Comment on essd-2022-415

<strong class="journal-contentHeaderColor">Abstract.</strong> Young forest age mapping at a fine spatial resolution is important for increasing the accuracy of estimating land-atmosphere carbon fluxes and guiding forest management practices. In recent decades, China has actively conducted afforestation and forest protection projects, thereby, laying the foundation for the realization of carbon neutrality. However, very few studies have been conducted which map the ages of young forests for the whole of China at a fine spatial resolution. In this research, a continuous change detection and classification (CCDC)-based method suitable for large-scale forest age mapping is proposed, and used to estimate young forest ages across China in 2020 at a spatial resolution of 30 m. First, a 10 m spatial resolution land cover dataset (WorldCover2020) from the European Space Agency (ESA) was used to determine the forest cover areas in 2020. Then, the CCDC algorithm was used to identify stand-replacing disturbances to determine the stand age based on 436,967 Landsat tiles across China from 1990 to 2020. A validation sample set composed of multiple land use/land cover (LULC) products was used to calculate the overall accuracy (OA) of the 2020 young forest age (1&ndash;31 years) map of China, and the OA was 90.28 %. The reliability and applicability of the proposed CCDC-based forest age mapping method was validated by comparing the forest age map with Hansen&rsquo;s forest change dataset, Max Planck Institute for Biogeochemistry (MPI-BGC) 1 km global forest age datasets and field measurements. The CCDC-based method has strong application potential in real-time mapping of the age of young forests at the global scale. The produced forest age map provides a basic dataset for research on the forest carbon cycle and forest ecosystem services, and important guidance for government departments, such as the National Forestry and Grassland Administration and National Development and Reform Commission in China.

The accurate mapping of age structure and access to spatially explicit information are essential to optimal planning and policy-making for forest ecosystems, including forest management and sustainable economic development. Specifically, surveying and mapping the age structure of forests is crucial for calculating the carbon sequestration capacity of forest ecosystems. However, spatial heterogeneity and limited accessibility make forest age mapping in mountainous areas challenging. Here, we present a new workflow using ICESat-2 LiDAR data integrated with multisource remote sensing imagery to estimate forest age in Shangri-La, China. Two methods—a climate-driven exponential model and a random forest algorithm—are compared to infer the age structure of the five dominant species in Shangri-La. The climate-driven model, with an R2 of 0.67 and an RMSE of 12.79 years, outperforms the random forest model. The derived wall-to-wall forest age map at 30 m resolution reveals that nearly all forests in Shangri-La are mature or overmature, especially among the high-elevation species Abies fabri (Mast.) Craib and Picea asperata Mast., compared with Pinus yunnanensis Franch., Quercus aquifolioides Rehd. and E.H. Wils. and Pinus densata Mast., where the age structure is more evenly distributed across different elevation ranges. Younger forests are frequently found around human settlements and along the Jinsha River valley, whereas older forests are located in remote and high-elevation areas that are less disturbed. The combined use of active and passive remote sensing data has resulted in substantial improvements in the spatial detail and accuracy of wall-to-wall age mapping, which is expected to be a cost-effective approach for supporting forest management and carbon accounting in this important ecological region. The method developed here can be scaled to other mountain areas both to understand the age patterns and structure of mountain forests and to provide critical information for forestation, reforestation and carbon accounting in surface-to-high mountain areas, which are increasingly crucial for climate mitigation.

Uncovering forest dynamics using historical forest inventory data and Landsat time series

A fast hybrid approach for continuous land cover change monitoring and semantic segmentation using satellite time series

Spatio-temporal monitoring of marsh vegetation phenology and its response to hydro-meteorological factors using CCDC algorithm with optical and SAR images: In case of Honghe National Nature Reserve, China

Abstract. Forest age can determine the capacity of a forest to uptake carbon from the atmosphere. However, a lack of global diagnostics that reflect the forest stage and associated disturbance regimes hampers the quantification of age-related differences in forest carbon dynamics. This study provides a new global distribution of forest age circa 2010, estimated using a machine learning approach trained with more than 40 000 plots using forest inventory, biomass and climate data. First, an evaluation against the plot-level measurements of forest age reveals that the data-driven method has a relatively good predictive capacity of classifying old-growth vs. non-old-growth (precision = 0.81 and 0.99 for old-growth and non-old-growth, respectively) forests and estimating corresponding forest age estimates (NSE = 0.6 – Nash–Sutcliffe efficiency – and RMSE = 50 years – root-mean-square error). However, there are systematic biases of overestimation in young- and underestimation in old-forest stands, respectively. Globally, we find a large variability in forest age with the old-growth forests in the tropical regions of Amazon and Congo, young forests in China, and intermediate stands in Europe. Furthermore, we find that the regions with high rates of deforestation or forest degradation (e.g. the arc of deforestation in the Amazon) are composed mainly of younger stands. Assessment of forest age in the climate space shows that the old forests are either in cold and dry regions or warm and wet regions, while young–intermediate forests span a large climatic gradient. Finally, comparing the presented forest age estimates with a series of regional products reveals differences rooted in different approaches and different in situ observations and global-scale products. Despite showing robustness in cross-validation results, additional methodological insights on further developments should as much as possible harmonize data across the different approaches. The forest age dataset presented here provides additional insights into the global distribution of forest age to better understand the global dynamics in the forest water and carbon cycles. The forest age datasets are openly available at https://doi.org/10.17871/ForestAgeBGI.2021 (Besnard et al., 2021).

