Global Forest Types Based on Climatic and Vegetation Data

Forest types are generally identified using vegetation or land-use types. However, vegetation classifications less frequently consider the actual forest attributes within each type. To address this in an objective way across different regions and to link forest attributes with their climate, we aimed to improve the distribution of forest types to be more realistic and useful for biodiversity preservation, forest management, and ecological and forestry research. The forest types were classified using an unsupervised cluster analysis method by combining climate variables with normalized difference vegetation index (NDVI) data. Unforested regions were masked out to constrict our study to forest type distributions, using a 20% tree cover threshold. Descriptive names were given to the defined forest types based on annual temperature, precipitation, and NDVI values. Forest types had distinct climate and vegetation characteristics. Regions with similar NDVI values, but with different climate characteristics, which would be merged in previous classifications, could be clearly distinguished. However, small-range forest types, such as montane forests, were challenging to differentiate. At macroscale, the resulting forest types are largely consistent with land-cover types or vegetation types defined in previous studies. However, considering both potential and current vegetation data allowed us to create a more realistic type distribution that differentiates actual vegetation types and thus can be more informative for forest managers, conservationists, and forest ecologists. The newly generated forest type distribution is freely available to download and use for non-commercial purposes as a GeoTIFF file via doi: 10.13140/RG.2.2.19197.90082).

Synthesizing Disparate LiDAR and Satellite Datasets through Deep Learning to Generate Wall-to-Wall Regional Inventories for the Complex, Mixed-Species Forests of the Eastern United States

Light detection and ranging (LiDAR) has become a commonly-used tool for generating remotely-sensed forest inventories. However, LiDAR-derived forest inventories have remained uncommon at a regional scale due to varying parameters among LiDAR data acquisitions and the availability of sufficient calibration data. Here, we present a model using a 3-D convolutional neural network (CNN), a form of deep learning capable of scanning a LiDAR point cloud, combined with coincident satellite data (spectral, phenology, and disturbance history). We compared this approach to traditional modeling used for making forest predictions from LiDAR data (height metrics and random forest) and found that the CNN had consistently lower uncertainty. We then applied the CNN to public data over six New England states in the USA, generating maps of 14 forest attributes at a 10 m resolution over 85% of the region. Aboveground biomass estimates produced a root mean square error of 36 Mg ha−1 (44%) and were within the 97.5% confidence of independent county-level estimates for 33 of 38 or 86.8% of the counties examined. CNN predictions for stem density and percentage of conifer attributes were moderately successful, while predictions for detailed species groupings were less successful. The approach shows promise for improving the prediction of forest attributes from regional LiDAR data and for combining disparate LiDAR datasets into a common framework for large-scale estimation.

Hierarchical Bayesian Small Area Estimation Using Weakly Informative Priors in Ecologically Homogeneous Areas of the Interior Western Forests

The U.S. Forest Inventory and Analysis Program (FIA) collects inventory data on and computes estimates for many forest attributes to monitor the status and trends of the nation's forests. Increasingly, FIA needs to produce estimates in small geographic and temporal regions. In this application, we implement area level hierarchical Bayesian (HB) small area estimators of several forest attributes for ecosubsections in the Interior West of the US. We use a remotely-sensed auxiliary variable, percent tree canopy cover, to predict response variables derived from ground-collected data such as basal area, biomass, tree count, and volume. We implement four area level HB estimators that borrow strength across ecological provinces and sections and consider prior information on the between-area variation of the response variables. We compare the performance of these HB estimators to the area level empirical best linear unbiased prediction (EBLUP) estimator and to the industry-standard post-stratified (PS) direct estimator. Results suggest that when borrowing strength to areas which are believed to be homogeneous (such as the ecosection level) and a weakly informative prior distribution is placed on the between-area variation parameter, we can reduce variance substantially compared the analogous EBLUP estimator and the PS estimator. Explorations of bias introduced with the HB estimators through comparison with the PS estimator indicates little to no addition of bias. These results illustrate the applicability and benefit of performing small area estimation of forest attributes in a HB framework, as they allow for more precise inference at the ecosubsection level.

