Landsat Surface Reflectance Research Articles

A global forest burn severity dataset from Landsat imagery (2003–2016)

Abstract. Forest fires, while destructive and dangerous, are important to the functioning and renewal of ecosystems. Over the past 2 decades, large-scale, severe forest fires have become more frequent globally, and the risk is expected to increase as fire weather and drought conditions intensify. To improve quantification of the intensity and extent of forest fire damage, we have developed a 30 m resolution global forest burn severity (GFBS) dataset of the degree of biomass consumed by fires from 2003 to 2016. To develop this dataset, we used the Global Fire Atlas product to determine when and where forest fires occurred during that period and then we overlaid the available Landsat surface reflectance products to obtain pre-fire and post-fire normalized burn ratios (NBRs) for each burned pixel, designating the difference between them as dNBR and the relative difference as RdNBR. We compared the GFBS dataset against the Canada Landsat Burned Severity (CanLaBS) product, showing better agreement than the existing Moderate Resolution Imaging Spectrometer (MODIS)-based global burn severity dataset (MOdis burn SEVerity, MOSEV) in representing the distribution of forest burn severity over Canada. Using the in situ burn severity category data available for the 2013 wildfires in southeastern Australia, we demonstrated that GFBS could provide burn severity estimation with clearer differentiation between the high-severity and moderate-/low-severity classes, while such differentiation among the in situ burn severity classes is not captured in the MOSEV product. Using the CONUS-wide composite burn index (CBI) as a ground truth, we showed that dNBR from GFBS was more strongly correlated with CBI (r=0.63) than dNBR from MOSEV (r=0.28). RdNBR from GFBS also exhibited better agreement with CBI (r=0.56) than RdNBR from MOSEV (r=0.20). On a global scale, while the dNBR and RdNBR spatial patterns extracted by GFBS are similar to those of MOSEV, MOSEV tends to provide higher burn severity levels than GFBS. We attribute this difference to variations in reflectance values and the different spatial resolutions of the two satellites. The GFBS dataset provides a more precise and reliable assessment of burn severity than existing available datasets. These enhancements are crucial for understanding the ecological impacts of forest fires and for informing management and recovery efforts in affected regions worldwide. The GFBS dataset is freely accessible at https://doi.org/10.5281/zenodo.10037629 (He et al., 2023).

A dataset of trophic state index for nation-scale lakes in China from 40-year Landsat observations

Trophic state index (TSI) serves as a key indicator for quantifying and understanding the lake eutrophication, which has not been fully explored for long-term water quality monitoring, especially for small and medium inland waters. Landsat satellites offer an effective complement to facilitate the temporal and spatial monitoring of multi-scale lakes. Landsat surface reflectance products were utilized to retrieve the annual average TSI for 2693 lakes over 1 km2 in China from 1984 to 2023. Our method first distinguishes lake types by pixels with a decision tree and then derives relationships between trophic state and algal biomass index. Validation with public reports and existing datasets confirmed the good consistency and reliability. The dataset provides reliable annual TSI results and credible trends for lakes under different area scales, which can serve as a reference for further research and provide convenience for lake sustainable management.

Soil erosion assessment and identification of erosion hotspot areas in the upper Tekeze Basin, Northern Ethiopia

Soil erosion is a major environmental problem in Ethiopia, reducing topsoil and agricultural land productivity. Soil loss estimation is a critical component of sustainable land management practices because it provides important information about soil erosion hotspot areas and prioritizes areas that require immediate management interventions. This study integrates the Revised Universal Soil Loss Equation (RUSLE) with Google Earth Engine (GEE) to estimate soil erosion rates and map soil erosion in the Upper Tekeze Basin, Northern Ethiopia. SoilGrids250 m, CHIRPS-V2, SRTM-V3, MERIT Hydrograph, NDVI from sentinel collections and land use land cover (LULC) data were accessed and processed in the GEE Platform. LULC was classified using Random forest (RF) classification algorithm in the GEE platform. Landsat surface reflectance images from Landsat 8 Operational land imager (OLI) sensors (2021) was used for LULC classification. Besides, different auxiliary data were utilized to improve the classification accuracy. Using the RUSLE-GEE framework, we analyzed the soil loss rate in different agroecologies and LULC types in the upper Tekeze basin in Waghimra zone. The results showed that the average soil loss rate in the Upper Tekeze basin is 25.5 t ha−1 yr−1. About 63 % of the basin is experiencing soil erosion above the maximum tolerable rate, which should be targeted for land management interventions. Specifically, 55 % of the study area, which is covered by unprotected shrubland is experiencing mean annual soil loss of 34.75 t ha−1 yr−1 indicating the need for immediate soil conservation intervention. The study also revealed evidence that this high mean soil loss rate of the basin can be reduced to a tolerable rate by implementing integrative watershed management and exclosures. Furthermore, this study demonstrated that GEE could be a good source of datasets and a computing platform for RUSLE, in particular for data scarce semi-arid and arid environments. The results from this study are reliable for decision-making for rapid soil erosion assessment and intervention prioritization.