Mapping the annual dynamics of land cover in Beijing from 2001 to 2020 using Landsat dense time series stack

Dimensionality reduction techniques have been successfully applied in remote sensing to reduce redundant information. However, achieving dimensionality reduction and lossless recovery for multispectral data at any global location remains a challenge, particularly given the complex and variable nature of surface conditions. Furthermore, it is still unclear if the reduced features maintain temporal continuity and can be effectively integrated with existing time series algorithms for disturbance detection. This study trains a Uniform Manifold Approximation and Projection (UMAP) model based on Harmonized Landsat Sentinel-2 (HLS) imagery to accomplish multispectral dimensionality reduction. Subsequently, the manifold embeddings are used in the Continuous Change Detection and Classification (CCDC) algorithm for land disturbance detection. Two key conclusions are drawn from this study: 1) a general multispectral dimensionality reduction model was constructed based on UMAP, which is applicable to all global land surfaces and any seasons. The manifold embeddings exhibit a stable value range and preserve the coherence of the time series. 2) compared to full-spectrum multispectral data, the manifold embeddings achieved comparable performance in image prediction and disturbance detection. Our study demonstrates the potential of manifold learning-based representation of global land surface reflectance spectra for lightweight storage and processing of dense satellite image time series, while keeping the ability to detect any kinds of land disturbance.

Carbon emissions from forest ecosystems are greatly impacted by the acceleration of fragmentation and edge effects. Understanding these effects requires accurate monitoring of changes in fragmented forest landscapes. However, these changes are often low-intensity and small-scale, making it difficult to detect them using medium spatial resolution satellite images (e.g., Landsat). To address this challenge, this study developed the Pure Forest Index (PFI), which uses a combination of the existing vegetation index (VI) and spectral mixture analysis (SMA) to more effectively detect and characterize the contribution of forests to the observed spectral response of a pixel. The PFI was applied to detect forest changes in the Amazon rainforest from 1986 to 2020 using the Continuous Change Detection and Classification (CCDC) algorithm (hereafter referred to as the CCDC-PFI algorithm). The results showed reliable performance in mapping forest changes, with an overall accuracy of 0.94 (±0.03) at the spatial scale and a temporal accuracy of 91.1 % (within a two-year window). Comparison with other indices revealed that the PFI improves the ability to monitor forest dynamics with an increased overall accuracy of 0.02–0.35. The PFI also demonstrated advantages in enhancing sub-pixel forest information and suppressing non-forest backgrounds in various scenes compared to conventional VIs. The proposed approach is expected to benefit further research on forests and ecosystems.

Improved access to remotely sensed imagery and time series algorithms in combination with increased availability of cloud computing resources and platforms such as Google Earth Engine have significantly expanded the community of users processing and analyzing time series of satellite observations. Though individual time series analysis methods and their applications tend to be well-documented, comparisons of different approaches are beneficial to new users faced with the choice of different algorithms and parameterizations. We review two temporal segmentation approaches that have become increasingly prevalent in land cover mapping and monitoring: LandTrendr (Landsat-based detection of Trends in Disturbance and Recovery) and CCDC (Continuous Change Detection and Classification). We examine differences in the way these approaches use the temporal and spectral domains and compare model specifications and outputs. This review highlights previous work and applications, current limitations, ongoing challenges, and opportunities for future integration and comparison of methods and map products, and is expected to benefit both user and developer communities.

Coastal wetland ecosystems, one of the most important ecosystems in the world, play an important role in regulating climate, sequestering blue carbon, and maintaining sustainable development of coastal zones. Wetland landscapes are notoriously difficult to map with satellite data, particularly in highly complex, dynamic coastal regions. The Liao River Estuary (LRE) wetland in Liaoning Province, China, has attracted major attention due to its status as Asia’s largest coastal wetland, with extensive Phragmites australis (reeds), Suaeda heteroptera (seepweed, red beach), and other natural resources that have been continuously encroached upon by anthropogenic land-use activities. Using the Continuous Change Detection and Classification (CCDC) algorithm and all available Landsat images, we mapped the spatial–temporal changes of LRE coastal wetlands (e.g., seepweed, reed, tidal flats, and shallow marine water) annually from 1986 to 2018 and analyzed the changes and driving forces. Results showed that the total area of coastal wetlands in the LRE shrank by 14.8% during the study period. The tidal flats were the most seriously affected type, with 45.7% of its total area lost. One of the main characteristics of wetland change was the concurrent disappearance and emergence of wetlands in different parts of the LRE, creating drastically different mixtures of wetland quality (e.g., wetland age composition) in addition to area change. The reduction and replacement/translocation of coastal wetlands were mainly caused by human activities related to urbanization, tourism, land reclamation, and expansion of aquaculture ponds. Our efforts in mapping annual changes of wetlands provide direct, specific, and spatially explicit information on rates, patterns, and causes of coastal wetland change, both in coverage and quality, so as to contribute to the effective plans and policies for coastal management, preservation, and restoration of coastal ecosystem services.

