Loading...
Loading...
Crop yield, soil data, pest surveillance, livestock, food composition, precision farming
19,247 datasets
426-band hyperspectral files, along with RGB imagery, LiDAR tiles, and canopy height data, form the training data for the NeonTreeEvaluation benchmark. This dataset provides co-registered, multi-sensor data for individual tree detection across multiple National Ecological Observatory Network (NEON) field sites. Annotations for the image tiles are available in a separate version-controlled repository.
NeonTreeEvaluation Benchmark provides multi-modal airborne remote sensing data for developing and testing individual tree detection algorithms. The dataset includes RGB images, LiDAR tiles, 426-band hyperspectral files, and 1-meter canopy height models for multiple geographic sites identified by NEON codes. Annotations for tree locations are provided separately, and the evaluation set includes unannotated data for testing or semi-supervised learning.
NSW Department of Climate Change, Energy, the Environment and Water produced this land use dataset capturing how the New South Wales landscape was used for food production, forestry, nature conservation, infrastructure, and urban development in September 2017. Version 1.5, published in December 2023, incorporates fine-scale mapping for regions like the Central Coast and Illawarra, enhanced horticulture data, and an updated road network. The mapping is based on aerial and satellite imagery, ancillary datasets like cadastre and land zoning, and was captured using GIS software at a scale of 1:8,000 or better.
10-meter flux tower measurements from a mulga woodland in central Australia, processed to provide gap-filled data on ecosystem carbon, water, and energy exchange. The site, established in 2012 and managed by the University of Technology Sydney, monitors a mosaic of grassy mulga woodland and Corymbia/Triodia savanna on Pine Hill Station. Data processing used PyFluxPro (v3.5.0) to produce final Net Ecosystem Exchange (NEE) partitioned into Gross Primary Productivity (GPP) and Ecosystem Respiration (ER).
A Swiss field experiment established in August 2013 on 168 plots (3 × 5 m) with six grass species in monoculture. The dataset likely contains measurements from control and seasonal drought treatments (spring, summer, fall) applied for ten weeks using rainout shelters, with hourly sensor data for soil water potential, humidity, and temperature. Claudia Hahn from the University of Basel conducted the research, which received standardized fertilization and watering interventions.
A cross-sectional study analyzes the micronutrient content of diets based on data from 32,898 individuals in the 2008-2009 Brazilian Household Budget Survey. It uses two 24-hour food records per person to assess the association between ultra-processed food consumption and micronutrient intake. The research was authored by Maria Laura da Costa Louzada and highlights dietary impacts aligned with Brazilian dietary guidelines.
An experiment evaluating Saccharomyces cerevisiae yeast at levels 0, 0.2, 0.4, and 0.6 g/L on the in vitro digestibility of dry matter, crude protein, and neutral detergent fiber in diets containing 100, 75, 50, and 25% Cynodon dactylon grass. The data were collected using a two-stage artificial rumen incubation technique and analyzed by variance and regression. The work was authored by Fábio José Ferreira Figueiroa.
16 folders contain raw event camera data for autofocus evaluation. The dataset includes aedat4 files with event streams and synchronized APS frames, ground truth images, and results from PBF and EGS algorithms. It was created by Yuhan Bao and shared on paperswithcode.
Oneida and Cazenovia Lakes in New York State have ice cover records spanning from 1826 to 2009. The data were collected by personnel from the Cornell Biological Field Station, part of Cornell University's Department of Natural Resources. This dataset is part of a broader 50-year aquatic research program focused on Oneida Lake.
ELSA-Brasil baseline data was used for a cross-sectional study investigating the association between fat and fiber intakes and the hypertriglyceridemic waist phenotype. The dataset likely contains anthropometric measurements, dietary intake assessments, and socioeconomic, demographic, and behavioral variables collected via questionnaires. The study found no association between fat/fiber intake and the phenotype, but identified associations with factors like physical activity, smoking, income, and obesity.
Terrestrial laser scanning (TLS) data from a UK forest plot collected in April 2012 by Eric Casella of Forest Research. The data was used to automatically reconstruct individual trees as cylinder models, a process detailed in an open-access journal article. The animation demonstrating this reconstruction was produced by the Inverse Problems research group at Tampere University of Technology.
A geospatial dataset mapping Tree Preservation Orders (TPOs) within the Rochdale Borough Council area in England. TPOs are legally enforceable orders protecting specific trees, groups, or woodlands, prohibiting activities like cutting down without written consent from the local planning authority. The data is maintained and updated by the council's planning department and is provided under a CC-BY-4.0 license.
Modelled annual average percentage yield loss due to ozone damage for four global staple crops, 2010-2012 version 2 provides modelled global yield loss percentages for maize, rice, soybean, and wheat. The data, produced by the Environmental Information Data Centre for the NERC-funded SUNRISE project, is based on FAO GAEZ production data and EMEP MSC-W chemical transport model outputs. It highlights the impact of ground-level ozone on agriculture and identifies high-risk areas for mitigation.
A review of environmental factors document for a proposed road upgrade project in New South Wales, Australia. The NSW Government produced this document, which is scheduled for an update in June 2026. It concerns the upgrade of Nelson Bay Road (Main Road 108) between Bobs Farm and Anna Bay, a major arterial route to the Tomaree Peninsula.
Forestry Commission data tracks the percentage of woodland in England under active management, a key corporate performance indicator. The dataset defines 'actively managed' woodland using ten administrative criteria, including grant schemes, felling licenses, and public estate land. It supports reporting for specific fiscal years, with sources referencing the 2011/12, 2018/19, and 2020/21 indicator cycles.
DAFF and SELTMP compiled this 2013 dataset on Queensland's commercial harvest, line, net, pot, and trawl fisheries. It contains 20 geospatial layers for active licenses, effort days, harvest weight, and gross value of product (GVP) aggregated on a 30-minute grid. The data originates from mandatory daily logbooks completed by commercial fishers and is managed by the Department of Agriculture, Fisheries and Forestry.
1.2 MB of data from a laboratory and artificial flower experiment investigating the sub-lethal effects of pesticides on solitary bees. The dataset, authored by Janine M. Schwarz and last updated in June 2026, contains results showing sulfoxaflor reduced sugar solution intake by an average of 23.8% and impaired learning efficiency.
Two global raster maps of tree density, one modeled at the biome scale and the other at the ecoregion scale, produced by Thomas W. Crowther and colleagues. The maps were generated using linear regression models based on over 420,000 ground-sourced estimates and vegetative, climatic, topographic, and anthropogenic variables. They were initially published in 2015 and the data was explicitly published in 2016.
A 2004 census of topsoil macroinvertebrates in spruce-fir forests along a pollution gradient from the Middle Ural Copper Smelter. The dataset consists of 340 sampling events across two soil layers, recording 64,658 occurrences and 8,430 collected individuals across multiple taxa. It was created by Е. Л. Воробейчик of the Institute of Plant and Animal Ecology for ecotoxicological monitoring of forest ecosystems.
Soil macrofauna communities were sampled in September 2016 at the Canguiri Farm in Pinhais, Paraná, Brazil. The data likely contains abundance counts for 42 taxa across four different land use systems, each with three replicates. The study was conducted by Talita Ferreira of the Federal University of Paraná using the Tropical Soil Biology and Fertility Programme (TSBF) methodology.