Monitoring Water Quality Indicators over Matagorda Bay, Texas, Using Landsat-8

Remote sensing datasets offer a unique opportunity to observe spatial and temporal trends in water quality indicators (WQIs), such as chlorophyll-a, salinity, and turbidity, across various aquatic ecosystems. In this study, we used available in situ WQI measurements (chlorophyll-a: 17, salinity: 478, and turbidity: 173) along with Landsat-8 surface reflectance data to examine the capability of empirical and machine learning (ML) models in retrieving these indicators over Matagorda Bay, Texas, between 2014 and 2023. We employed 36 empirical models to retrieve chlorophyll-a (12 models), salinity (2 models), and turbidity (22 models) and 4 ML families—deep neural network (DNN), distributed random forest, gradient boosting machine, and generalized linear model—to retrieve salinity and turbidity. We used the Nash–Sutcliffe efficiency coefficient (NSE), correlation coefficient (r), and normalized root mean square error (NRMSE) to assess the performance of empirical and ML models. The results indicate that (1) the empirical models displayed minimal effectiveness when applied over Matagorda Bay without calibration; (2) once calibrated over Matagorda Bay, the performance of the empirical models experienced significant improvements (chlorophyll-a—NRMSE: 0.91 ± 0.03, r: 0.94 ± 0.04, NSE: 0.89 ± 0.06; salinity—NRMSE: 0.24 ± 0, r: 0.24 ± 0, NSE: 0.06 ± 0; turbidity—NRMSE: 0.15 ± 0.10, r: 0.13 ± 0.09, NSE: 0.03 ± 0.03); (3) ML models outperformed calibrated empirical models when used to retrieve turbidity and salinity, and (4) the DNN family outperformed all other ML families when used to retrieve salinity (NRMSE: 0.87 ± 0.09, r: 0.49 ± 0.09, NSE: 0.23 ± 0.12) and turbidity (NRMSE: 0.63± 0.11, r: 0.79 ± 0.11, NSE: 0.60 ± 0.20). The developed approach provides a reference context, a structured framework, and valuable insights for using empirical and ML models and Landsat-8 data to retrieve WQIs over aquatic ecosystems. The modeled WQI data could be used to expand the footprint of in situ observations and improve current efforts to conserve, enhance, and restore important habitats in aquatic ecosystems.

Characterizing post-fire northern boreal forest height dynamics

ABSTRACT The forest structure of Canada’s northern unmanaged boreal forests are not well characterized despite their critical importance for ecosystem services such as carbon storage and caribou habitat, as well as their vulnerability to climate change and increasing disturbance rates, especially wildfire. Spaceborne lidar observations acquired from the Advanced Topographic Laser Altimeter System (ATLAS) instrument onboard the Ice, Cloud and Elevation Satellite-2 (ICESat-2), enables monitoring of forest structure in these remote northern ecosystems and characterization of long-term post-fire recovery; however, ICESat-2 is a sampling instrument, acquiring data transects rather than wall-to-wall data coverage. By imputing ICESat-2 estimates of canopy height using annual time-series (1984–2021) of spectral indices, change metrics, and land cover metrics from Landsat surface reflectance composites, we derived annual, spatially explicit, wall-to-wall estimates of canopy height for the period 1984–2021 across a 19.6 Mha area of boreal forests in northwestern Ontario and a 630,000 ha area of managed boreal forest further south that had coincident airborne laser scanning (ALS) data for validation. The accuracy of the imputation of derived canopy height estimates in the northwestern site was assessed using a reserved set of ICESat-2 observations (r = 0.76; RMSD = 3.04 m, pRMSD = 33.9%, MD = −0.10 m, and pMD = −1.0%). Using coincident ALS data, the accuracy of the imputed canopy heights in the southern study site were also assessed (r = 0.81; RMSD = 3.65 m, pRMSD = 25.87%), MD = −0.70 m, pMD = 4.96%). Examination of height dynamics post fire indicated that canopy height decreased 8–10 years post fire and recovered between 60% and 85% of pre-fire canopy height within 30 years post fire. The approach presented could readily be extended to similar northern boreal forest areas to provide broad-scale characterizations of fire impacts on forest structure and subsequent recovery.

