Extending Estimates of Tree and Tree Species Presence-Absence through Space and Time Using Landsat Composites

Strickland, Guy; Luther, Joan E.; White, Joanne C.; Wulder, Michael A.

doi:10.1080/07038992.2020.1811083

Cited by 9 publications

(8 citation statements)

References 85 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For forest management purposes, a large, consistent, standardized, long-term and high resolution image collection such as the one provided by the Landsat program can help extending in space and time information on tree species presence, composition and abundance. A spatial resolution of 30 m is particularly well suited for NFI applications: Strickland et al (2020) derived probability maps of forest tree species for a 25 years time period (1985–2010) using yearly Landsat composites to extend missing information from the Canadian NFI and estimating changes in forest cover, species composition and forest disturbances. The increasing availability of even higher-spatial resolution satellite data from the European Copernicus program ( i.e.…”

Section: Discussionmentioning

confidence: 99%

Forest tree species distribution for Europe 2000–2020: mapping potential and realized distributions using spatiotemporal machine learning

Bonannella

Hengl²,

Heisig

et al. 2022

PeerJ

View full text Add to dashboard Cite

This article describes a data-driven framework based on spatiotemporal machine learning to produce distribution maps for 16 tree species (Abies alba Mill., Castanea sativa Mill., Corylus avellana L., Fagus sylvatica L., Olea europaea L., Picea abies L. H. Karst., Pinus halepensis Mill., Pinus nigra J. F. Arnold, Pinus pinea L., Pinus sylvestris L., Prunus avium L., Quercus cerris L., Quercus ilex L., Quercus robur L., Quercus suber L. and Salix caprea L.) at high spatial resolution (30 m). Tree occurrence data for a total of three million of points was used to train different algorithms: random forest, gradient-boosted trees, generalized linear models, k-nearest neighbors, CART and an artificial neural network. A stack of 305 coarse and high resolution covariates representing spectral reflectance, different biophysical conditions and biotic competition was used as predictors for realized distributions, while potential distribution was modelled with environmental predictors only. Logloss and computing time were used to select the three best algorithms to tune and train an ensemble model based on stacking with a logistic regressor as a meta-learner. An ensemble model was trained for each species: probability and model uncertainty maps of realized distribution were produced for each species using a time window of 4 years for a total of six distribution maps per species, while for potential distributions only one map per species was produced. Results of spatial cross validation show that the ensemble model consistently outperformed or performed as good as the best individual model in both potential and realized distribution tasks, with potential distribution models achieving higher predictive performances (TSS = 0.898, R2logloss = 0.857) than realized distribution ones on average (TSS = 0.874, R2logloss = 0.839). Ensemble models for Q. suber achieved the best performances in both potential (TSS = 0.968, R2logloss = 0.952) and realized (TSS = 0.959, R2logloss = 0.949) distribution, while P. sylvestris (TSS = 0.731, 0.785, R2logloss = 0.585, 0.670, respectively, for potential and realized distribution) and P. nigra (TSS = 0.658, 0.686, R2logloss = 0.623, 0.664) achieved the worst. Importance of predictor variables differed across species and models, with the green band for summer and the Normalized Difference Vegetation Index (NDVI) for fall for realized distribution and the diffuse irradiation and precipitation of the driest quarter (BIO17) being the most frequent and important for potential distribution. On average, fine-resolution models outperformed coarse resolution models (250 m) for realized distribution (TSS = +6.5%, R2logloss = +7.5%). The framework shows how combining continuous and consistent Earth Observation time series data with state of the art machine learning can be used to derive dynamic distribution maps. The produced predictions can be used to quantify temporal trends of potential forest degradation and species composition change.

show abstract

Section: Discussionmentioning

confidence: 99%

Forest tree species distribution for Europe 2000–2020: mapping potential and realized distributions using spatiotemporal machine learning

Bonannella

Hengl²,

Heisig

et al. 2022

PeerJ

View full text Add to dashboard Cite

show abstract

“…Despite its complexity, modeling the realized distribution in current conditions may still be feasible, for example using Earth Observation (EO) data, but there are only few examples of such exercises (Bonannella et al, 2022;Gelfand and Shirota, 2021;Hefley and Hooten, 2016). Compared to previous studies focusing on long term future predictions using climate models, using EO data based models allows for timely and consistent assessments of ongoing changes (Strickland et al, 2020); however, a comprehensive understanding of changes in tree species distributions trends in the face of climate change is still lacking.…”

Section: Introductionmentioning

confidence: 99%

Multi-decadal trend analysis and forest disturbance assessment of European tree species: concerning signs of a subtle shift

