Systematic review for a set of indicators supporting the Common International Classification of Ecosystem Services

Ecological Indicators, Journal Year: 2023, Volume and Issue: 147, P. 109978 - 109978

Published: Feb. 8, 2023

Ecosystem services (ES) contribute to human well-being and provide an important contribution economies at all scales. However, ES are often difficult measure quantify, thus, it is adequately account for the true value of their contributions. The use indicators, understood as proxies estimating provision ES, has been proposed a solution this obstacle. In context, indicators physical elements ecosystems that can be relatively easily quantified with available tools knowledge, usually communicated decision-makers practitioners. study, we conducted literature review peer reviewed publications, aiming complete up-to-date list ES. total, generated 85 individual have previously used in practice linked them each one described by CICES (v5.1) classification system. Moreover, identified which those could derived from remotely sensed (RS) data following three categories: i) RS direct relation indicator, ii) indirect indicator requires additional information or modelling, iii) Indicators not derivable currently without enough available. Only minority these (6) directly data, while most (46) indirectly, some (33) data.

Language: Английский

---

Remotely sensed functional diversity and its association with productivity in a subtropical forest

Remote Sensing of Environment, Journal Year: 2023, Volume and Issue: 290, P. 113530 - 113530

Published: March 10, 2023

Functional diversity is a critical component driving ecosystem functioning. Spatially explicit data of plant functional traits and are essential for understanding biodiversity effects on Here we retrieved three morphological (95th quantile height, leaf area index, foliage height diversity) physiological (chlorophyll + b content, specific area, equivalent water thickness) from airborne laser scanning multispectral Sentinel-2 data, respectively. We found LiDAR-derived parameters correlated well with in-situ plot-level (R2 ≥ 0.67). For satellite-derived traits, partial least squares regression (PLSR) obtained higher prediction accuracy = 0.26–0.43, cross-validation community-weighted mean (CWM) trait data) than vegetation index (VI) approach. The remotely-sensed were used as input to estimate multi-trait (FD) indices in species-rich subtropical mountainous forest. Finally, investigated the influence single-trait CWMs, FD environmental variables remotely-derived aboveground carbon stocks (aboveground biomass, AGB) primary productivity (kernel normalized difference kNDVI). CWMs all significant predictors AGB kNDVI, suggested by mass-ratio hypothesis. Morphological also important indicating complementarity crown architectures. In best-fit multivariate models, first principal CWM that most richness was additionally selected models kNDVI at landscape scales. Our work highlights potential using assess relationship between functioning across large, contiguous areas.

Language: Английский

---

Mapping leaf area index in a mixed temperate forest using Fenix airborne hyperspectral data and Gaussian processes regression

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2020, Volume and Issue: 95, P. 102242 - 102242

Published: Oct. 17, 2020

Machine learning algorithms, in particular, kernel-based machine methods such as Gaussian processes regression (GPR) have shown to be promising alternatives traditional empirical for retrieving vegetation parameters from remotely sensed data. However, the performance of GPR predicting forest biophysical has hardly been examined using full-spectrum airborne hyperspectral The main objective this study was evaluate potential estimate leaf area index (LAI) To achieve this, field measurements LAI were collected Bavarian Forest National Park (BFNP), Germany, concurrent with acquisition Fenix images (400−2500 nm) July 2017. further compared three commonly used (i.e., narrowband indices (VIs), partial least square (PLSR), and artificial neural network (ANN)). cross-validated coefficient determination (Rcv2) root mean error (RMSEcv) between retrieved field-measured examine accuracy respective methods. Our results showed that entire spectral data nm), yielded most accurate estimation (Rcv2 = 0.67, RMSEcv 0.53 m2 m−2) best performing VIs SAVI2 0.54, 0.63 m−2), PLSR 0.74, 0.73 ANN 0.68, 0.54 m−2). Consequently, when a subset obtained analysis model input, predictive accuracies generally improved (GPR 0.52 m−2; 0.55 0.69 indicating extracting useful information vast bands is crucial improving performance. In general, there an agreement measured estimated different approaches (p > 0.05). generated map BFNP endorsed spatial distribution across dominant classes (e.g., deciduous stands associated higher values). accompanying uncertainty by shows uncertainties observed mainly regions low values (low cover) areas which not well represented sample plots. This demonstrated estimating Owing its capability generate predictions estimates, evaluated candidate operational retrieval applications traits.

Language: Английский

---

Early detection of spruce vitality loss with hyperspectral data: Results of an experimental study in Bavaria, Germany

Remote Sensing of Environment, Journal Year: 2021, Volume and Issue: 266, P. 112676 - 112676

Published: Sept. 21, 2021

Vitality loss of trees caused by extreme weather conditions, drought stress or insect infestations, are expected to increase with ongoing climate change. The detection vitality at an early stage is thus vital importance for forestry and forest management minimize ecological economical damage. Remote sensing instruments able detect changes over large areas down the level individual trees. scope our study investigate whether it possible stress-related spectral using hyperspectral sensors. For this purpose, two Norway spruce (Picea abies) stands, both different in age maintenance, were monitored field vegetation periods. In parallel, time series airborne remote data acquired. each stand 70 artificially stressed (ring-barked) used as control collected south-eastern Germany consists measurements multiple times scales: (1) crown conditions visually assessed (2) needle reflectance spectra acquired laboratory a FieldSpec spectrometer, (3) (HySpex) flown 0.5 m spatial resolution. We aimed simultaneous acquisition three levels. This unique set was investigated any feature can be discriminated stage. Several transformations applied tree spectra, such derivatives, indices angle indices. All features examined their separability (ring-barked vs. trees) Random Forest (RF) classification algorithm. As result, younger, well maintained only showed minor 2-year period, whereas older observable respectively. These could even detected before visible observations. reactions ring-barking first noticeable 11 months after 6 weeks they inspection. most discriminative separating groups VIs separated RF classifier 79% overall accuracy beginning second period 1 month later 92% high kappa index. results clearly demonstrate great potential detecting

