Synergistic Use of Sentinel-1 and Sentinel-2 to Map Natural Forest and Acacia Plantation and Stand Ages in North-Central Vietnam

Forest age estimation in northern Arkhangelsk region based on machine learning pipeline on Sentinel-2 and auxiliary data

Tree age is one of the key characteristics of a forest, along with tree species and height. It affects management decisions of forest owners and allows researchers to analyze environmental characteristics in support of sustainable development. Although forest age is of primary significance, it can be unknown for remote areas and large territories. Currently, remote sensing (RS) data supports rapid information gathering for wide territories. To automate RS data processing and estimate forest characteristics, machine learning (ML) approaches are applied. Although there are different data sources that can be used as features in ML models, there is no unified strategy on how to prepare a dataset and define a training task to estimate forest age. Therefore, in this work, we aim to conduct a comprehensive study on forest age estimation using remote sensing observations of the Sentinel-2 satellite and two ML-based approaches for forestry inventory data, namely stand-based and pixel-based. We chose the CatBoost algorithm to assess these two approaches. To establish the robustness of the pipeline, an in-depth analysis is conducted, embracing diverse scenarios incorporating dominant species information, tree height, Digital Elevation Model (DEM), and vegetation indices. We performed experiments on forests in the northern Arkhangelsk region and obtained the best Mean Absolute Error (MAE) result of 7 years in the case of the stand-based approach and 6 years in the case of the pixel-based approach. These results are achieved for all available input data such as spectral satellites bands, vegetation indices, and auxiliary forest characteristics (dominant species and height). However, when only spectral bands are used, the MAE metric is the same both for per-pixel and per-stand approaches and equals 11 years. It was also shown that, despite high correlation between forest age and height, only height can not be used for accurate age estimation: the MAE increases to 18 and 26 years for per-pixel and per-stand approaches, respectively. The conducted study might be useful for further investigation of forest ecosystems through remote sensing observations.

Implications of tropical cyclones on damage and potential recovery and restoration of logged forests in Vietnam

Many natural forests in Southeast Asia are degraded following decades of logging. Restoration of these forests is delayed by ongoing logging and tropical cyclones, but the implications for recovery are largely uncertain. We analysed meteorological, satellite and forest inventory plot data to assess the effect of Typhoon Doksuri, a major tropical cyclone, on the forest landscapes of central Vietnam consisting of natural forests and plantations. We estimated the return period for a cyclone of this intensity to be 40 years. Plantations were almost twice as likely to suffer cyclone damage compared to natural forests. Logged natural forests (9-12 years after cessation of government-licensed logging) were surveyed before and after the storm with 2 years between measurements and remained a small biomass carbon sink (0.1 ± 0.3 Mg C ha-1 yr-1) over this period. The cyclone reduced the carbon sink of recovering natural forests by an average of 0.85 Mg C ha-1 yr-1, less than the carbon loss due to ongoing unlicensed logging. Restoration of forest landscapes in Southeast Asia requires a reduction in unlicensed logging and prevention of further conversion of degraded natural forests to plantations, particularly in landscapes prone to tropical cyclones where natural forests provide a resilient carbon sink. This article is part of the theme issue 'Understanding forest landscape restoration: reinforcing scientific foundations for the UN Decade on Ecosystem Restoration'.

A Planted Forest Mapping Method Based on Long-Term Change Trend Features Derived from Dense Landsat Time Series in an Ecological Restoration Region

Planted forests provide a variety of meaningful ecological functions and services, which is a major approach for ecological restoration, especially in arid areas. However, mapping planted forests with remote-sensed data remains challenging due to the similarities in canopy spectral and structure characteristics and associated phenology features between planted forests and other vegetation types. In this study, taking advantage of the Google Earth Engine (GEE) platform and taking the Ningxia Hui Autonomous Region in northwestern China as an example, we developed an approach to map planted forests in an arid region by applying long-term features of the NDVI derived from dense Landsat time series. Our land cover map achieved a satisfactory accuracy and relatively low uncertainty, with an overall accuracy of 93.65% and a kappa value of 0.92. Specifically, the producer (PA) and user accuracies (UA) were 92.48% and 91.79% for the planted forest class, and 93.88% and 95.83% for the natural forest class, respectively. The total planted forest area was estimated as 3608.72 km2 in 2020, accounting for 20.60% of the study area. The proposed mapping approach can facilitate assessment of the restoration effects of ecological engineering and research on ecosystem services and stability of planted forests.

Integrating Remotely Sensed Leaf Area Index with Biome-BGC to Quantify the Impact of Land Use/Land Cover Change on Water Retention in Beijing

Maintaining or increasing water retention in ecosystems (WRE) can reduce floods and increase water resource provision. However, few studies have taken the effect of the spatial information of vegetation structure into consideration when assessing the effects of land use/land cover (LULC) change on WRE. In this study, we integrated the remotely sensed leaf area index (LAI) into the ecosystem process-based Biome-BGC model to analyse the impact of LULC change on the WRE of Beijing between 2000 and 2015. Our results show that the volume of WRE increased by approximately 8.58 million m3 in 2015 as compared with 2000. The volume of WRE in forests increased by approximately 26.74 million m3, while urbanization, cropland expansion and deforestation caused the volume of WRE to decline by 11.96 million m3, 5.86 million m3 and 3.20 million m3, respectively. The increased WRE contributed by unchanged forests (14.46 million m3) was much greater than that of new-planted forests (12.28 million m3), but the increase in WRE capacity per unit area in new-planted forests (124.69 ± 14.30 m3/ha) was almost tenfold greater than that of unchanged forests (15.60 ± 7.85 m3/ha). The greater increase in WRE capacity in increased forests than that of unchanged forests was mostly due to the fact that the higher LAI in unchanged forests induced more evapotranspiration to exhaust more water. Meanwhile, the inverted U-shape relationship that existed between the forest LAI and WRE implied that continued increased LAI in forests probably caused the WRE decline. This study demonstrates that integrating remotely sensed LAI with the Biome-BGC model is feasible for capturing the impact of LULC change with the spatial information of vegetation structure on WRE and reduces uncertainty.