Methodology for generating a global forest management layer

Remote Sensing, 2022

Forest canopy cover (FCC) is an important ecological parameter of forest ecosystems, and is correlated with forest characteristics, including plant growth, regeneration, biodiversity, light regimes, and hydrological properties. Here, we present an approach of combining Sentinel-2 data, high-resolution aerial images, and machine learning (ML) algorithms to model FCC in the Hyrcanian mixed temperate forest, Northern Iran. Sentinel-2 multispectral bands and vegetation indices were used as variables for modeling and mapping FCC based on UAV ground truth to a wider spatial extent. Random forest (RF), support-vector machine (SVM), elastic net (ENET), and extreme gradient boosting (XGBoost) were the ML algorithms used to learn and generalize on the remotely sensed variables. Evaluation of variable importance indicated that vegetation indices including NDVI, NDVI-A, NDRE, and NDI45 were the dominant predictors in most of the models. Model accuracy estimation results showed that among the te...

European journal of theoretical and applied sciences, 2024

Sustainable forest management necessitates the mapping and estimation of forest stand attributes such as density, volume, basal area, and aboveground biomass. This study was conducted to explore the potential of geographic information systems (GIS), remote sensing, machine learning, and field inventories to estimate the forest stand volume of natural and plantation forests within watersheds in the Abra River Basin. The common machine learning regression techniques, which are random forest (RF), k-nearest neighbors (KNN), and support vector machines (SVM), were used to model and predict forest stand volume. The validation of the three machine learning methods showed that the best model to estimate and map forest stand volume is the RF algorithm (R 2 = 0.42, RMSE = 0.40 m 3 /plot, MAE = 0.31 m 3 /plot). Topographic variables such as the Digital Elevation Model (DEM) and the spectral band Near Infrared (NIR) were the most important variables in predicting forest stand volume. The estimated forest stand volume using the RF model ranged from 33 to 115 m 3 /ha, with a mean of 59 m 3 /ha. The results of this study revealed that forest volume can be measured using freely available satellite data and machine learning techniques.

Environmental Conservation, 2022

Selective logging in tropical forests may lead to deforestation and forest degradation, so accurate mapping of it will assist in forest restoration, among other ecological applications. This study aimed to track canopy tree loss due to illegal logging of the important hardwood tree Ocotea usambarensis in a closed-canopy submontane tropical forest by evaluating the mapping potential of the very-high-resolution WorldView-3 multispectral dataset using random forest (RF) and support vector machine (SVM) with radial basis function kernel classifiers. The results show average overall accuracies of 92.3 ± 2.6% and 94.0 ± 2.1% for the RF and SVM models, respectively. Average kappa coefficients were 0.88 ± 0.03 for RF and 0.90 ± 0.02 for SVM. The user's and producer's accuracies for both classifiers were in the range of 84-100%. This study further indicates that vegetation indices derived from bands 5 and 6 helped detect canopy gaps in the study area. Both variable importance measurement in the RF algorithm and pairwise feature selection proved useful in identifying the most pertinent variables in the classification of canopy gaps. These findings could allow forest managers to improve methods of detecting canopy gaps at larger scales using remote sensing data and relatively little additional fieldwork.

Remote Sensing

The assessment of changes in forest coverage is crucial for managing protected forest areas, particularly in the face of climate change. This study monitored forest cover dynamics in a 6535 ha mountain area located in north-west Romania as part of the Apuseni Natural Park from 2003 to 2019. Two approaches were used: vectorization from orthophotos and Google Earth images (in 2003, 2005, 2009, 2012, 2014, 2016, 2017, and 2019) and satellite imagery (Landsat 5 TM, 7 ETM, and 8 OLI) pre-processed to Surface Reflectance (SR) format from the same years. We employed four standard classifiers: Support Vector Machine (SVM), Random Forest (RF), Maximum Likelihood Classification (MLC), Spectral Angle Mapper (SAM), and three combined methods: Linear Spectral Unmixing (LSU) with Natural Breaks (NB), Otsu Method (OM) and SVM, to extract and classify forest areas. Our study had two objectives: 1) to accurately assess changes in forest cover over a 17-year period and 2) to determine the most effici...

Remote Sensing

Dilijan National Park is one of the most important national parks of Armenia, established in 2002 to protect its rich biodiversity of flora and fauna and to prevent illegal logging. The aim of this study is to provide first, a mapping of forest degradation and deforestation, and second, of land cover/land use changes every 5 years over a 28-year monitoring cycle from 1991 to 2019, using Sentinel-2 and Landsat time series and Machine Learning methods. Very High Spatial Resolution imagery was used for calibration and validation purposes of forest density modelling and related changes. Correlation coefficient R2 between forest density map and reference values ranges from 0.70 for the earliest epoch to 0.90 for the latest one. Land cover/land use classification yield good results with most classes showing high users’ and producers’ accuracies above 80%. Although forest degradation and deforestation which initiated about 30 years ago was restrained thanks to protection measures, anthropo...

Scientific Data, 2023

State-of-the-art cloud computing platforms such as Google Earth Engine (GEE) enable regionalto-global land cover and land cover change mapping with machine learning algorithms. However, collection of high-quality training data, which is necessary for accurate land cover mapping, remains costly and labor-intensive. To address this need, we created a global database of nearly 2 million training units spanning the period from 1984 to 2020 for seven primary and nine secondary land cover classes. Our training data collection approach leveraged GEE and machine learning algorithms to ensure data quality and biogeographic representation. We sampled the spectral-temporal feature space from Landsat imagery to efficiently allocate training data across global ecoregions and incorporated publicly available and collaborator-provided datasets to our database. To reflect the underlying regional class distribution and post-disturbance landscapes, we strategically augmented the database. We used a machine learning-based cross-validation procedure to remove potentially mis-labeled training units. Our training database is relevant for a wide array of studies such as land cover change, agriculture, forestry, hydrology, urban development, among many others.