Application of allometric systems for compatible area-based LiDAR-assisted estimation in the Province of Nova Scotia

Light detection and ranging (LiDAR) is used to estimate tree, stand, and forest characteristics across large geographic areas. In the province of Nova Scotia, an enhanced forest inventory (EFI) was developed to provide high-resolution spatial forest inventory estimates across the landscape. For various forest attributes, independent LiDAR-based relationships were built leading to mathematical and biological inconsistency among forest attribute estimates. A systems approach, composed of allometric equations describing the relationships between volume per unit area, Lorey’s average height, basal area, quadratic mean diameter, and density, is developed to address these inconsistencies. Previous results showed that applying the systems approach provided reasonable and compatible estimates and eliminated inconsistency issues among forest attributes. This study evaluates application of the systems approach applied to eastern Nova Scotia using field data from a network of permanent sample plots and recent LiDAR acquisitions. The independent EFI estimates had inconsistencies of greater than 100% for basal area and implied stand-level form factor. These inconsistencies were eliminated using the systems approach. Results show that the systems approach can be scaled to larger landscape areas and that long-term field data can be leveraged to fit the allometric systems producing mathematically and biologically consistent estimates.

Comparison of Low-Cost Commercial Unpiloted Digital Aerial Photogrammetry to Airborne Laser Scanning across Multiple Forest Types in California, USA

Science-based forest management requires quantitative estimation of forest attributes traditionally collected via sampled field plots in a forest inventory program. Three-dimensional (3D) remotely sensed data such as Light Detection and Ranging (lidar), are increasingly utilized to supplement and even replace field-based forest inventories. However, lidar remains cost prohibitive for smaller areas and repeat measurements, often limiting its use to single acquisitions of large contiguous areas. Recent advancements in unpiloted aerial systems (UAS), digital aerial photogrammetry (DAP) and high precision global positioning systems (HPGPS) have the potential to provide low-cost time and place flexible 3D data to support forest inventory and monitoring. The primary objective of this study was to assess the ability of low-cost commercial off the shelf UAS DAP and HPGPS to create accurate 3D data and predictions of key forest attributes, as compared to both lidar and field observations, in a wide range of forest conditions in California, USA. A secondary objective was to assess the accuracy of nadir vs. off-nadir UAS DAP, to determine if oblique imagery provides more accurate 3D data and forest attribute predictions. UAS DAP digital terrain models (DTMs) were comparable to lidar DTMS across most sites and nadir vs. off-nadir imagery collection (R2 = 0.74–0.99), although model accuracy using off-nadir imagery was very low in mature Douglas-fir forest (R2 = 0.17) due to high canopy density occluding the ground from the image sensor. Surface and canopy height models were shown to have less agreement to lidar (R2 = 0.17–0.69), with off-nadir imagery surface models at high canopy density sites having the lowest agreement with lidar. UAS DAP models predicted key forest metrics with varying accuracy compared to field data (R2 = 0.53–0.85), and were comparable to predictions made using lidar. Although lidar provided more accurate estimates of forest attributes across a range of forest conditions, this study shows that UAS DAP models, when combined with low-cost HPGPS, can accurately predict key forest attributes across a range of forest types, canopies densities, and structural conditions.

Effects of topographic variability and forest attributes on fine-scale soil fertility in late-secondary succession of Atlantic Forest

Background Understanding how soil fertility changes due to topographical conditions and forest attributes is an essential premise for local-scale forest management practices. We evaluated the effects of topographic variables and forest attributes on soil fertility along a local topographical gradient in a Brazilian Atlantic Forest. We hypothesised that soil fertility is positively affected by topographic variability and forest attributes (structure and diversity). We used tree species richness, composition, abundance, and aboveground biomass as forest attributes. We analysed two 1-ha forest patches with contrasting topographical conditions. We used different linear mixed effects models (LMMs) to test the main effects of different forest attributes and topography variables on soil fertility. Results The results showed that higher topographic variability determines soil fertility along a fine-scale gradient. The first two axes of the PCA explained 66.8% of the variation in soil data, with the first axis (PCA1) explaining 49.6% of the variation in soil data and positively correlating with fertility-related soil properties. The second axis (PCA2) explained 17.2% of the variation in topographical data and positively correlated with convexity (the elevation of a plot minus the average elevation of all immediate neighbour plots) and elevation. Our best models showed that topographic variables (elevation and convexity) are the main predictors that affect fine-scale soil fertility. Conclusions Our study demonstrates that the topographic variability, mainly elevation and convexity, determines fine-scale soil fertility in an Atlantic Forest. These results advance our understanding that context-dependent conditions based on topography and soil properties have a high variability at a fine scale, which can influence variations in forest attributes (i.e., species distribution, diversity and structure of tree communities). In addition, the information generated in this research may be important for planning forest restoration activities (passive and active) based on the high variability of environmental variables at a fine scale.

