Satellite remote sensing of vegetation phenology: Progress, challenges, and opportunities

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2024, Volume and Issue: 217, P. 149 - 164

Published: Aug. 29, 2024

Language: Английский

---

A Sentinel-2 machine learning dataset for tree species classification in Germany

Earth system science data, Journal Year: 2025, Volume and Issue: 17(2), P. 351 - 367

Published: Feb. 3, 2025

Abstract. We present a machine learning dataset for tree species classification in Sentinel-2 satellite image time series of bottom-of-atmosphere reflectance. It is geared towards training classifiers but less suitable validating the resulting maps. The based on German National Forest Inventory 2012 as well analysis-ready imagery computed using Framework Operational Radiometric Correction Environmental monitoring (FORCE) processing pipeline. From data, we extracted positions, filtered 387 775 trees upper canopy layer, and automatically corresponding reflectance from L2A images. These are labeled with species, which allows pixel-wise tasks. Furthermore, provide auxiliary information such approximate position, year possible disturbance events, or diameter at breast height. Temporally, spans years July 2015 to end October 2022, approx. 75.3 million data points 48 3 groups 13.8 observations non-tree backgrounds. Spatially, it covers whole Germany. available following DOI (Freudenberg et al., 2024): https://doi.org/10.3220/DATA20240402122351-0.

Language: Английский

---

From UAV to PlanetScope: Upscaling fractional cover of an invasive species Rosa rugosa

Journal of Environmental Management, Journal Year: 2023, Volume and Issue: 336, P. 117693 - 117693

Published: March 11, 2023

Invasive plant species pose a direct threat to biodiversity and ecosystem services. Among these, Rosa rugosa has had severe impact on Baltic coastal ecosystems in recent decades. Accurate mapping monitoring tools are essential quantify the location spatial extent of invasive support eradication programs. In this paper we combined RGB images obtained using an Unoccupied Aerial Vehicle, with multispectral PlanetScope map R. at seven locations along Estonian coastline. We used RGB-based vegetation indices 3D canopy metrics combination random forest algorithm thickets, obtaining high accuracies (Sensitivity = 0.92, specificity 0.96). then presence/absence maps as training dataset predict fractional cover based derived from constellation Extreme Gradient Boosting (XGBoost). The XGBoost yielded prediction (RMSE 0.11, R2 0.70). An in-depth accuracy assessment site-specific validations revealed notable differences between study sites (highest 0.74, lowest 0.03). attribute these various stages invasion density thickets. conclusion, UAV is cost-effective method highly heterogeneous ecosystems. propose approach valuable tool extend local geographical scope assessments into wider areas regional evaluations.

Language: Английский

---

Sentinel-2 versus PlanetScope Images for Goldenrod Invasive Plant Species Mapping

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(4), P. 636 - 636

Published: Feb. 8, 2024

A proliferation of invasive species is displacing native species, occupying their habitats and degrading biodiversity. One these the goldenrod (Solidago spp.), characterized by aggressive growth that results in habitat disruption as it outcompetes plants. This invasiveness also leads to altered soil composition through release allelopathic chemicals, complicating control efforts making challenging maintain ecological balance affected areas. The research goal was develop methods allow analysis changes heterogeneous with high accuracy repeatability. For this reason, we used open source classifiers Support Vector Machine (SVM), Random Forest (RF), satellite images Sentinel-2 (free) PlanetScope (commercial) assess potential classification. Due fact invasions begin invasion footholds, created small patches invasive, autochthonous plants different land cover patterns (asphalt, concrete, buildings) forming areas, based our studies on field-verified polygons, which allowed selection randomized pixels for training validation iterative classifications. confirmed optimal solution use multitemporal RF classifier, combination gave F1-score 0.92–0.95 polygons dominated 0.85–0.89 areas where minority (mix class; smaller share canopy than plants). mean decrease (MDA), indicating an informativeness individual spectral bands, showed bands coastal aerosol, NIR, green, SWIR, red were comparably important, while case data, NIR definitely most remaining less informative, yellow (B5) did not contribute significant information even during flowering period, when plant covered intensely perianth, red-edge, or green II much more important. maximum classification values are similar (F1-score > 0.9), but medians lower especially SVM algorithm.

