Leaf Water Potential in a Mixed Mediterranean Forest from Machine Learning and Unmanned Aerial Vehicle (UAV)-Based Hyperspectral Imaging DOI Creative Commons

Netanel Fishman,

Yehuda Yungstein,

Assaf Yaakobi

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 17(1), P. 106 - 106

Published: Dec. 31, 2024

Leaf water potential (ψleaf) is a key indicator of plant status, but its measurement labor-intensive and limited in spatial coverage. While remote sensing has emerged as useful tool for estimating vegetation ψleaf remains unexplored, particularly mixed forests. Here, we use spectral indices derived from unmanned aerial vehicle-based hyperspectral imaging machine learning algorithms to assess mixed, multi-species Mediterranean forest comprised five woody species: Pinus halepensis, Quercus calliprinos, Cupressus sempervirens, Ceratonia siliqua, Pistacia lentiscus. Hyperspectral images (400–1000 nm) were acquired monthly over one year, concurrent with measurements each species. Twelve thousands normalized difference index (NDSI) combinations evaluated. Three algorithms—random (RF), extreme gradient boosting (XGBoost), support vector (SVM)—were used model ψleaf. We compared the results linear models based on NDSI. SVM, using species information feature, performed best relatively good assessment (R2 = 0.53; RMSE 0.67 MPa; rRMSE 28%), especially considering small seasonal variance (±σ 0.8 MPa). Predictions sempervirens 0.80) lentiscus 0.49), which had largest variances > 1 Aggregating data at plot scale ‘general’ markedly improved 0.79, 0.31 13%), providing promising monitoring The fact that non-species-specific, could predict implies such can also be coarser resolution satellite data. Our study demonstrates combining imagery non-invasive estimation forests while highlighting challenges capturing interspecies variability.

Language: Английский

Drought legacies in mixed Mediterranean forests: Analysing the effects of structural overshoot, functional traits and site factors DOI Creative Commons
Santain Settimio Pino Italiano, J. Julio Camarero, Marco Borghetti

et al.

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 927, P. 172166 - 172166

Published: April 3, 2024

Previous favorable climate conditions stimulate tree growth making some forests more vulnerable to hotter droughts. This so-called structural overshoot may contribute forest dieback, but there is little evidence on its relative importance depending site and species because of limited field data. Here, we analyzed remote sensing (NDVI) tree-ring width data evaluate the impacts 2017 drought canopy cover in mixed Mediterranean (Fraxinus ornus, Quercus pubescens, Acer monspessulanum, Pinus pinaster) located southern Italy. Legacy effects were assessed by calculating differences between observed predicted basal area increment (BAI). Overall, response study stands was contingent characteristics. Most sites presented BAI reductions during drought. Growth decline followed a quick recovery positive legacy effects, particularly case F. ornus. However, found negative legacies (e.g., Q. A. monspessulanum) sites. In those showing legacies, high rates prior previous wet winter-spring have predisposed trees damage. Vice versa, ornus linked post-drought release due pubescens dieback mortality. Therefore, evidences overshoot, it restricted specific species. Our findings highlight considering settings such as stand composition, pre-drought different when studying overshoot. Droughts modify composition dynamics forests.

Language: Английский

Citations

5
Analysing resilience of European beech tree to recurrent extreme drought events through ring growth, wood anatomy and stable isotopes DOI Creative Commons
Guangqi Zhang, Nathalie Bréda,

Nicolas Steil

et al.

Journal of Ecology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 17, 2025

Abstract Recent extreme drought events in Central Europe have caused widespread forest dieback with detrimental effects on functioning and carbon water balance. This impact has been notable European beech ( Fagus sylvatica L.), particularly at the core of its distribution, causing concern among stakeholders questions about resilience capacity trees. The objective this study is to investigate physiological processes linked constraints involved cambial growth drought. We selected 56 trees distributed four plots north‐eastern France different soil deficits characterized retrospectively by a balance model. Functional traits including tree ring width, wood anatomical stable isotopes (e.g. δ 13 C 18 O) were measured assess effect recent recurrent 2015, 2018–2020, 2022. Decreased increased O intrinsic use efficiency (iWUE) observed due shortage, whereas xylem vessel size theoretical specific hydraulic conductivity K th ) did not show obvious changes. Vessel density was negatively correlated annual width index highly sensitive C, iWUE significantly related index. plot that experienced most severe intensity 2018–2020 showed significant decrease resistance compared during 2015 event. Surprisingly, associated isotopic traits. Synthesis . Our results demonstrate structure responds adjusting radial rings relatively diameter. also highlights consecutive or reducing resilience, sites higher intensity. Tree does seem involve changes would promote better cope future shortages.

Language: Английский

Citations

0
Linking remotely sensed growth-related canopy attributes to interannual tree-ring width variations: A species-specific study using Sentinel optical and SAR time series DOI Creative Commons
Vahid Nasiri, Paweł Hawryło, Piotr Tompalski

et al.

