9 research outputs found
Multi-Sensor Aboveground Biomass Estimation in the Broadleaved Hyrcanian Forest of Iran
No full textPublisher Copyright: © 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.In this study, the capability of Landsat-8 (L8), Sentinel-2 (S2), Sentinel-1 (S1), and their combination was investigated for estimating aboveground biomass (AGB). A pure stand of Fagus Orientalis located in the Hyrcanian forest of Iran was selected as the study area. The performance of a parametric approach, i.e., Multiple Linear Regression (MLR) model and non-parametric approaches, i.e., k-Nearest Neighbor (k-NN), Random Forest (RF), and Support Vector Regression (SVR), were also evaluated for AGB estimations. Our results indicated that among S2 metrics, the FAPAR canopy biophysical index and NDVI index based on the red-edge band (NIR-b8a) have the highest correlation coefficient (r) of 0.420 and 0.417, respectively. The results of AGB estimation showed that a combination of S2 and S1 datasets using the k-NN algorithm had the best accuracy (R 2 of 0.57 and rRMSE of 14.68%). The best rRMSE using L8, S2, and S1 datasets was 18.95, 16.99, and 19.17% using k-NN, k-NN, and MLR algorithms, respectively. The combination of L8 with S1 dataset also improved the rRMSE relative to L8 and S1 separately by 0.96 and 1.18%, respectively. We concluded that the combination of optical data (L8 or S2) with SAR data (S1) improves the broadleaved Hyrcanian AGB estimation.Peer reviewe
Detection of Forest Windstorm Damages with Multitemporal SAR Data—A Case Study: Finland
No full textThe purpose of this study was to develop methods to localize forest windstorm damages, assess their severity and estimate the total damaged area using space-borne SAR data. The development of the methods is the first step towards an operational system for near-real-time windstorm damage monitoring, with a latency of only a few days after the storm event in the best case. Windstorm detection using SAR data is not trivial, particularly at C-band. It can be expected that a large-area and severe windstorm damage may affect backscatter similar to clear cutting operation, that is, decrease the backscatter intensity, while a small area damage may increase the backscatter of the neighboring area, due to various scattering mechanisms. The remaining debris and temporal variation in the weather conditions and possible freeze–thaw transitions also affect observed backscatter changes. Three candidate windstorm detection methods were suggested, based on the improved k-nn method, multinomial logistic regression and support vector machine classification. The approaches use multitemporal ESA Sentinel-1 C-band SAR data and were evaluated in Southern Finland using wind damage data from the summer 2017, together with 27 Sentinel-1 scenes acquired in 2017 and other geo-referenced data. The stands correctly predicted severity category corresponded to 79% of the number of the stands in the validation data, and already 75% when only one Sentinel-1 scene after the damage was used. Thus, the damaged forests can potentially be localized with proposed tools within less than one week after the storm damage. In this study, the achieved latency was only two days. Our preliminary results also indicate that the damages can be localized even without separate training data
Detection of forest windstorm damages with multitemporal sar data—a case study:Finland
No full textThe purpose of this study was to develop methods to localize forest windstorm dam-ages, assess their severity and estimate the total damaged area using space-borne SAR data. The development of the methods is the first step towards an operational system for near-real-time windstorm damage monitoring, with a latency of only a few days after the storm event in the best case. Windstorm detection using SAR data is not trivial, particularly at C-band. It can be expected that a large-area and severe windstorm damage may affect backscatter similar to clear cutting operation, that is, decrease the backscatter intensity, while a small area damage may increase the backscatter of the neighboring area, due to various scattering mechanisms. The remaining debris and temporal variation in the weather conditions and possible freeze–thaw transitions also affect observed backscatter changes. Three candidate windstorm detection methods were suggested, based on the improved k-nn method, multinomial logistic regression and support vector machine classification. The approaches use multitemporal ESA Sentinel-1 C-band SAR data and were evaluated in Southern Finland using wind damage data from the summer 2017, together with 27 Sentinel-1 scenes acquired in 2017 and other geo-referenced data. The stands correctly predicted severity category corresponded to 79% of the number of the stands in the validation data, and already 75 % when only one Sentinel-1 scene after the damage was used. Thus, the damaged forests can potentially be localized with proposed tools within less than one week after the storm damage. In this study, the achieved latency was only two days. Our preliminary results also indicate that the damages can be localized even without separate training data.</p