Language: Английский

---

Transforming forest management through rewilding: Enhancing biodiversity, resilience, and biosphere sustainability under global change

One Earth, Journal Year: 2025, Volume and Issue: 8(3), P. 101195 - 101195

Published: March 1, 2025

Language: Английский

---

Identification of Acacia invasive species in protected areas of Spain using PlanetScope high-resolution satellite images and machine learning models in time series: an important action for protective management of forests

Forest Ecology and Management, Journal Year: 2025, Volume and Issue: 586, P. 122696 - 122696

Published: April 9, 2025

Language: Английский

---

Advanced Detection of Invasive Neophytes in Agricultural Landscapes: A Multisensory and Multiscale Remote Sensing Approach

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(3), P. 500 - 500

Published: Jan. 28, 2024

The sustainable provision of ecological products and services, both natural man-made, faces a substantial threat emanating from invasive plant species (IPS), which inflict considerable economic harm on global scale. They are widely recognized as one the primary drivers biodiversity decline have become focal point an increasing number studies. integration remote sensing (RS) geographic information systems (GIS) plays pivotal role in their detection classification across diverse range research endeavors, emphasizing critical significance accounting for phenological stages targeted when endeavoring to accurately delineate distribution occurrences. This study is centered this fundamental premise, it endeavors amass terrestrial data encompassing spectral attributes specified IPS, with overarching objective ascertaining most opportune time frames detection. Moreover, involves development validation algorithm, harnessing array RS datasets, including satellite unmanned aerial vehicle (UAV) imagery spanning spectrum RGB multispectral near-infrared (NIR). Taken together, our investigation underscores advantages employing datasets conjunction stages, offering economically efficient adaptable solution monitoring species. Such insights hold potential inform present future policymaking pertaining management agricultural ecosystems.

Language: Английский

---

Deep learning and satellite remote sensing for biodiversity monitoring and conservation

Remote Sensing in Ecology and Conservation, Journal Year: 2024, Volume and Issue: unknown

Published: June 17, 2024

Abstract In the context of current nature crisis, being able to reliably and cost‐effectively track subtle changes in biosphere across adequate spatial temporal extents resolutions is crucial. Deep learning represents a group versatile approaches image processing tasks that are increasingly combined with satellite remote sensing imagery monitor biodiversity inform ecology conservation, yet an overview opportunities challenges associated this development has so far been lacking. Here, we provide interdisciplinary perspective on research technological developments deep have potential make difference monitoring wildlife conservation; highlight broader adoption these by experts operating at interface between discuss how can be overcome. By enabling leveraging big data providing new ways learn about its dynamics, promise become powerful tool help address needs knowledge gaps. certain situations, may moreover substantially reduce time resources required analyse imagery. However, issues relating capacity building, reference access, environmental costs as well model interpretability, robustness alignment need addressed successfully capitalize opportunities.

Language: Английский

---

Mapping invasive noxious weed species in the alpine grassland ecosystems using very high spatial resolution UAV hyperspectral imagery and a novel deep learning model

GIScience & Remote Sensing, Journal Year: 2024, Volume and Issue: 61(1)

Published: March 13, 2024

The term "invasive noxious weed species" (INWS), which refers to plants that invade native alpine grasslands, has increasingly become an ecological and economic threat in the grassland ecosystem of Qinghai-Tibetan Plateau (QTP). Both INWS grass species are small physical size share a habitat. Using remote sensing data distinguish from remains challenge. High spatial resolution hyperspectral imagery provides alternative for addressing this problem. Here, we explored use unmanned aerial vehicle (UAV) deep learning methods with sample mapping mixed grasslands. To assess method, UAV very high 2 cm were collected study site, novel convolutional neural network (CNN) model called 3D&2D-INWS-CNN was developed take full advantage rich information provided by imagery. results indicate proposed applied ground truth training samples is robust sufficient, overall classification accuracy exceeding 95% kappa value 98.67%. F1 score each ranged 92% 99%. In conclusion, our highlight potential using combined state-of-the-art even degraded ecosystems. Studies such as ours can aid development invasive management practices provide more decision-making controlling spread similar ecosystems or, widely, terrestrial

Language: Английский

---

Wetland vegetation mapping improved by phenological leveraging of multitemporal nanosatellite images

Geocarto International, Journal Year: 2025, Volume and Issue: 40(1)

Published: Jan. 21, 2025

Language: Английский

---