Effect of Field Sample Size on Large-Scale Subtropic Forest Inventory Attribute Estimation Based on Airborne Laser Scanner Data

Exploring the effect of the sample size on the estimation accuracy of airborne LiDAR forest attributes in a large-scale area can help in optimizing the technical application scheme of operational ALS-based large-scale forest stand inventories. In our study, sample datasets composed of different sample plots were constructed by repeated sampling from 1003 sample plots in a subtropical study area covering 2376 × 103 km2. Sixteen multiplicative power models were built in each forest type consisting of four forest attributes. Through these models, the variations of standard deviation (SD) and coefficient of variation (CV) of R2 and rRMSE of forest attribute estimation models for different quantity levels of sample plots were also analyzed. The results showed that, first, when the sample size increased from 30 to the top limit, the SD of the forest attributes and LiDAR variables showed a decreasing trend. Second, as the sample size increased, the rRMSE of the 16 forest attribute estimation models gradually decreased, while the R2 gradually increased. Third, when the sample size was small, both the SD of R2 and rRMSE of the models were large, and the SD of R2 and rRMSE gradually decreased as the sample size increased. In 50 models conducted for each attribute at the same sample size, for the mean standard deviations of forest attributes, the ten best performing models were lower than those of the total 50 models, and the worst ten models were the opposite. When the sample size increased, the accuracy of each forest attribute estimation model for each forest type gradually improved. The variation of forest attributes and the LiDAR variable of the construction model are critical factors that affect the model’s accuracy. To efficiently apply airborne LiDAR in order to survey large-scale subtropical forest resources, the sample size of the Chinese fir forest, pine forest, eucalyptus forest, and broad-leaved forest should be 110, 80, 85, and 70, respectively.

Assessing Environmental Criteria to Support Forest Connectivity

Establishing forest connection in landscapes under urban sprawl is essential for maintaining the ecological processes and ensuring biodiversity conservation. However, the major challenge is incorporated the ecological network in the land-use/land-cover planning. This way, the main objective of the study was the evaluation of environmental criteria for prioritizing areas to obtain forest functional connectivity in a landscape subject to the urban sprawl. The second objective was to understand how the criteria are associated with the structural forest attributes represented by traditional landscape ecology metrics. The criteria were defined through the literature review, representing the landscape characteristics as the topographic, conflicts, and biotics. The metrics used to characterize the forest structure were perimeter, shape index, and distance to the nearest neighbor. They were generated to a selected group of forest remnants, which represent the landscape forest structure. Sampling the criteria and forest fragments maps (i.e., different maps representing the metrics-values) through the hexagon network, we assessed how the criteria are associated with the structural forest attributes. The statistical analysis used to evaluate these sampled values were The Moran Global (Moran I) and Moran Local (LISA). We obtained that the urban expansion process is diffuse, although it does not occur randomly in our landscape. The criteria slope, TWI, distance from drainage network, distance from highways, distance from the low-density urban area, and distance from forest patches have characteristics that support this process. Furthermore, our results indicated a spatial autocorrelation among metrics and after, among metrics and these criteria. Also, we obtained that the external influences on the fragments did not occur randomly and that the criteria act on the landscape. This way, through these criteria, we can identify regions where it is possible to have the persistence of forest fragments, even though in places under the impact of urban sprawl.

The Road to Operationalization of Effective Tropical Forest Monitoring Systems

The urgency to preserve tropical forest remnants has encouraged the development of remote sensing tools and techniques to monitor diverse forest attributes for management and conservation. State-of-the-art methodologies for mapping and tracking these attributes usually achieve accuracies greater than 0.8 for forest cover monitoring; r-square values of ~0.5–0.7 for plant diversity, vegetation structure, and plant functional trait mapping, and overall accuracies of ~0.8 for categorical maps of forest attributes. Nonetheless, existing operational tropical forest monitoring systems only track single attributes at national to global scales. For the design and implementation of effective and integrated tropical forest monitoring systems, we recommend the integration of multiple data sources and techniques for monitoring structural, functional, and compositional attributes. We also recommend its decentralized implementation for adjusting methods to local climatic and ecological characteristics and for proper end-user engagement. The operationalization of the system should be based on all open-source computing platforms, leveraging international support in research and development and ensuring direct and constant user engagement. We recommend continuing the efforts to address these multiple challenges for effective monitoring.