A5074170911- Remote Sensing in Agriculture
- Land Use and Ecosystem Services
- Remote Sensing and LiDAR Applications
- Remote Sensing and Land Use
- Soil Geostatistics and Mapping
- Remote-Sensing Image Classification
- Impact of Light on Environment and Health
- Agricultural risk and resilience
- Species Distribution and Climate Change
- Fire effects on ecosystems
- Opportunistic and Delay-Tolerant Networks
- Mobile Ad Hoc Networks
- Conservation, Biodiversity, and Resource Management
- Forest Management and Policy
- Climate change impacts on agriculture
- Agriculture and Biological Studies
- Forest ecology and management
- Soil and Land Suitability Analysis
- Atmospheric and Environmental Gas Dynamics
- Geographic Information Systems Studies
- Urban Transport and Accessibility
- Vehicular Ad Hoc Networks (VANETs)
- Sustainable Agricultural Systems Analysis
- Agriculture Sustainability and Environmental Impact
- Flood Risk Assessment and Management
Joint Research Centre
2011-2024
Universitat Politècnica de Catalunya
2011
Joint Research Centre
2006
A recently completed research program (TREES) employing the global imaging capabilities of Earth-observing satellites provides updated information on status world's humid tropical forest cover. Between 1990 and 1997, 5.8 ± 1.4 million hectares were lost each year, with a further 2.3 0.7 visibly degraded. These figures indicate that net rate change in cover for tropics is 23% lower than generally accepted rate. This result affects calculation carbon fluxes budget means terrestrial sink...
Accurate characterization of tropical moist forest changes is needed to support conservation policies and quantify their contribution global carbon fluxes more effectively. We document, at pantropical scale, the extent (degradation, deforestation, recovery) these forests over past three decades. estimate that 17% have disappeared since 1990 with a remaining area 1071 million hectares in 2019, from which 10% are degraded. Our study underlines importance degradation process ecosystems,...
The Joint Research Centre of the European Commission (JRC), in partnership with 30 institutions, has produced a global land cover map for year 2000, GLC 2000 map. validation GLC2000 product now been completed. accuracy assessment relied on two methods: confidence-building method (quality control based comparison ancillary data) and quantitative stratified random sampling reference data. sample site stratification used an underlying grid Landsat data was proportion priority classes landscape...
Accurate and reliable information on the spatial distribution of major crops is needed for detecting possible production deficits with aim preventing food security crises anticipating response planning. In this paper, we compared some most widely used global land cover datasets to examine their comparative advantages cropland monitoring. Cropland class areas are following datasets: FAO-GLCshare (FAO Global Land Cover Network), Geowiki IIASA-Hybrid (Hybrid map from International Institute...
Population figures are usually collected by national statistical institutes at small enumeration units (e.g. census tracts or building units). However, still for many countries in Europe, data distributed coarser geographical like municipalities. This level of resolution is insufficient analysis fields. In addition, the heterogeneity size causes great distortions analysis, i.e. Modifiable Areal Unit Problem (MAUP). Dasymetric mapping techniques have long been applied world-wide to derive...
Abstract A global reference data set on cropland was collected through a crowdsourcing campaign using the Geo-Wiki tool. The lasted three weeks, with over 80 participants from around world reviewing almost 36,000 sample units, focussing identification. For quality assessment purposes, two additional sets are provided. first is control of 1,793 locations validated by students trained in satellite image interpretation. This used to assess crowd as progressed. second contains 60 expert...
This paper outlines the methods and results for monitoring forest change resulting carbon emissions 1990–2000 200–2005 periods carried out over tropical Central South America. To produce our estimates we used a systematic sample of medium resolution satellite data processed to maps covering 1230 sites 20 km by km, each located at degree confluence. Biomass were spatially associated individual site so that annual could be estimated. For study area estimate cover in had fallen from 763 Mha...
This study presents an approach to forest canopy disturbance monitoring in evergreen forests continental Southeast Asia, based on temporal differences of a modified normalized burn ratio (NBR) vegetation index. We generate NBR values from each available Landsat 8 scene given period. A step ‘self-referencing’ normalizes the values, largely eliminating illumination/topography effects, thus maximizing inter-comparability. then create yearly composites these self-referenced (rNBR) selecting per...
