The use of unmanned aerial vehicles (UAVs) in vegetation remote sensing allows a time-flexible and cost-effective acquisition of very high-resolution imagery. Still, current methods for the mapping ...of forest tree species do not exploit the respective, rich spatial information. Here, we assessed the potential of convolutional neural networks (CNNs) and very high-resolution RGB imagery from UAVs for the mapping of tree species in temperate forests. We used multicopter UAVs to obtain very high-resolution (<2 cm) RGB imagery over 51 ha of temperate forests in the Southern Black Forest region, and the Hainich National Park in Germany. To fully harness the end-to-end learning capabilities of CNNs, we used a semantic segmentation approach (U-net) that concurrently segments and classifies tree species from imagery. With a diverse dataset in terms of study areas, site conditions, illumination properties, and phenology, we accurately mapped nine tree species, three genus-level classes, deadwood, and forest floor (mean F1-score 0.73). A larger tile size during CNN training negatively affected the model accuracies for underrepresented classes. Additional height information from normalized digital surface models slightly increased the model accuracy but increased computational complexity and data requirements. A coarser spatial resolution substantially reduced the model accuracy (mean F1-score of 0.26 at 32 cm resolution). Our results highlight the key role that UAVs can play in the mapping of forest tree species, given that air- and spaceborne remote sensing currently does not provide comparable spatial resolutions. The end-to-end learning capability of CNNs makes extensive preprocessing partly obsolete. The use of large and diverse datasets facilitate a high degree of generalization of the CNN, thus fostering transferability. The synergy of high-resolution UAV imagery and CNN provide a fast and flexible yet accurate means of mapping forest tree species.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZRSKP
Structural analysis of forests by UAV is currently growing in popularity. Given the reduction in platform costs, and the number of algorithms available to analyze data output, the number of ...applications has grown rapidly. Forest structures are not only linked to economic value in forestry, but also to biodiversity and vulnerability issues. LiDAR remains the most promising technique for forest structural assessment, but small LiDAR sensors suitable for UAV applications are expensive and are limited to a few manufactures. The estimation of 3D-structures from two-dimensional image sequences called ‘Structure from motion’ (SfM) overcomes this limitation by photogrammetrically reconstructing point clouds similar to those rendered from LiDAR sensors. The result of these techniques in highly structured terrain strongly depends on the methods employed during image acquisition, therefore structural indices might be vulnerable to misspecifications in flight campaigns. In this paper, we outline how image overlap and ground sampling distances affect image reconstruction completeness in 2D and 3D. Higher image overlaps and coarser GSDs have a clearly positive influence on reconstruction quality. Therefore, higher accuracy requirements in the GSD must be compensated by a higher image overlap. The best results are achieved with an image overlap of > 95% and a resolution of > 5 cm. The most important environmental factors have been found to be wind and terrain elevation, which could be an indicator of vegetation density.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
In recent years, technological advances have led to the increasing use of unmanned aerial vehicles (UAVs) for forestry applications. One emerging field for drone application is forest health ...monitoring (FHM). Common approaches for FHM involve small-scale resource-extensive fieldwork combined with traditional remote sensing platforms. However, the highly dynamic nature of forests requires timely and repetitive data acquisition, often at very high spatial resolution, where conventional remote sensing techniques reach the limits of feasibility. UAVs have shown that they can meet the demands of flexible operation and high spatial resolution. This is also reflected in a rapidly growing number of publications using drones to study forest health. Only a few reviews exist which do not cover the whole research history of UAV-based FHM. Since a comprehensive review is becoming critical to identify research gaps, trends, and drawbacks, we offer a systematic analysis of 99 papers covering the last ten years of research related to UAV-based monitoring of forests threatened by biotic and abiotic stressors. Advances in drone technology are being rapidly