Language: Английский

---

The EnMAP spaceborne imaging spectroscopy mission: Initial scientific results two years after launch

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: unknown, P. 114379 - 114379

Published: Sept. 1, 2024

Language: Английский

---

Leaf pigment retrieval using the PROSAIL model: Influence of uncertainty in prior canopy-structure information

The Crop Journal, Journal Year: 2022, Volume and Issue: 10(5), P. 1251 - 1263

Published: May 16, 2022

Leaf pigments are critical indicators of plant photosynthesis, stress, and physiological conditions. Inversion radiative transfer models (RTMs) is a promising method for robustly retrieving leaf biochemical traits from canopy observations, adding prior information has been effective in alleviating the "ill-posed" problem, major challenge model inversion. Canopy structure parameters, such as area index (LAI) average inclination angle (ALA), can serve pigment retrieval. Using spectra simulated PROSAIL model, we estimated effects uncertainty LAI ALA used lookup table-based inversions chlorophyll (Cab) carotenoid (Car). The retrieval accuracies two were increased by use priors (RMSE Cab 7.67 to 6.32 μg cm−2, Car 2.41 2.28 cm−2) 5.72 2.23 cm−2). However, this improvement deteriorated with an increase additive multiplicative uncertainties, when 40% 20% noise was added respectively, these ceased accuracy. Validation using experimental winter wheat dataset also showed that compared Car, estimation accuracy more or less structure. This study demonstrates possible limitations RTM biochemistry, large uncertainties present.

Language: Английский

---

Climate-Change-Driven Droughts and Tree Mortality: Assessing the Potential of UAV-Derived Early Warning Metrics

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(10), P. 2627 - 2627

Published: May 18, 2023

Protecting and enhancing forest carbon sinks is considered a natural solution for mitigating climate change. However, the increasing frequency, intensity, duration of droughts due to change can threaten stability growth existing sinks. Extreme weaken plant hydraulic systems, lead tree mortality events, may reduce diversity, making forests more vulnerable subsequent disturbances, such as fires or pest infestations. Although early warning metrics (EWMs) derived using satellite remote sensing data are now being tested predicting post-drought physiological stress mortality, applications unmanned aerial vehicles (UAVs) yet be explored extensively. Herein, we provide twenty-four prospective approaches classified into five categories: (i) complexities, (ii) site-specific confounding (abiotic) factors, (iii) interactions with biotic agents, (iv) monitoring optimization, (v) technological infrastructural developments, adoption, future operationalization, upscaling UAV-based frameworks EWM applications. These UAV considerations paramount they hold potential bridge gap between field inventory assessing characteristics their responses drought conditions, identifying prioritizing conservation needs and/or high-carbon-efficient species efficient allocation resources, optimizing management adaptation mitigation practices in timely cost-effective manner.

Language: Английский

---

High resolution retrieval of leaf chlorophyll content over Himalayan pine forest using Visible/IR sensors mounted on UAV and radiative transfer model

Ecological Informatics, Journal Year: 2023, Volume and Issue: 75, P. 102099 - 102099

Published: April 8, 2023

Language: Английский

---

Deep Learning and Machine Learning in Hydrological Processes, Climate Change and Earth Systems: A Systematic Review

Published: Aug. 15, 2019

Artificial intelligence methods and application have recently shown great contribution in modeling prediction of the hydrological processes, climate change, earth systems. Among them, deep learning machine mainly reported being essential for achieving higher accuracy, robustness, efficiency, computation cost, overall model performance. This paper presents state art applications this realm current state, future trends are discussed. The survey advances presented through a novel classification methods. concludes that is still first stages development, research progressing. On other hand, already established fields, with performance emerging ensemble techniques hybridization.

Language: Английский

---

Monitoring LAI, Chlorophylls, and Carotenoids Content of a Woodland Savanna Using Hyperspectral Imagery and 3D Radiative Transfer Modeling

Remote Sensing, Journal Year: 2019, Volume and Issue: 12(1), P. 28 - 28

Published: Dec. 19, 2019

Leaf pigment contents, such as chlorophylls a and b content (C ) or carotenoid (Car), the leaf area index (LAI) are recognized indicators of plants’ forests’ health status that can be estimated through hyperspectral imagery. Their measurement on seasonal yearly basis is critical to monitor plant response adaptation stress, droughts. While extensively done over dense canopies, estimation these variables tree-grass ecosystems with very low overstory LAI (mean site < 1 m 2 /m ), woodland savannas, lacking. We investigated use look-up table (LUT)-based inversion radiative transfer model retrieve C Car from AVIRIS images at an 18 spatial resolution multiple dates broadleaved savanna during California drought. compared performances different cost functions in step. demonstrated consistency our LAI, , estimations using validation data high canopy cover parts site, their temporal by qualitatively confronting variations two years those would expected. concluded LUT-based inversions medium-resolution images, achieved simple geometric representation within 3D (RTM), valid means monitoring savannas more generally sparse forests, although for maximum applicability, should selected dates. Validation revealed use: The normalized difference vegetation (NDVI) outperformed other indices (root mean square error (RMSE) = 0.22 R 0.81); band ratio ρ 0.750 μ 0.550 retrieved accurately than (RMSE 5.21 g/cm 0.73); RMSE 0.5–0.55 interval showed encouraging results estimations.

Language: Английский

---