Bonannella,

Parente,

de Bruin

et al. 2023

Preprint

View full text Add to dashboard Cite

Climate change poses a significant threat to the distribution and composition of forest tree species worldwide. European forest tree species' range is expected to shift to cope with the increasing frequency and intensity of extreme weather events, pests and diseases caused by climate change. Despite numerous regional studies, a continental scale assessment of current changes in species distributions in Europe is missing due to the difficult task of modeling a species realized distribution and to quantify the influence of forest disturbances on each species. In this study we conducted a trend analysis on the realized distribution of 6 main European forest tree species (Abies alba Mill., Fagus sylvatica L., Picea abies L. H. Karst., Pinus nigra J. F. Arnold, Pinus sylvestris L. and Quercus robur L.) to capture and map the prevalent trends in probability of occurrence for the period 2000–2020. We also analyzed the impact of forest disturbances on each species' range and identified the dominant disturbance drivers. Our results revealed an overall trend of stability in species' distributions (85% of the pixels are considered stable by 2020 for all species) but we also identified some hot spots characterized by negative trends in probability of occurrence, mostly at the edges of each species' latitudinal range. Additionally, we identified a steady increase in disturbance events in each species' range by disturbance (affected range doubled by 2020, from 3.5% to 7% on average) and highlighted species-specific responses to forest disturbance drivers such as wind and fire. Overall, our study provides insights into distribution trends and disturbance patterns for the main European forest tree species. The identification of range shifts and the intensifying impacts of disturbances call for proactive conservation efforts and long-term planning to ensure the resilience and sustainability of European forests.

show abstract

“…Compared to previous studies focusing on long term future predictions using climate models, using EO data based models allows for timely and consistent assessments of ongoing changes (Strickland et al, 2020): EO data does not only provide higher temporal resolution compared to, for example, periodic surveys such as National Forest Inventories (NFIs) programs, but also offers a broader spatial coverage, including remote or inaccessible regions which are difficult to monitor from the ground. EO data also offers a compelling advantage to extensive field surveys in terms of cost efficiency: accessing and processing EO data is nowadays remarkably streamlined due to the free availability of large collections of data, like the Landsat mission (Zhu et al, 2019), or tools able to analyze such large collections like the cloud computing platform Google Earth Engine (Gorelick et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

Multi-decadal trend analysis and forest disturbance assessment of European tree species: concerning signs of a subtle shift

Bonannella,

Parente,

de Bruin

et al. 2023

Preprint

View full text Add to dashboard Cite

Climate change poses a significant threat to the distribution and composition of forest tree species worldwide. European forest tree species’ range is expected to shift to cope with the increasing frequency and intensity of extreme weather events, pests and diseases caused by climate change. Despite numerous regional studies, a continental scale assessment of current changes in species distributions in Europe is missing due to the difficult task of modeling a species realized distribution and to quantify the influence of forest disturbances on each species. In this study we conducted a trend analysis on the realized distribution of 6 main European forest tree species (Abies alba Mill., Fagus sylvatica L., Picea abies L. H. Karst., Pinus nigra J. F. Arnold, Pinus sylvestris L. and Quercus robur L.) to capture and map the prevalent trends in probability of occurrence for the period 2000–2020. We also analyzed the impact of forest disturbances on each species’ range and identified the dominant disturbance drivers. Our results revealed an overall trend of stability in species’ distributions (85% of the pixels are considered stable by 2020 for all species) but we also identified some hot spots characterized by negative trends in probability of occurrence, mostly at the edges of each species’ latitudinal range. Additionally, we identified a steady increase in disturbance events in each species’ range by disturbance (affected range doubled by 2020, from 3.5% to 7% on average) and highlighted species-specific responses to forest disturbance drivers such as wind and fire. Overall, our study provides insights into distribution trends and disturbance patterns for the main European forest tree species. The identification of range shifts and the intensifying impacts of disturbances call for proactive conservation efforts and long-term planning to ensure the resilience and sustainability of European forests.

show abstract

Extending Estimates of Tree and Tree Species Presence-Absence through Space and Time Using Landsat Composites

Abstract: Space and Time Using Landsat Composites Etendre les estimations de la pr esence-absence d'arbres et d'esp eces d'arbres dans l'espace et le temps au moyen de compos es Landsat

Cited by 9 publications

References 85 publications

Forest tree species distribution for Europe 2000–2020: mapping potential and realized distributions using spatiotemporal machine learning

Forest tree species distribution for Europe 2000–2020: mapping potential and realized distributions using spatiotemporal machine learning

Multi-decadal trend analysis and forest disturbance assessment of European tree species: concerning signs of a subtle shift

Multi-decadal trend analysis and forest disturbance assessment of European tree species: concerning signs of a subtle shift

Contact Info

Product

Resources

About