An integrated approach combining bi-temporal airborne laser scanning and X-ray microdensitometry in assessing wood properties
No full textInformation on wood properties across trees and stands is needed to support silviculture and wood procurement. Wood properties of standing trees are usually measured by destructive sampling limiting the number of observations that can be collected across a range of forest structural and environmental conditions. In contrast, airborne laser scanning (ALS), with its capability to characterize tree crowns and their increment over time, could provide a non-destructive approach for assessing wood properties. This study aimed at relating ALS-derived mean annual increments in tree height and crown dimensions between 2009 (T1) and 2023 (T2) to X-ray microdensitometry-measured mean ring width (RWmean-tree) and basal-area weighted mean wood density (WDmean-tree) formed during the same period. The experimental design comprised 257 Scots pines (Pinus sylvestris L.) and 142 Norway spruces (Picea abies (L.) Karst.) across 59 sample plots representing varying boreal forest conditions. As per our investigations, the mean annual increment in tree height (ΔHmean_tree) represented the strongest correlations with RWmean_tree for both tree species (r = 0.43–0.44) and a weak but statistically significant correlation with WDmean_tree for Norway spruces only (r = −0.17). When aggregating individual tree observations for plot-level, ΔHmean_plot exhibited moderate correlations (r = 0.47–0.48) with RWmean_plot for both species. WDmean_plot of Scots pines showed a correlation of 0.36 with the averaged mean annual increments of crown surface area. However, none of the metrics were significant for WDmean_plot of Norway spruces. By utilizing the linear-mixed effect model 40–41 % of the variations in RWmean_tree of Scots pines and Norway spruces could be explained when accounting for the variability between sample plots. Based on our study, it appears that some of the variation in wood properties, particularly in ring width, can be captured using bi-temporal ALS measurements. However, assess
Characterizing the competitive stress of individual trees using point clouds
No full textThe competitive stress of individual trees can be quantified by considering their positions and dimensions such as diameter at breast height (dbh) and height with respect to their neighbor trees. However, measurements of these attributes in the field limit the number of trees and stands that can be assessed with given resources. In recent years, terrestrial laser scanning (TLS) and airborne laser scanning (ALS) data have become prominent in characterizing three-dimensional forest structures. These data could also provide efficient and reliable tools to assess the competitive stress of trees within a stand. Therefore, we aimed to investigate the capability of TLS and low-altitude ALS in characterizing the competitive stress affecting individual trees in boreal forests. We compared: a) an object-based approach that quantified competition through the identification and characterization of competing neighbor trees from the TLS and ALS point clouds, and b) a point cloud-based approach where the presence of point cloud structures representing competitive vegetation around a target tree was considered. Accordingly, three object-based competition indices (CIs) utilizing dbh (CIdbh), height (CIH), and maximum crown diameter (CIMCD) as weights were calculated using the Hegyi equation. For the point cloud-based approach, the canopy density index (CDI), and the competitive pressure index (CPI) were derived using an upside-down search cone set at 60 % relative tree height, while the CICylinder was calculated by counting the number of voxels occupied by the competitive vegetation inside a fixed-radius cylinder. These laser scanning-based CIs were assessed against in situ-based CIs where dbh and height were used as weights in the Hegyi equation. The results showed that the object-based CIs were more correlated (r = 0.33–0.48, p-value < 0.001) with the in situ-based CIs in comparison with the point cloud-based CIs (r = −0.22–0.37). The object-based CIs showed a high correlation (r = 0.65–0.71, p-value < 0.001) when compared between TLS and ALS, while a greater variation was observed for the point cloud-based CIs (r = 0.29–0.53, p-value < 0.001). Tree detection rate and the number of neighboring trees in the field affected how well the CIs derived from the TLS and ALS data were in line with the in situ-based CIs, especially when the competitive stress was assessed using the object-based CIs. In conclusion, the object-based CIs derived using TLS and ALS provided consistent characterization of competition in managed boreal forests compared to the in situ-based CIs. While TLS is ideal for small-scale assessments, low-altitude ALS offers a rather similar capacity for assessing competition but with broader coverage. In complex forest structures, reliable tree detection is essential to avoid underestimating the competitive stress of trees