Hybrid inversion of radiative transfer models based on topographically corrected Landsat surface reflectance improves leaf area index and aboveground biomass retrievals of grassland on the hilly Loess Plateau

ABSTRACT Accurate monitoring of the leaf area index (LAI) and aboveground biomass (AGB) using remote sensing at a fine scale is crucial for understanding the spatial heterogeneity of vegetation structure in mountainous ecosystems. Understanding discrepancies in various retrieval strategies considering topographic effects or not is necessary to improve LAI and AGB estimations over mountainous areas. In this study, the performances of the look-up table method (LUT) using radiative transfer model (RTM), machine learning algorithms (MLAs), and hybrid RTM integrating RTM and MLAs based on Landsat surface reflectance (SR) before and after topographic correction were compared and analyzed. The results show that topographic correction improves the accuracies of retrieval methods involving RTM more significantly than the MLAs, meanwhile, it reduces the performance variability of different MLAs. Based on the topographically corrected Landsat SR, the random forest (RF) combined with RTM improves the retrieval accuracy of RTM-based LUT by 7.7% for LAI and 13.8% for AGB, and reduces the simulation error of MLA by 15.1% for LAI and 20.1% for AGB. Compared with available remote sensing products, the hybrid RTM based on Landsat SR with topographic correction has better feasibility to capture LAI and AGB variation at 30 m scale over mountainous areas.

National-scale remotely sensed lake trophic state from 1984 through 2020

Lake trophic state is a key ecosystem property that integrates a lake’s physical, chemical, and biological processes. Despite the importance of trophic state as a gauge of lake water quality, standardized and machine-readable observations are uncommon. Remote sensing presents an opportunity to detect and analyze lake trophic state with reproducible, robust methods across time and space. We used Landsat surface reflectance data to create the first compendium of annual lake trophic state for 55,662 lakes of at least 10 ha in area throughout the contiguous United States from 1984 through 2020. The dataset was constructed with FAIR data principles (Findable, Accessible, Interoperable, and Reproducible) in mind, where data are publicly available, relational keys from parent datasets are retained, and all data wrangling and modeling routines are scripted for future reuse. Together, this resource offers critical data to address basic and applied research questions about lake water quality at a suite of spatial and temporal scales.

Analyzing Land Use/Land Cover Changes Using Google Earth Engine and Random Forest Algorithm and Their Implications to the Management of Land Degradation in the Upper Tekeze Basin, Ethiopia.

Land use and land cover change (LULCC) without appropriate management practices has been identified as a major factor contributing to land degradation, with significant impacts on ecosystem services and climate change and hence on human livelihoods. Therefore, up-to-date and accurate LULCC data and maps at different spatial scales are significant for regular monitoring of existing ecosystems, proper planning of natural resource management, and promotion of sustainable regional development. This study investigates the temporal and spatial dynamics of land use land cover (LULC) changes over 31 years (1990-2021) in the upper Tekeze River basin, Ethiopia, utilizing advanced remote sensing techniques such as Google Earth Engine (GEE) and the Random Forest (RF) algorithm. Landsat surface reflectance images from Landsat Thematic Mapper (TM) (1990, 2000, and 2010) and Landsat 8 Operational land imager (OLI) sensors (2021) were used. Besides, auxiliary data were utilized to improve the classification of LULC classes. LULC was classified using the Random Forest (RF) classification algorithm in the Google Earth Engine (GEE). The OpenLand R package was used to map the LULC transition and intensity of changes across the study period. Despite the complexity of the topographic and climatic features of the study area, the RF algorithm achieved high accuracy with 0.83 and 0.75 overall accuracy and Kappa values, respectively. The LULC change results from 1990 to 2021 showed that forest, bushland, shrubland, and bareland decreased by 12.2, 24.8, 1.2, and 15.4%, respectively. Bareland has changed to farmland, settlement, and dry riverbed and stream channels. Expansion of dry stream channels and sandy land surfaces has been observed from 1990 to 2021. Bushland has shown an increment by 17.2% from 1900 to 2010 but decreased by 19.5% from 2010 to 2021. Throughout the study period, water, farmland, dry stream channels and riverbeds, and urban settlements showed positive net gains of 484, 8.7, 82, and 26778.5%, respectively. However, forest, bush, shrub, and bareland experienced 12.17, 24.8, 1.2, and 15.37% losses. The observed changes showed the existing land degradation and the future vulnerability of the basin which would serve as an evidence to mitigate land degradation by avoiding the future conversion of forest, bushland, and shrubland to farmland, on the one hand, and by scaling up sustainable farmland management, and afforestation practices on degraded and vulnerable areas, on the other hand.

