A5075715085- Remote Sensing in Agriculture
- Remote-Sensing Image Classification
- Species Distribution and Climate Change
- Land Use and Ecosystem Services
- Remote Sensing and Land Use
- Remote Sensing and LiDAR Applications
- Advanced Image Fusion Techniques
- Plant Water Relations and Carbon Dynamics
- Climate variability and models
- Data Management and Algorithms
- Cultural and Social Studies in Latin America
- Geochemistry and Geologic Mapping
- Cryospheric studies and observations
- Scientific Computing and Data Management
- Data Mining Algorithms and Applications
- Viral Infections and Vectors
- Smart Agriculture and AI
- Geographic Information Systems Studies
- Vector-borne infectious diseases
- Data-Driven Disease Surveillance
- Comparative Literary Analysis and Criticism
- Research Data Management Practices
- Advanced Clustering Algorithms Research
- Time Series Analysis and Forecasting
- Social Sciences and Policies
University of Twente
2016-2025
GeoInformation (United Kingdom)
2024-2025
Global Initiative on Psychiatry
2011-2020
Diego Portales University
2008-2019
Cornell University
2015
Wageningen University & Research
2004-2011
Université d'Avignon et des Pays de Vaucluse
2004
EMMAH - Environnement Méditerranéen et Modélisation des Agro-Hydrosystèmes
2003
Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement
2003
Lyme borreliosis (LB) is the most prevalent tick-borne disease in Europe and its incidence has steadily increased over last two decades. In Netherlands alone, more than 20,000 citizens are affected by LB each year. Because of this, Dutch citizen science projects were started to monitor tick bites. Both have collected nearly 50,000 geo-located bite reports period 2006-2016. The number per area unit a proxy risk. This risk can also be modelled as result interaction hazard (e.g. activity) human...
Remote sensing studies of vegetation phenology increasingly benefit from freely available satellite imagery acquired with high temporal frequency at fine spatial resolution. Particularly for heterogeneous landscapes this is good news, given the drawback medium-resolution sensors commonly used retrieval (e.g., MODIS) to properly represent fine-scale variability types. The Sentinel-2 mission acquires spectral data globally 10 60 m resolution every five days. To illustrate mission's potential...
An unmixing-based data fusion technique is used to generate images that have the spatial resolution of Landsat Thematic Mapper (TM) and spectral provided by Medium Resolution Imaging Spectrometer (MERIS) sensor. The method requires optimization following two parameters: number classes classify TM image size MERIS ldquowindowrdquo (neighborhood) solve unmixing equations. ERGAS index assess quality fused at resolutions assist with identification best combination parameters need be optimized....
New Earth observation missions and technologies are delivering large amounts of data. Processing this data requires developing evaluating novel dimensionality reduction approaches to identify the most informative features for classification regression tasks. Here we present an exhaustive evaluation Guided Regularized Random Forest (GRRF), a feature selection method based on Forest. GRRF does not require fixing priori number be selected or setting threshold importance. Moreover, use...
Abstract Despite the increasing availability of Open Science (OS) infrastructure and rise in policies to change behaviour, OS practices are not yet norm. While pioneering researchers developing practices, majority sticks status quo. To transition common practice, we must engage a critical proportion academic community. In this transition, Communities (OSCs) play key role. OSCs bottom-up learning groups scholars that discuss within across disciplines. They make knowledge more accessible...
Random cross-validation (CV) is often used to evaluate geospatial machine learning models, particularly when a limited amount of sample data are available, and collecting an extra test set unfeasible. However, the prediction locations can be substantially different from available sample, leading over-optimistic evaluation results. This has fostered development spatial CV methods. Yet these methods only focus on autocorrelation cannot sufficiently guarantee that validation subset good proxy...
Wastewater-based epidemiology was utilized during the COVID-19 outbreak to monitor circulation of SARS-CoV-2, virus causing this disease. However, approach is limited by need for additional methods accurately translate concentrations in wastewater disease-positive human counts. Combined modelling disease cases and concentration its causative virus, will necessarily deepen our understanding. requires addressing technical differences between disease, population mobility models. To that end, we...
The potential applicability of the leaf radiative transfer model PROSPECT (version 3.01) was tested for Norway spruce (Picea abies (L.) Karst.) needles collected from stress resistant and resilient trees. Direct comparison measured simulated optical properties between 450–1000 nm revealed requirement to recalibrate chlorophyll dry matter specific absorption coefficients k ab(λ) m(λ). subsequent validation modified 3.01.S) showed close agreement with spectral measurements all three needle...
Abstract Climate change is expected to modify the timing of seasonal transitions this century, impacting wildlife migrations, ecosystem function, and agricultural activity. Tracking in a consistent manner across space through time requires indices that can be used for monitoring managing biophysical ecological systems during coming decades. Here new gridded dataset spring described understand interannual, decadal, secular trends coterminous United States. This derived from daily interpolated...
Smallholder farmers cultivate more than 80% of the cropland area available in Africa. The intrinsic characteristics such farms include complex crop-planting patterns, and small fields that are vaguely delineated. These pose challenges to mapping crops from space. In this study, we evaluate use a cloud-based multi-temporal ensemble classifier map smallholder farming systems case study for southern Mali. combines selection spatial spectral features derived multi-spectral Worldview-2 images,...
Applications of machine-learning-based approaches in the geosciences have witnessed a substantial increase over past few years. Here we present an approach that accounts for spatial autocorrelation by introducing features to models. In particular, explore two types features, namely lag and eigenvector filtering (ESF). These are used within widely random forest (RF) method, their effect is illustrated on public datasets varying sizes (Meuse California housing datasets). The least absolute...
The production of land cover maps through satellite image classification is a frequent task in remote sensing. Random Forest (RF) and Support Vector Machine (SVM) are the two most well-known recurrently used methods for this task. In paper, we evaluate pros cons using an RF-based kernel (RFK) SVM compared to conventional Radial Basis Function (RBF) standard RF classifier. A time series seven multispectral WorldView-2 images acquired over Sukumba (Mali) single hyperspectral AVIRIS Salinas...
Abstract Although developments in remote sensing have greatly improved land cover mapping, the mixed pixel problem has not yet been fully addressed. Soft classification techniques introduced to address problem, but they do show spatial location of class proportions a pixel. Subpixel mapping drawbacks soft classifications. In this work, feedforward backpropagating neural network (FFBPNN) was used for subpixel mapping. A set proportion images, which are be treated as results, were created from...
Data from current medium-spatial-resolution imaging spectroradiometers are used for land-cover mapping and change detection at regional to global scales. However, few landscapes homogeneous these scales, this creates the so-called mixed-pixel problem. In context, study explores use of linear spectral mixture model extract subpixel composition data. particular, a time series MEdium Resolution Imaging Spectrometer (MERIS) full-resolution (FR; pixel size 300 m) images acquired over The...
Geographical information systems support the application of statistical techniques to map spatially referenced crop data. To do this in optimal way, errors and uncertainties have be minimized that are often associated with operations on This paper applies a spatial approach upscale yields from field level toward scale Burkina Faso. Observed were related Normalized Difference Vegetation Index derived SPOT-VEGETATION. The objective was quantify at subsequent steps. First, we applied point...
Even though many studies have shown the usefulness of clustering for exploration spatio-temporal patterns, until now there is no systematic description methods geo-referenced time series (GTS) classified as one-way clustering, co-clustering and tri-clustering methods. Moreover, selection a suitable method given dataset task remains to be challenge. Therefore, we present an overview existing GTS, using aforementioned classification, compare different provide suggestions appropriate For this...