Feasibility of Bi-Temporal Airborne Laser Scanning Data in Detecting Species-Specific Individual Tree Crown Growth of Boreal Forests
No full textThe tree crown, with its functionality of assimilation, respiration, and transpiration, is a key forest ecosystem structure, resulting in high demand for characterizing tree crown structure and growth on a spatiotemporal scale. Airborne laser scanning (ALS) was found to be useful in measuring the structural properties associated with individual tree crowns. However, established ALS-assisted monitoring frameworks are still limited. The main objective of this study was to investigate the feasibility of detecting species-specific individual tree crown growth by means of airborne laser scanning (ALS) measurements in 2009 (T1) and 2014 (T2). Our study was conducted in southern Finland over 91 sample plots with a size of 32 × 32 m. The ALS crown metrics of width (WD), projection area (A2D), volume (V), and surface area (A3D) were derived for species-specific individually matched trees in T1 and T2. The Scots pine (Pinus sylvestris), Norway spruce (Picea abies (L.) H. Karst), and birch (Betula sp.) were the three species groups that studied. We found a high capability of bi-temporal ALS measurements in the detection of species-specific crown growth (Δ), especially for the 3D crown metrics of V and A3D, with Cohen’s D values of 1.09–1.46 (p-value < 0.0001). Scots pine was observed to have the highest relative crown growth (rΔ) and showed statistically significant differences with Norway spruce and birch in terms of rΔWD, rΔA2D, rΔV, and rΔA3D at a 95% confidence interval. Meanwhile, birch and Norway spruce had no statistically significant differences in rΔWD, rΔV, and rΔA3D (p-value < 0.0001). However, the amount of rΔ variability that could be explained by the species was only 2–5%. This revealed the complex nature of growth controlled by many biotic and abiotic factors other than species. Our results address the great potential of ALS data in crown growth detection that can be used for growth studies at large scales
Terrestrial Laser Scanning in Assessing the Effect of Different Thinning Treatments on the Competition of Scots Pine (Pinus sylvestris L.) Forests
No full textThinning is a forest management activity that regulates the competition between the trees within a forest. However, the effect of different thinning treatments on competition is largely unexplored, especially because of the difficulty in measuring crown characteristics. This study aimed to investigate how different type and intensity thinning treatments affect the stem- and crown-based competition of trees based on terrestrial laser scanning (TLS) point clouds. The research was conducted in three study sites in southern Finland where the Scots pine (Pinus sylvestris L.) is the dominant tree species. Nine rectangular sample plots of varying sizes (1000 m2 to 1200 m2) were established within each study site, resulting in 27 sample plots in total. The experimental design of each study site included two levels of thinning intensities and three thinning types, resulting in six different thinning treatments. To assess the competition between the trees, six distance-dependent competition indices were computed for each tree. The indices were based on diameter at breast height (DBH) (CIDBH), height (CIH), maximum crown diameter (CIMCD), crown projection area (CICA), crown volume (CICV), and crown surface area (CICS). The results showed that for both moderate and intensive intensities, the competition decrease was 45.5–82.5% for thinning from below, 15.6–73.6% for thinning from above, and 12.8–66.8% for systematic thinning when compared with control plots. In most cases, the crown- and stem-based metrics were affected by thinning treatments significantly when compared with control plots at a 95% confidence interval. Moreover, moderate from-below and from-above thinning showed no statistical difference with each other in both crown- and stem-based competition indices except for CIDBH (p-value ≤ 0.05). Our results confirm the great potential of TLS point clouds in quantifying stem- and crown-based competition between trees, which could be beneficial for enhancing ecological knowledge on how trees grow in response to competition
Enhancing Hyrcanian Forest Height and Aboveground Biomass Predictions:A Synergistic Use of TanDEM-X InSAR Coherence, Sentinel-1, and Sentinel-2 Data