Satellite Estimation of pCO2 and Quantification of CO2 Fluxes in China’s Chagan Lake in the Context of Climate Change

The massive increase in the amount of greenhouse gases in the atmosphere, especially carbon dioxide (CO2), has had a significant impact on the global climate. Research has revealed that lakes play an important role in the global carbon cycle and that they can shift between the roles of carbon sources and sinks. This study used Landsat satellite images to analyze the spatiotemporal characteristics and factors influencing the CO2 changes in Chagan Lake in China. We conducted six water sampling campaigns at Chagan Lake in 2020–2021 and determined the partial pressure of carbon dioxide (pCO2) from 110 water samples. Landsat surface reflectance was matched with water sampling events within ±7 days of satellite overpasses, yielding 75 matched pairs. A regression analysis indicated strong associations between pCO2 and both the band difference model of the near-infrared band and green band (Band 5-Band 3, R2 = 0.83, RMSE = 27.55 μatm) and the exponential model [((exp(b3) − exp(b5))2/(exp(b3) + exp(b5))2, R2 = 0.82, RMSE = 27.99 μatm]. A comparison between the performances of a linear regression model and a machine learning model found that the XGBoost model had the highest fitting accuracy (R2 = 0.94, RMSE = 16.86 μatm). We used Landsat/OLI images acquired mainly in 2021 to map pCO2 in Chagan Lake during the ice-free period. The pCO2 in the surface water of Chagan Lake showed considerable spatiotemporal variability within a range of 0–200 μatm. pCO2 also showed significant seasonal variations, with the lowest and highest mean values in autumn (30–50 μatm) and summer (120–150 μatm), respectively. Spatially, the pCO2 values in the southeast of Chagan Lake were higher than those in the northwest. The CO2 fluxes were calculated based on the pCO2 and ranged from −3.69 to −2.42 mmol/m2/d, indicating that Chagan Lake was absorbing CO2 (i.e., it was a weak carbon sink). Temperature, chlorophyll a, total suspended matter, and turbidity were found to have reinforcing effects on the overall trend of pCO2, while the Secchi disk depth was negatively correlated with pCO2. The results of this study provide valuable insights for assessing the role of lakes in the carbon cycle in the context of climate change.

Classification and Transition of Grassland in Qinghai, China, from 1986 to 2020 with Landsat Archives on Google Earth Engine

The lack of long-duration, high-frequency grassland classification products limits further understanding of the grasslands’ long-term succession. This study first explored the annual mapping of grassland with fourteen categories at 30 m in Qinghai, China, from 1986 to 2020 based on Google Earth Engine (GEE) and the Integrated Orderly Classification System (IOCSG). Specifically, we proposed an image composite strategy to obtain annual source images for classification, by quarterly compositing multi-sensor and multi-temporal Landsat surface reflectance images. Subsequently, the 35-year area time series of each category was analyzed in terms of trend, degree of change, and succession of each category. The results indicate that the different grasslands of the IOCSG can be effectively differentiated by utilizing the designed feature bands of remote sensing data. Additionally, the proposed annual image composition strategy can not only decrease the invalid pixels but also promote classification accuracy. The grasslands transition analysis from 1986 to 2020 implies the progressive urbanization, warming, and wetting trend in Qinghai. The generated 35-year annual grassland thematic data in Qinghai can serve as an elementary dataset for further regional ecological and climate change studies. The proposed methodology of large-scale grassland classification can also be referenced to other applications like land use/cover mapping and ecological resource monitoring.

Validity of the Landsat surface reflectance archive for aquatic science: Implications for cloud‐based analysis

AbstractOriginally developed for terrestrial science and applications, the US Geological Survey Landsat surface reflectance (SR) archive spanning ~ 40 yr of observations has been increasingly utilized in large‐scale water‐quality studies. These products, however, have not been rigorously validated using in situ measured reflectance. This letter quantifies and demonstrates the quality of the SR products by harnessing a sizeable global dataset (N = 1100). We found that the Landsat 8/9 SR in the green and red bands marginally meet the targeted accuracy requirements (30%), whereas the uncertainties in the blue and coastal‐aerosol bands ranged from 48% to 110%. We further observed > +25% biases in the visible bands of Landsat 5/7 SR, which can introduce an apparent downward trend when applied in time‐series analyses combined with Landsat 8/9. Users must exercise caution when using this archive for trend analyses, and progress in atmospheric correction is required to foster advanced applications of the Landsat archive for aquatic science.

