sustainability of bioenergy, so that they can represent a sustainable solution to the energy issue and climate change. [Display omitted] [ABSTRACT FROM AUTHOR]

This paper investigates the energy system of the Basilicata region (southern Italy) to highlight, through a scenario analysis, its possible future development in compliance with the European Commission's long-term vision. The IEA-ETSAP methodology was utilized to model and analyse the Basilicata energy system in a business- as -usual scenario, based on the assumptions of the 2010 Regional Energy and Environmental Policy Plan, and in two low-carbon scenarios, decarbonisation and high energy efficiency, which, in line with the long-term European strategy, foresee an 85% carbon dioxide emission reduction and a 20% energy efficiency increase in 2050. The results highlight that electricity production from renewable energy sources (in particular wind energy), strongly supported by the regional policy, goes beyond the Basilicata Regional Energy-Environmental Plan forecast, with a further increase in the decarbonisation scenarios. In addition, solar thermal and highly efficient technologies are adopted in commercial and residential end-use sectors. An additional 9% reduction in energy consumption is achieved in the high energy efficiency scenario by passive houses. The proposed approach shows the usefulness of a modelling framework typically used for national and supranational analyses to support regional authorities in their decision-making process to obtain results in line with the European and national strategies. Application to the Basilicata region shows the usefulness of a consolidated framework for policy assessment at a local scale to assess regional contribution to the achievement of national climate mitigation targets that tend towards climate neutrality. • The role of regional authorities is essential to reach national decarbonisation goals. • TIMES can be advantageously used for regional energy system modelling and analysis. • Scenario analysis can be a powerful tool for policy evaluation. • A cap on carbon dioxide emissions leads to a 28% increase of electricity production. • Passive houses in Residential allow a further 5% reduction of energy consumption. [ABSTRACT FROM AUTHOR]

A reliable energy supply is fundamental to ensure energy security and support the mitigation of climate change by promoting the use of renewable sources and reducing carbon emissions. Energy system analysis provides a sound methodology to assess energy needs, allowing to investigate the energy system behavior and to individuate the optimal energy-technology configurations for the achievement of strategic energy and environmental policy targets. In this framework, the estimation of future trends of exogenous variables such as energy demand has a fundamental importance to obtain reliable and effective solutions, contributing remarkably to the accuracy of models' input data. This study illustrates an application of regression analysis to predict energy demand trends in end use sectors. The proposed procedure is applied to characterize statistically the relationships between population and gross domestic product (independent variables) and energy demands of Residential, Transport and Commercial in order to determine the energy demand trends over a long-term horizon. The effectiveness of linear and nonlinear regression models for energy demand forecasting has been validated by classical statistical tests. Energy demand projections have been tested as input data of the bottom-up TIMES model in two applications (the TIMES-Basilicata and TIMES-Italy models) confirming the validity of the forecasting approach. • Energy systems modeling requires a detailed description of energy demand development (85). • Relationships among energy demand, population and economic variables are emphasized (83). • Regression analysis is applied to end-use demand forecasts as input for TIMES models (84). • The efficacy of regression analysis is validated by classical statistical tests (79). • Energy trends are tested with two energy models: TIMES-Basilicata and TIMES-Italy (81). [ABSTRACT FROM AUTHOR]

Global forecasts for the growth of urban agglomerations and the expected effects of ongoing climate changes highlight the need to reflect on urban and regional planning issues.The real challenge is to combine urban growth with adequate density and compactness to prevent and contain land use change, loss of productive land and reduction of ecosystem services.The city is therefore observed because its compact or dispersed form and its regulation through appropriate planning tools can make an important contribution both in terms of mitigation and adaptation to climate change, and in limiting future impacts.The present work aims to provide a useful methodology to investigate the relationship between the increase of urbanized areas and their growth dynamics, and the surface temperatures variation recorded within urban, peri-urban and rural contexts.The building environment evolution and the Land Surface Temperatures (LST) time series surveyed by Modis satellites are jointly analyzed in the context of the Basilicata region. The results show a strong relationship between the increase in recorded minimum temperatures and urban areas expansion, especially where the main growth dynamics is compaction. [ABSTRACT FROM AUTHOR]