No full textForest height (FH) is an important driver for aboveground biomass (AGB) that can be obtained using interferometric SAR (InSAR). However, the limited access to the quad-polarimetric data or high-accuracy terrain model makes FH retrieval a challenging task. This study aimed to retrieve FH and further predict AGB by combining TanDEM-X InSAR coherence, Sentinel-1 (S-1), and Sentinel-2 (S-2) data. A total of 125 sample plots with a size of 900 m2 were established in a broadleaved forest of Kheyroud, Iran. The Linear and Sinc models obtained by simplification of the Random Volume over Ground (RVoG) model were used for deriving FHLin and FHSinc. Further investigation was conducted when S-1 and S-2 features including backscatters and multispectral information were added to FH predictions. Using the abovementioned datasets and FH as an additional predictor, AGB was also predicted. K-nearest neighbor (k-NN), random forest (RF), and support vector regression (SVR) were employed for prediction. Lorey&#x0027;s mean height and AGB at sample plots were used in the accuracy assessment. Using the SVR method and synergy of FHSinc, S-1, and S-2 features, the FH prediction was improved (FHimp) with RMSE of 3.18 m and R2 &#x003D; 0.59. The AGB prediction with RF and the combination of S-1 and S-2 features resulted in RMSE &#x003D; 62.88 Mg.ha-1 (19.77&#x0025;) that was improved to RMSE &#x003D; 51.27 Mg.ha-1 (16.12&#x0025;) when FHimp included. This study highlighted the capability of TanDEM-X InSAR coherence with certain geometry for FH prediction. Also, the importance of FH in AGB predictions can stimulate further attempts aiming at higher spatiotemporal accuracies.</p
Understanding tree growth dependencies using multisensorial point clouds
No full textAbstract
Individual tree crowns are the primary interface with the environment and closely relate to tree growth, yet accurately characterizing them remains challenging. This study aimed to understand how individual tree stem volume growth (ΔV) depends on crown metrics both at the beginning of the monitoring period (T1_C) and on their changes over time (ΔC), using close-range multisensorial point clouds obtained from terrestrial and airborne laser scanning (TLS and ALS). Data were collected from 22 sample plots in boreal forests of Finland in 2014 (T1) and 2021 (T2). Spearman’s rank correlation coefficient (ρ) was employed to assess the relationships between ΔV and crown metrics across different tree species. Additionally, Random Forest regression (RF) was applied to explore the relative importance of these metrics in explaining ΔV. A strong correlation (ρ = 0.60–0.63) was found between ΔV of Scots pine (Pinus sylvestris L.) and crown metrics, including volume (T1_CV), perimeter (T1_CP), projection area (T1_CA2D), and top height (T1_CHmax). In contrast, ΔV of Norway spruce (Picea abies (L.) H. Karst.) showed only weak correlations, with the best metrics being crown base height (T1_CHmin), T1_CV, and its change (ΔCV) (ρ = 0.32–0.38). For birches (Betula sp.), ΔV also exhibited weak correlations (ρ = 0.27–0.34), mainly with crown surface area (T1_CA3D), ΔCV, and T1_CHmax. RF analyses further highlighted species-specific drivers of ΔV. Scots pine with the most important metric of T1_CHmax explained 50% of variation in ΔV. However, ΔCV was the most important metric in explaining ΔV of Norway spruce and birch, with explained variability of 20% and 6%, respectively. In conclusion, this study demonstrated that multisensorial point clouds provide an effective approach to analyze the relationship between ΔV and tree crown structure. Nevertheless, challenges persist in consistently measuring various crown metrics over time and distinguishing actual changes from measurement errors.Abstract
Individual tree crowns are the primary interface with the environment and closely relate to tree growth, yet accurately characterizing them remains challenging. This study aimed to understand how individual tree stem volume growth (ΔV) depends on crown metrics both at the beginning of the monitoring period (T1_C) and on their changes over time (ΔC), using close-range multisensorial point clouds obtained from terrestrial and airborne laser scanning (TLS and ALS). Data were collected from 22 sample plots in boreal forests of Finland in 2014 (T1) and 2021 (T2). Spearman’s rank correlation coefficient (ρ) was employed to assess the relationships between ΔV and crown metrics across different tree species. Additionally, Random Forest regression (RF) was applied to explore the relative importance of these metrics in explaining ΔV. A strong correlation (ρ = 0.60–0.63) was found between ΔV of Scots pine (Pinus sylvestris L.) and crown metrics, including volume (T1_CV), perimeter (T1_CP), projection area (T1_CA2D), and top height (T1_CHmax). In contrast, ΔV of Norway spruce (Picea abies (L.) H. Karst.) showed only weak correlations, with the best metrics being crown base height (T1_CHmin), T1_CV, and its change (ΔCV) (ρ = 0.32–0.38). For birches (Betula sp.), ΔV also exhibited weak correlations (ρ = 0.27–0.34), mainly with crown surface area (T1_CA3D), ΔCV, and T1_CHmax. RF analyses further highlighted species-specific drivers of ΔV. Scots pine with the most important metric of T1_CHmax explained 50% of variation in ΔV. However, ΔCV was the most important metric in explaining ΔV of Norway spruce and birch, with explained variability of 20% and 6%, respectively. In conclusion, this study demonstrated that multisensorial point clouds provide an effective approach to analyze the relationship between ΔV and tree crown structure. Nevertheless, challenges persist in consistently measuring various crown metrics over time and distinguishing actual changes from measurement errors