adopted and put into practice, further improving the economical use of UAVs. Despite the many advantages of UAVs, such as their flexibility, relatively low costs, and the possibility to fly below cloud cover, we also identified some shortcomings: (1) multitemporal and long-term monitoring of forests is clearly underrepresented; (2) the rare use of hyperspectral and LiDAR sensors must drastically increase; (3) complementary data from other RS sources are not sufficiently being exploited; (4) a lack of standardized workflows poses a problem to ensure data uniformity; (5) complex machine learning algorithms and workflows obscure interpretability and hinders widespread adoption; (6) the data pipeline from acquisition to final analysis often relies on commercial software at the expense of open-source tools.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
The retention of structural elements such as habitat trees in forests managed for timber production is essential for fulfilling the objectives of biodiversity conservation. This paper seeks to ...predict tree-related microhabitats (TreMs) by close-range remote sensing parameters. TreMs, such as cavities or crown deadwood, are an established tool to quantify the suitability of habitat trees for biodiversity conservation. The aim to predict TreMs based on remote sensing (RS) parameters is supposed to assist a more objective and efficient selection of retention elements. The RS parameters were collected by the use of terrestrial laser scanning as well as unmanned aerial vehicles structure from motion point cloud generation to provide a 3D distribution of plant tissue. Data was recorded on 135 1-ha plots in Germany. Statistical models were used to test the influence of 28 RS predictors, which described TreM richness (R2: 0.31) and abundance (R2: 0.31) in moderate precision and described a deviance of 44% for the abundance and 38% for richness of TreMs. Our results indicate that multiple RS techniques can achieve moderate predictions of TreM occurrence. This method allows a more efficient and objective selection of retention elements such as habitat trees that are keystone features for biodiversity conservation, even if it cannot be considered a full replacement of TreM inventories due to the moderate statistical relationship at this stage.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
One of the most important drivers for the coexistence of plant species is the resource heterogeneity of a certain environment, and several studies in different ecosystems have supported this resource ...heterogeneity–diversity hypothesis. However, to date, only a few studies have measured heterogeneity of light and soil resources below forest canopies to investigate their influence on understory plant species richness. Here, we aim to determine (1) the influence of forest stand structural complexity on the heterogeneity of light and soil resources below the forest canopy and (2) whether heterogeneity of resources increases understory plant species richness. Measures of stand structural complexity were obtained through inventories and remote sensing techniques in 135 1‐ha study plots of temperate forests, established along a gradient of forest structural complexity. We measured light intensity and soil chemical properties on six 25 m² subplots on each of these 135 plots and surveyed understory vegetation. We calculated the coefficient of variation of light and soil parameters to obtain measures of resource heterogeneity and determined understory plant species richness at plot level. Spatial heterogeneity of light and of soil pH increased with higher stand structural complexity, although heterogeneity of soil pH did not increase in conditions of generally high levels of light availability. Increasing light heterogeneity was also associated with increasing understory plant species richness. However, light heterogeneity had no such effects in conditions where soil resource heterogeneity (variation in soil C:N ratios) was low. Our results support the resource heterogeneity–diversity hypothesis for temperate forest understory at the stand scale. Our results also highlight the importance of interaction effects between the heterogeneity of both light and soil resources in determining plant species richness.
One of the most important drivers of the coexistence of species is the resource heterogeneity of a certain environment. We measured forest structural complexity, vegetation, and resources on forest floor to test whether stand structure has an effect on the distribution of resources, and if so, whether it influences species richness. Our study revealed that forest structure creates a heterogeneous light and soil‐resource environment which then increases species richness.