Accessibility of multispectral, multitemporal imagery combined with recent advances in cloud computing and machine learning approaches have enhanced our ability to model habitat characteristics across broad spatial and temporal scales. We integrated a large dataset of known nest and roost sites of a threatened species, the Mexican spotted owl (Strix occidentalis lucida), in the southwestern USA with Landsat imagery processed using the Continuous Change Detection and Classification (CCDC) time series algorithm on Google Earth Engine. We then used maximum entropy modeling (Maxent) to classify the landscape into four ‘spectral similarity’ classes that reflected the degree to which 30-m pixels contained a multispectral signature similar to that found at known owl nest/roost sites and mapped spectral similarity classes from 1986–2020. For map interpretation, we used nationally consistent forest inventory data to evaluate the structural and compositional characteristics of each spectral similarity class. We found a monotonic increase of structural characteristics typically associated with owl nesting and roosting over classes of increasing similarity, with the ‘very similar’ class meeting or exceeding published minimum desired management conditions for owl nesting and roosting. We also found an increased rate of loss of forest vegetation typical of owl nesting and roosting since the beginning of the 21st century that can be partly attributed to increased frequency and extent of large (≥400 ha) wildfires. This loss resulted in a 38% reduction over the 35-year study period in forest vegetation most similar to that used for owl nesting and roosting. Our modelling approach using cloud computing with time series of Landsat imagery provided a cost-effective tool for landscape-scale, multidecadal monitoring of vegetative components of a threatened species’ habitat. Our approach could be used to monitor trends in the vegetation favored by any other species, provided that high-quality location data such as we presented here are available.

Monitoring forest disturbances is crucial for the preservation of biodiversity, climate change mitigation, and overall ecosystem health. Remote sensing offers a powerful tool for monitoring forest disturbances as it can cover large areas, detect changes in near real-time, and cost effective as compared to survey method. Thus, this study tries to utilize the use of Sentinel-2 imagery to detect forest disturbances in Forest Research Institute Malaysia (FRIM) campus. Continuous change detection and classification (CCDC) algorithm was employed to identify disturbances that occurred between 2019 and 2023. NDVI, NDMI, and NBR were also used to improve the accuracy of disturbance detection. The study returned a promising result, achieving a precision of 85 percent in detecting forest disturbances. The effective utilization of Sentinel-2 imagery and the CCDC method in this study highlights not only the efficiency of remote sensing in precisely detecting forest disturbances but also underscores its potential to contribute significantly to sustainable conservation initiatives.

A time series analysis of urbanization induced land use and land cover change and its impact on land surface temperature with Landsat imagery

The forest age can be defined as the time since the last stand-replacement event. Within the local context, it determines the forest successional stage, a fundamental variable for diagnosing the net carbon fluxes in terrestrial ecosystems. To accurately quantify the carbon sink-source strength in these ecosystems and inform adaptation-mitigation strategies, it is essential to be able to infer forest age at high resolution.This study presents an updated version of the MPI-BGC forest age product, featuring global distributions of forest age for 2010, 2017, 2018, and 2020 at 100m spatial resolution. We employed two machine learning approaches, XGBoost and a multi-layer perceptron model, to create data-driven estimates of forest age based on over 40,000 forest inventory plots, biomass, remote-sensing, and climate data. One key innovation of our approach is the incorporation of Landsat-based disturbance history metrics as input variables. Our updated estimates show better precision in identifying old-growth forests and reduce overestimation biases in young forests and underestimation biases in old forests, but not completely. Additionally, we found substantial regional variations related to changes in covariate strength and improvement in the model. Also, we discussed the uncertainty layers, created using model ensembles, that materialize the quantification of methodological uncertainty in the forest age estimates.An analysis of the global distribution of forest age reveals significant variations across the years studied. We also quantified the changes in forest age in regions with high deforestation or forest degradation rates, where younger stands are becoming more prevalent. We discuss the challenges and limitations of using regression-based mapping approaches, including the choice of machine learning algorithm, spatial cross-validation techniques, and the caveats of extrapolation, given data limitations. Our research highlights the complementary biomass-based approaches for determining forest age and underscores the importance of detailed global estimates at high spatial resolutions. Overall, this study advances our understanding of forest age, a key variable for understanding the carbon cycle in terrestrial ecosystems.