{"type": "FeatureCollection", "features": [{"id": "10.1016/j.scitotenv.2021.152880", "type": "Feature", "geometry": null, "properties": {"updated": "2026-05-25T16:16:43Z", "type": "Journal Article", "created": "2022-01-06", "title": "Exploring the potential role of environmental and multi-source satellite data in crop yield prediction across Northeast China", "description": "Open AccessLe d\u00e9veloppement d'un syst\u00e8me pr\u00e9cis de pr\u00e9diction du rendement des cultures \u00e0 grande \u00e9chelle est d'une importance primordiale pour la gestion des ressources agricoles et la s\u00e9curit\u00e9 alimentaire mondiale. L'observation de la Terre fournit une source unique d'informations pour surveiller les cultures \u00e0 partir d'une diversit\u00e9 de gammes spectrales. Cependant, l'utilisation int\u00e9gr\u00e9e de ces donn\u00e9es et de leurs valeurs dans la pr\u00e9diction du rendement des cultures est encore peu \u00e9tudi\u00e9e. Ici, nous avons propos\u00e9 la combinaison de donn\u00e9es environnementales (climat, sol, g\u00e9ographie et topographie) avec de multiples donn\u00e9es satellitaires (indices de v\u00e9g\u00e9tation optiques, fluorescence induite par le soleil (SIF), temp\u00e9rature de surface du sol (LST) et profondeur optique de la v\u00e9g\u00e9tation micro-ondes (VOD)) dans le cadre pour estimer le rendement des cultures de ma\u00efs, de riz et de soja dans le nord-est de la Chine, et leur valeur unique et leur influence relative sur la pr\u00e9diction du rendement ont \u00e9t\u00e9 \u00e9valu\u00e9es. Deux m\u00e9thodes de r\u00e9gression lin\u00e9aire, trois m\u00e9thodes d'apprentissage automatique (ML) et un mod\u00e8le d'ensemble ML ont \u00e9t\u00e9 adopt\u00e9s pour construire des mod\u00e8les de pr\u00e9diction de rendement. Les r\u00e9sultats ont montr\u00e9 que les m\u00e9thodes individuelles de ML surpassaient les m\u00e9thodes de r\u00e9gression lin\u00e9aire, le mod\u00e8le d'ensemble de ML a encore am\u00e9lior\u00e9 les mod\u00e8les de ML uniques. De plus, les mod\u00e8les avec plus d'intrants ont obtenu de meilleures performances, la combinaison de donn\u00e9es satellitaires avec des donn\u00e9es environnementales, qui expliquaient respectivement 72\u00a0%, 69\u00a0% et 57\u00a0% de la variabilit\u00e9 du rendement du ma\u00efs, du riz et du soja, a d\u00e9montr\u00e9 des performances de pr\u00e9diction du rendement sup\u00e9rieures \u00e0 celles des intrants individuels. Alors que les donn\u00e9es satellitaires ont contribu\u00e9 \u00e0 la pr\u00e9diction du rendement des cultures principalement au d\u00e9but de la pointe de la saison de croissance, les donn\u00e9es climatiques ont fourni des informations suppl\u00e9mentaires principalement \u00e0 la pointe de la fin de la saison. Nous avons \u00e9galement constat\u00e9 que l'utilisation combin\u00e9e de l'IVE, du LST et du SIF a am\u00e9lior\u00e9 la pr\u00e9cision du mod\u00e8le par rapport au mod\u00e8le d'IVE de r\u00e9f\u00e9rence. Cependant, les indices de v\u00e9g\u00e9tation bas\u00e9s sur l'optique partageaient des informations similaires et ne fournissaient pas beaucoup d'informations suppl\u00e9mentaires au-del\u00e0 de l'IVE. Les pr\u00e9visions de rendement en cours de saison ont montr\u00e9 que les rendements des cultures peuvent \u00eatre pr\u00e9vus de mani\u00e8re satisfaisante deux \u00e0 trois mois avant la r\u00e9colte. La g\u00e9ographie, la topographie, la VOD, l'IVE, les param\u00e8tres hydrauliques du sol et les param\u00e8tres nutritifs sont plus importants pour la pr\u00e9diction du rendement des cultures.", "keywords": ["Atmospheric sciences", "Climate", "Multi-source satellite data", "Normalized Difference Vegetation Index", "Engineering", "Pathology", "Climate change", "Urban Heat Islands and Mitigation Strategies", "Linear regression", "2. Zero hunger", "Global and Planetary Change", "Vegetation Monitoring", "Ecology", "Geography", "Statistics", "Agriculture", "Geology", "Remote Sensing in Vegetation Monitoring and Phenology", "04 agricultural and veterinary sciences", "Remote sensing", "Aerospace engineering", "Archaeology", "Physical Sciences", "Metallurgy", "Medicine", "Seasons", "Global Vegetation Models", "Biomass Estimation", "Regression analysis", "Vegetation (pathology)", "Crops", " Agricultural", "Environmental Engineering", "Environmental data", "Yield (engineering)", "Zea mays", "Environmental science", "Machine learning", "FOS: Mathematics", "Crop yield", "Biology", "Global Forest Drought Response and Climate Change", "FOS: Environmental engineering", "Predictive modelling", "Food security", "FOS: Earth and related environmental sciences", "15. Life on land", "Agronomy", "Materials