Full text
Available for:
FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Spartina alterniflora (S. alterniflora) is one of the worst plant invaders in the coastal wetlands of China. Accurate and repeatable mapping of S. alterniflora invasion is essential to develop ...cost-effective management strategies for conserving native biodiversity. Traditional remote-sensing-based mapping methods require a lot of fieldwork for sample collection. Moreover, our ability to detect this invasive species is still limited because of poor spectral separability between S. alterniflora and its co-dominant native plants. Therefore, we proposed a novel scheme that uses an ensemble one-class classifier (EOCC) in combination with phenological Normalized Difference Vegetation Index (NDVI) time-series analysis (TSA) to detect S. alterniflora. We evaluated the performance of the EOCC algorithm in two scenarios, i.e., single-scene analysis (SSA) and NDVI-TSA in the core zones of Yancheng National Natural Reserve (YNNR). Meanwhile, a fully supervised classifier support vector machine (SVM) was tested in the two scenarios for comparison. With these scenarios, the crucial phenological stages and the advantage of phenological NDVI-TSA in S. alterniflora recognition were also investigated. Results indicated the EOCC using only positive training data performed similarly well with the SVM trained on complete training data in the YNNR. Moreover, the EOCC algorithm presented a more robust transferability with notably higher classification accuracy than the SVM when being transferred to a second site, without a second training. Furthermore, when combined with the phenological NDVI-TSA, the EOCC algorithm presented more balanced sensitivity–specificity result, showing slightly better transferability than it performed in the best phenological stage (i.e., senescence stage of November). The achieved results (overall accuracy (OA), Kappa, and true skill statistic (TSS) were 92.92%, 0.843, and 0.834 for the YNNR, and OA, Kappa, and TSS were 90.94%, 0.815, and 0.825 for transferability to the non-training site) suggest that our detection scheme has a high potential for the mapping of S. alterniflora across different areas, and the EOCC algorithm can be a viable alternative to traditional supervised classification method for invasive plant detection.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Forests are one of the most important terrestrial ecosystems for the protection of biodiversity, but at the same time they are under heavy production pressures. In many cases, management optimized ...for timber production leads to a simplification of forest structures, which is associated with species loss. In recent decades, the concept of retention forestry has been implemented in many parts of the world to mitigate this loss, by increasing structure in managed stands. Although this concept is widely adapted, our understanding what forest structure is and how to reliably measure and quantify it is still lacking. Thus, more insights into the assessment of biodiversity-relevant structures are needed, when aiming to implement retention practices in forest management to reach ambitious conservation goals. In this study we compare expert ratings on forest structural richness with a modern light detection and ranging (LiDAR) -based index, based on 52 research sites, where terrestrial laser scanning (TLS) data and 360° photos have been taken. Using an online survey (n = 444) with interactive 360° panoramic image viewers, we sought to investigate expert opinions on forest structure and learn to what degree measures of structure from terrestrial laser scans mirror experts’ estimates. We found that the experts’ ratings have large standard deviance and therefore little agreement. Nevertheless, when averaging the large number of participants, they distinguish stands according to their structural richness significantly. The stand structural complexity index (SSCI) was computed for each site from the LiDAR scan data, and this was shown to reflect some of the variation of expert ratings (p = 0.02). Together with covariates describing participants’ personal background, image properties and terrain variables, we reached a conditional R2 of 0.44 using a linear mixed effect model. The education of the participants had no influence on their ratings, but practical experience showed a clear effect. Because the SSCI and expert opinion align to a significant degree, we conclude that the SSCI is a valuable tool to support forest managers in the selection of retention patches.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
•Stereo WV-3 imagery is used to map standing dead trees in the Black Forest.•The POBIF approach significantly outperforms pixel- and object- based methods.•Adding CHM greatly reduces commission and ...omission errors caused by bare ground.•Vegetation indices cannot significantly improve the accuracy of SDT mapping.