2021 Second International Conference on Intelligent Data Science Technologies and Applications (IDSTA), 2021

Due to the urgency of the Climate Change phenomenon, it has become important to estimate the rate of deforestation in various countries since forests are essential for oxygen generation. The advent of popular machine learning algorithms, such as those that are applied to the field of computer vision, has led to the use of approaches, such as fully convolutional neural networks (FCN), for performing semantic segmentation on satellite images to determine forested and non-forested areas. However, these models tend to be computationally intensive and, even with the advancement of specialized hardware such as GPU's (Graphical Processing Units), these approaches can still be quite costly especially for Small Island Developing States (SIDS) which are dis-proportionally affected by Climate Change. We consider the use of less computationally intensive approaches such as, logistic regression, linear support vector machine and Naive Bayes to achieve similar performance to a FCN but at a much lower cost. We perform semantic segmentation on satellite images to determine the percentage of forested areas and use this information to determine the rate of change of deforestation over a period of time. We compare the performance, computing requirements, storage requirements and robustness of different Machine Learning techniques.

Remote Sensing of Environment, 2021

The southeastern United States is unique in terms of both the intensity and scale of forest management, which includes substantial thinning and other forms of harvesting. Because thinning is not a land use transition, and the disturbance signal is relatively subtle compared to a clear cut, there is a dearth of studies that attempt to detect thinning over large areas. Our goal was to detect pine thins as an indicator of active forest management using Landsat data. Areas which undergo thinning are indicative of active forest management in the region. Our approach uses a machine learning method which combines first-order harmonics and metrics from 3-year Fourier regression of Landsat time series stacks, layers from the Global Forest Change product, and other vetted national products into a random forests model to classify forest thins in the southeastern US. Forest Harvest Inspection Records for Virginia were used for training and validation. Models were successful separating thins from clear cuts and non-harvested pines (overall accuracy 86%, clear cut accuracy 90%, thin accuracy 83% for a simplified 10-predictor variable model). Examination of variable importance illustrates the physical meaning behind the models. The curve fit statistics (R 2 or RMSE) of the NDVI, Pan, and SWIR1 harmonic curve fits, which are an indication of a departure from typical vegetation phenology caused by thinning or other disturbances, were consistently among the top predictors. The harmonic regression constant, sine and cosine from the Landsat 8 panchromatic band were also important. These describe the visible reflectance (500-680 nm) phenology over the time period at a high spatial resolution (15 m). The Loss Year from the Global Forest Change product, which is an indication of stand replacing disturbance, was also consistently among the most important variables in the classifiers. High performance computing, such as Google Earth Engine, and analysis-ready data are important for this approach. This work has importance for quantification of actively managed forests in a region of the world where production forestry is the dominant land disturbance signal and a significant economic engine.

2023

The contribution of forests to carbon storage and biodiversity conservation highlights the need for accurate forest height and biomass mapping and monitoring. In France, forests are managed mainly by private owners and divided into small stands, requiring 10 to 50 m spatial resolution data to be correctly separated. Further, 35 % of the French forest territory is covered by mountains and Mediterranean forests which are managed very extensively. In this work, we used a deep-learning model based on multi-stream remote sensing measurements (NASA's GEDI LiDAR mission and ESA's Copernicus Sentinel 1 & 2 satellites) to create a 10 m resolution canopy height map of France for 2020 (FORMS-H). In a second step, with allometric equations fitted to the French National Forest Inventory (NFI) plot data, we created a 30 m resolution above-ground biomass density (AGBD) map (Mg ha -1 ) of France (FORMS-B). Extensive validation was conducted. First, independent datasets from Airborne Laser Scanning (ALS) and NFI data from thousands of plots reveal a mean absolute error (MAE) of 2.94 m for FORMS-H, which outperforms existing canopy height models. Second, FORMS-B was validated using two independent forest inventory datasets from the Renecofor permanent forest plot network and from the GLORIE forest inventory with MAE of 59.6 Mg ha -1 and 19.6 Mg.ha -1 respectively, providing greater performance than other AGBD products sampled over France. These results highlight the importance of coupling remote sensing technologies with recent advances in computer science to bring material insights to climate-efficient forest management policies. Additionally, our approach is based on open-access data having global coverage and a high spatial and temporal resolution, making the maps reproducible and easily scalable. FORMS products can be accessed from

The recent deployment of new satellite-based hyperspectral imaging sensors, along with advances in classification methods, has enhanced the acquisition of information on the spatiotemporal patterns of Land Use/Land Cover (LULC).. With this context, the study presented in this paper seeks to assess the combined utilization of the Pixxel's technology demonstrator hyperspectral satellite mission with the Support Vector Machines (SVM) and Random Forest (RF) machine learning classifiers for LULC mapping in an agricultural area located in Queensland, Australia. Accuracy assessment of the resulting LULC maps has been done based on relevant classification accuracy metrics. It has been observed that SVMs outperformed RF in terms of both overall classification and urban vegetation cover mapping accuracy. An overall accuracy of 92.34% with kappa of 0.90 was reported for the SVMs' results, whereas for RF, the accuracy was 89.65% with 0.89 kappa, respectively. Both classifiers produced similar results for the classes of mango tree and sugarcane crops, whereas minor differences were observed in the depiction of water bodies and artificial surfaces. Our study findings provide insightful perspectives on the potential of Pixxel datasets for delineating the spatiotemporal distribution of LULC in agricultural landscapes, showcasing Pixxel's hyperspectral sensors' promising potential in this field.