science", "Yield prediction", "Satellite", "13. Climate action", "FOS: Biological sciences", "Environmental Science", "Growing season", "0401 agriculture", " forestry", " and fisheries", "Mathematics"], "contacts": [{"organization": "Zhenwang Li, Lei Ding, Donghui Xu,", "roles": ["creator"]}]}, "links": [{"href": "https://doi.org/10.1016/j.scitotenv.2021.152880"}, {"rel": "related", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/Science%20of%20The%20Total%20Environment", "name": "related record", "description": "related record", "type": "application/json"}, {"rel": "self", "type": "application/geo+json", "title": "10.1016/j.scitotenv.2021.152880", "name": "item", "description": "10.1016/j.scitotenv.2021.152880", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main/items/10.1016/j.scitotenv.2021.152880"}, {"rel": "collection", "type": "application/json", "title": "Collection", "name": "collection", "description": "Collection", "href": "https://repository.soilwise-he.eu/cat/collections/metadata:main"}], "time": {"date": "2022-04-01T00:00:00Z"}}, {"id": "10.1038/s42949-024-00154-z", "type": "Feature", "geometry": null, "properties": {"updated": "2026-05-25T16:17:43Z", "type": "Journal Article", "created": "2024-03-16", "title": "Urban greenspaces and nearby natural areas support similar levels of soil ecosystem services", "description": "Abstract

Greenspaces are important for sustaining healthy urban environments and their human populations. Yet their capacity to support multiple ecosystem services simultaneously (multiservices) compared with nearby natural ecosystems remains virtually unknown. We conducted a global field survey in 56 urban areas to investigate the influence of urban greenspaces on 23 soil and plant attributes and compared them with nearby natural environments. We show that, in general, urban greenspaces and nearby natural areas support similar levels of soil multiservices, with only six of 23 attributes (available phosphorus, water holding capacity, water respiration, plant cover, arbuscular mycorrhizal fungi (AMF), and arachnid richness) significantly greater in greenspaces, and one (available ammonium) greater in natural areas. Further analyses showed that, although natural areas and urban greenspaces delivered a similar number of services at low (>25% threshold) and moderate (>50%) levels of functioning, natural systems supported significantly more functions at high (>75%) levels of functioning. Management practices (mowing) played an important role in explaining urban ecosystem services, but there were no effects of fertilisation or irrigation. Some services declined with increasing site size, for both greenspaces and natural areas. Our work highlights the fact that urban greenspaces are more similar to natural environments than previously reported and underscores the importance of managing urban greenspaces not only for their social and recreational values, but for supporting multiple ecosystem services on which soils and human well-being depends.Greenspaces are important for sustaining healthy urban environments and their human populations. Yet their capacity to support multiple ecosystem services simultaneously (multiservices) compared with nearby natural ecosystems remains virtually unknown. We conducted a global field survey in 56 urban areas to investigate the influence of urban greenspaces on 23 soil and plant attributes and compared them with nearby natural environments. We show that, in general, urban greenspaces and nearby natural areas support similar levels of soil multiservices, with only six of 23 attributes (available phosphorus, water holding capacity, water respiration, plant cover, arbuscular mycorrhizal fungi (AMF), and arachnid richness) significantly greater in greenspaces, and one (available ammonium) greater in natural areas. Further analyses showed that, although natural areas and urban greenspaces delivered a similar number of services at low (>25% threshold) and moderate (>50%) levels of functioning, natural systems supported significantly more functions at high (>75%) levels of functioning. Management practices (mowing) played an important role in explaining urban ecosystem services, but there were no effects of fertilisation or irrigation. Some services declined with increasing site size, for both greenspaces and natural areas. Our work highlights the fact that urban greenspaces are more similar to natural environments than previously reported and underscores the importance of managing urban greenspaces not only for their social and recreational values, but for supporting multiple ecosystem services on which soils and human well-being depends.