Abstract Accurately characterizing land surface changes with Earth Observation requires geo-located ground truth. In the European Union (EU), a tri-annual surveyed sample of cover and use has been collected since 2006 under Land Use/Cover Area frame Survey (LUCAS). A total 1351293 observations at 651780 unique locations for 106 variables along 5.4 million photos were during five LUCAS surveys. Until now, these data have never harmonised into one database, limiting full exploitation...
A problem of crop plants classification for three regions Ukraine with an area 78.500 km2 is considered. Classification carried out using not a single satellite but time series images. The used data are characterized by different spatial resolution and temporal characteristics. By example this we assessed the efficiency algorithms (neural networks, decision trees support vector machines) substantially levels (of training testing samples) both extremely large sets under condition their lack (absence).
In this paper, we propose a new approach to pixel and parcel-based classification of multi-temporal optical satellite imagery. We first restore missing data due clouds shadows based on vector raster fusion in different phases methodology. Pixel-based maps are derived from an ensemble neural networks, particular multilayer perceptrons (MLPs). The proposed is applied for regional scale crop using Landsat-8 images the JECAM site Kyivska oblast Ukraine 2013. obtained results area estimates also...
Abstract Several global high-resolution built-up surface products have emerged over the last five years, taking full advantage of open sources satellite data such as Landsat and Sentinel. However, these sets require validation that is independent producers products. To fill this gap, we designed a sample set 50 K locations using stratified sampling approach any existing We launched crowdsourcing campaign Geo-Wiki ( https://www.geo-wiki.org/ ) to visually interpret for surfaces very images...
This work presents a design-based validation and calibration scheme for the Global Burned Area 2000 (GBA2000) products. The objective of such is to assess margins uncertainty associated with burned area products estimate coefficients needed convert pixel counts into areal estimates. As GBA2000 was performed long after 2000, given fact that areas are predominantly nonpermanent land cover change, reference data obtained from set Landsat-7 Enhanced Thematic Mapper Plus high-resolution remotely...
In this paper we discuss results of a pilot study conducted by Ukrainian Space Research Institute NASU-NSAU, in collaboration with the MARS team JRC, to explore feasibility, cost-efficiency and specific difficulties crop area estimation assisted satellite remote sensing Ukraine. The compares cost efficiency several image types (MODIS, Landsat TM, AWiFS, LISS-III RapidEye) combined field survey on stratified sample square segments. Additionally, data were collected "along road" as training...
A major component of the Joint Research Centre's TREES-II project is assessment deforestation rates in moist tropical regions for period 1992 to 1997 using a statistical sample fine spatial resolution satellite image pairs. It widely recognized that stratification can reduce variance estimates sampling designs. However, at pan-tropical scale little reliable information available stratify on basis rates. This paper describes novel scheme assessing Stratification performed percentages forest...
Harmonized forest area information provides an important basis for environmental modelling and policy-making at both national international levels. Traditionally, this has been provided by inventory statistics but is now increasingly complemented with remote sensing tools. Reliability harmonization of sources are aspects to ensure comparability enable the development scenarios. Initiatives purpose currently ongoing. Nevertheless, all estimates contain uncertainties, which must be quantified...
Abstract This study uses high-resolution (HR) satellite imagery to quantify the stock of buildings, referred herein as building stock. The risk assessment requires information on natural hazards and element at risk, that is in this article. combines (1) texture-based image processing map built-up areas, (2) statistical sampling allows locating samples (3) photo-interpretation encoding footprints. Statistical inference then used per class size. Legaspi Philippines a case study. results show...
ABSTRACT Accurate characterization of the tropical moist forests changes is needed to support conservation policies and better quantify their contribution global carbon fluxes. We document - at pantropical scale extent these (degradation, deforestation recovery) over last three decades. estimate that 17% have disappeared since 1990 with a remaining area 1060 million ha in 2019, from which 8.5% are degraded. Our study underlines importance degradation process such ecosystems, particular as...