Information about the distribution of standing dead trees (SDT) is essential for forest biodiversity estimation, forest disturbances monitoring, and forest management strategy planning. Although remote sensing techniques offer unique capabilities to map SDT over large areas, three major hurdles exist: (1) the sporadic distribution of SDT in the study area; (2) often poor spectral separability between SDT and bare ground in forests; (3) the prominent spectral variability within SDT due to variations in background effect and canopy illumination. To address these problems, we proposed a pixel- and object-based image fusion (POBIF) approach using very high-resolution (VHR) stereo WorldView-3 (WV-3) data. The stereo WV-3 derived spectral bands, canopy height model (CHM), vegetation indices (VIs), and texture features were used as inputs in six classification scenarios with different variable combinations. A deep learning algorithm, deep neural network (DNN), and two machine learning algorithms, support vector machine (SVM) and random forest (RF), were utilized to process the pixel-based (PB) and object-based (OB) information. All PB and OB classifiers were then combined using a stacked generalization strategy to develop the POBIF model. Comparing the six scenarios we assessed the importance of the CHM, VIs, and textures for accurate SDT mapping. As a result, we found (1) the POBIF outperformed both PB and OB methods for SDT mapping, generating notably higher F1-score (p < 0.05); (2) Adding the CHM reduced the commission and omission errors caused by bare ground and artificial surfaces, and the highest classification accuracy was achieved when combing all the WV-3 derived variables; (3) SDT covered 0.89% of the forest areas in the study area and was particularly distributed on higher and steeper north-facing (northeast and northwest) slopes. Large tracts of SDT were found in the strictly protected forests. The study highlights the potential of VHR stereo WV-3 imagery and the POBIF for SDT mapping in a temperate montane forest area with high accuracy. The map created in this study could be used for guiding field investigations and for planning management measures.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Increasingly available spaceborne sensors provide unprecedented opportunities for large‐scale, timely and continuous tree species diversity (TSD) monitoring. However, given differences in spectral ...and spatial resolutions, the choice of sensor is not always straightforward. In this work, we investigated the effects of spatial and spectral resolutions for four spaceborne sensors (RapidEye, Landsat‐8, Sentinel‐2 and PlanetScope) on TSD mapping in an area of approximately 4000 km2 within the Black Forest, Germany. We employed a random forest (RF) regression model to predict Shannon–Wiener diversity based on seven types of spectral heterogeneity metrics (texture, coefficient of variation, Rao's Q, convex hull volume, spectral angle mapper, convex hull area and spectral species diversity) and a full survey dataset from 135 one‐ha sample plots. We compared the RF model's performance across sensors and spatial resolutions. Our results demonstrated that the Sentinel‐2‐based TSD model achieved the highest accuracy (mean R2: 0.477, mean root‐mean‐square error (RMSE): 0.274). The RapidEye‐based TSD model produced lower accuracy (mean R2: 0.346, mean RMSE: 0.303), but it was better than the PlanetScope‐ and Landsat‐based TSD models. The 10 m (for Sentinel‐2 and RapidEye) and 15 m (for PlanetScope) were the best spatial resolutions for predicting TSD. The NIR band was the most favourable spectral band for predicting TSD. Texture metrics and Rao's Q outperformed the other spectral heterogeneity metrics. Our results highlighted that spaceborne optical imagery (especially Sentinel‐2) can be successfully used for large‐scale TSD mapping but that the choice of sensors can significantly affect the resulting mapping accuracy in temperate montane forests.
We investigated the effects of spatial and spectral resolutions for four spaceborne sensors (RapidEye, Landsat‐8, Sentinel‐2 and PlanetScope) on tree species diversity (TSD) mapping in an area of approximately 4000 km2 within the Black Forest, Germany. We employed a random forest regression model to predict Shannon–Wiener diversity based on seven types of spectral heterogeneity metrics (texture, coefficient of variation, Rao's Q, convex hull volume, spectral angle mapper, convex hull area and spectral species diversity) and a full survey dataset from 135 one‐ha sample plots. Our results demonstrated that the Sentinel‐2‐based TSD model achieved higher accuracy than the other satellite‐based TSD models. The 10 m (for Sentinel‐2 and RapidEye) and 15 m (for PlanetScope) were the best resolutions for predicting TSD. Texture metrics and Rao's Q outperformed the other spectral heterogeneity metrics. Our results highlighted that spaceborne optical imagery (especially Sentinel‐2) can be successfully used for large‐scale TSD mapping but that the choice of sensors can significantly affect the resulting mapping accuracy in temperate montane forests.
Full text
Available for:
FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Quantifying wind loads acting on forest trees remains a major challenge of wind-tree-interaction research. Under wind loading, trees respond with a complex motion pattern to the external forces that ...displace them from their rest position. To minimize the transfer of kinetic wind energy, crowns streamline to reduce the area oriented toward the flow. At the same time, the kinetic energy transferred to the trees is dissipated by vibrations of all aerial parts to a different degree. This study proposes a method to estimate the effective wind load acting on plantation-grown Scots pine trees. It evaluates the hypothesis that the effective wind load acting on the sample trees can be estimated using static, non-destructive pulling tests, using measurements of stem tilt under natural wind conditions and static, non-destructive pulling tests. While the analysis of wind-induced stem displacement reconstructs the temporal tree response dynamics to the effective wind load, results from the pulling tests enable the effective wind load quantification. Since wind-induced stem displacement correlates strongly with the sample trees’ diameter at breast height, the effective wind load estimation can be applied to all other trees in the studied stand for which diameter data is available. We think the method is suitable for estimating the effective wind load acting on trees whose wind-induced response is dominated by sway in the fundamental mode.