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2025, Volume and Issue: 221, P. 347 - 362

Published: Feb. 20, 2025

Language: Английский

Citations

0
Watering the trees for the forest: Drought alleviation in oaks and pines by ancestral ditches DOI Creative Commons
J. Julio Camarero, María J. Salinas‐Bonillo, Cristina Valeriano

et al.

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 950, P. 175353 - 175353

Published: Aug. 6, 2024

Traditional ditches ("acequias" in Spanish) derive meltwater and infiltrate groundwater providing ecological services downstream the semi-arid Sierra Nevada range (SE Spain). Therefore, they may act as a nature-based solution by alleviating drought stress trees growing near enhancing growth reducing their intrinsic water-use efficiency (iWUE). Such mitigation role of acequias is critical given that some oak (Quercus pyrenaica) pine (Pinus sylvestris) stands reach xeric distribution limits Europe. We compared tree-ring width data wood δ

Language: Английский

Citations

0
Evaluating Sentinel-2 Time Series for Monitoring Dieback Reveals Different Responses Among Temperate Conifer Species DOI

Hélène Carletti,

Jean‐Claude Gégout,

Raphaël Dutrieux

et al.

Published: Jan. 1, 2024

Language: Английский

Citations

0
Linking patterns of forest dieback to triggering climatic and weather events: an overview on Mediterranean forests DOI
Francesca Alderotti, Filippo Bussotti, Cecilia Brunetti

et al.

iForest - Biogeosciences and Forestry, Journal Year: 2024, Volume and Issue: 17(5), P. 309 - 316

Published: Sept. 30, 2024

Language: Английский

Citations

0
Elevational Effects of Climate Warming on Tree Growth in a Picea schrenkiana Forest in the Eastern Tianshan Mountains DOI Open Access
Jianing He,

Zehao Shen,

Caiwen Ning

et al.

Forests, Journal Year: 2024, Volume and Issue: 15(12), P. 2052 - 2052

Published: Nov. 21, 2024

Considerable uncertainty exists regarding the overall effects of future climate change on forests in arid mountains, and elevational range drought-induced tree growth decline remains unclear. Tianshan is largest mountain regions globally. Here, we analyzed ring data pure stands Schrenk spruce (Picea schrenkiana Fisch. et Mey.) Jiangbulake region eastern Mountains along an gradient (1800–2600 m a.s.l.). The radial P. trees declined three nine sample strips (1800–2100 a.s.l.) over last two decades. response (measured by width index, RWI) to temperature significantly changed at “inflection point” 2100–2200 m. RWI was negatively correlated with low elevations, whereas opposite observed high elevations. Precipitation minimum temperatures winter spring mean summer were main drivers growth, effect maximum concentrated spring. In addition warming study area since 1970s, (as measured basal increment, BAI) elevations below 2200 initially increased then decreased. Tree higher continues increase. Since 2000, average exceeded that BAI values have gradually approached each other recent decades, although lower exhibited past.

Language: Английский

Citations

0
Leaf Water Potential in a Mixed Mediterranean Forest from Machine Learning and Unmanned Aerial Vehicle (UAV)-Based Hyperspectral Imaging DOI Creative Commons

Netanel Fishman,

Yehuda Yungstein,

Assaf Yaakobi

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 17(1), P. 106 - 106

Published: Dec. 31, 2024

Leaf water potential (ψleaf) is a key indicator of plant status, but its measurement labor-intensive and limited in spatial coverage. While remote sensing has emerged as useful tool for estimating vegetation ψleaf remains unexplored, particularly mixed forests. Here, we use spectral indices derived from unmanned aerial vehicle-based hyperspectral imaging machine learning algorithms to assess mixed, multi-species Mediterranean forest comprised five woody species: Pinus halepensis, Quercus calliprinos, Cupressus sempervirens, Ceratonia siliqua, Pistacia lentiscus. Hyperspectral images (400–1000 nm) were acquired monthly over one year, concurrent with measurements each species. Twelve thousands normalized difference index (NDSI) combinations evaluated. Three algorithms—random (RF), extreme gradient boosting (XGBoost), support vector (SVM)—were used model ψleaf. We compared the results linear models based on NDSI. SVM, using species information feature, performed best relatively good assessment (R2 = 0.53; RMSE 0.67 MPa; rRMSE 28%), especially considering small seasonal variance (±σ 0.8 MPa). Predictions sempervirens 0.80) lentiscus 0.49), which had largest variances > 1 Aggregating data at plot scale ‘general’ markedly improved 0.79, 0.31 13%), providing promising monitoring The fact that non-species-specific, could predict implies such can also be coarser resolution satellite data. Our study demonstrates combining imagery non-invasive estimation forests while highlighting challenges capturing interspecies variability.

Language: Английский

Citations

0