Validation of Landsat-9 and Landsat-8 Surface Temperature and Reflectance during the Underfly Event

With the launch of Landsat-9 on 27 September 2021, Landsat continues its fifty-year continuity mission of providing users with calibrated Earth observations. It has become a requirement that an underflight experiment be performed during commissioning to support sensor cross-calibration. In this most recent experiment, Landsat-9 flew under Landsat-8 for nearly three days with over 50% ground overlap, from 13 to 15 November 2021. To address the scarcity of reference data that are available to support calibration and validation early-on in the mission, a ground campaign was planned and executed by the Rochester Institute of Technology (RIT) on 14 November 2021 to provide full spectrum measurements for early mission comparisons. The primary experiment was conducted in the Outer Banks, North Carolina at Jockey’s Ridge Sand Dunes. Full-spectrum ground-based measurements were acquired with calibrated reference equipment, while a novel Unmanned Aircraft System (UAS)-based platforms acquired hyperspectral visible and near-infrared (VNIR)/Short-wave infrared (SWIR) imagery data and coincident broadband cooled thermal infrared (TIR) imagery. Results of satellite/UAS/ground comparisons were an indicator, during the commissioning phase, that Landsat-9 is behaving consistently with Landsat-8, ground reference, and UAS measurements. In the thermal infrared, all measurements agree to be within 1 K over water and to within 2 K over sand, which represents the most challenging material for estimating surface temperature. For the surface reflectance product(s), Landsat-8 and -9 are in good agreement and only deviate slightly from ground reference in the SWIR bands; a deviation of 2% in the VNIR and 5–8% in the SWIR regime. Subsequent longer-term studies indicate that Landsat 9 continues to perform as expected. The behavior of Thermal Infrared Sensor-2 (TIRS-2) against reference is also shown for the first year of the mission to illustrate its consistent performance.

Spatial prediction of soil salinity based on the Google Earth Engine platform with multitemporal synthetic remote sensing images

Accurately analyzing the spatial distribution of salinity at the regional scale is crucial for regional sustainable development. Currently, there are no reliable approaches for mapping salinity due to the substantial spatial variability of soil salinity. However, remote sensing technology has been successfully used to monitor soil properties in various locations. This research investigates an approach to choose the most appropriate remote sensing time window using a cloud computing platform and synthesizes multitemporal remote sensing images for high spatial resolution digital mapping of salinization at the regional scale. The research site for this experiment is the Werigan-Kuqa Oasis in Xinjiang, China. In the first stage, we screen the Landsat-8 surface reflectance datasets for four-time windows: April–October, May–September, June–August multitemporal, and July single-date images using the Google Earth Engine (GEE) cloud platform. Then, we synthesize images based on minimum, maximum, mean, and median values and construct multiple spectral indices. We completed model training and digital mapping in the second stage using a geemap. For the first time, we used geemap to integrate a local scikit-learn machine learning library with GEE to train a machine learning model. Next, we performed salinity prediction and mapping based on the best data set. This study demonstrates that (1) Various environmental variables contribute differently to modeling in the same time window. Additionally, these variables exhibit variation across different time windows. However, certain indices, including clay index (CLEX), near infrared (NIR), difference vegetation index (DVI), carbonate index (CAEX), green ratio vegetation index (GRVI), canopy response salinity (CRSI), and elevation, demonstrate stable contributions across different time windows. (2) Employing GEE to choose the suitable time window for synthesizing remote sensing images has a better prediction effect compared to single-date image prediction accuracy. Through experimental comparison, the best time window is May–September. (3) The mean synthesis approach from May to September exhibits the best performance in the model, while the median synthesis approach exhibits stable performance in the other two individual time windows. (4) Combining scikit-learn with GEE using Geemap improves model optimization and enhances the mapping performance. This study broadens the application of GEE in soil mapping and improves salinity prediction using multitemporal synthetic pictures with time frames.

Utility of daily 3 m Planet Fusion Surface Reflectance data for tillage practice mapping with deep learning

Tillage practices alter soil surface structure that can be potentially captured by satellite images with both high spatial and temporal resolution. This study explored tillage practice mapping using the daily Planet Fusion surface reflectance (PF-SR) gap-free 3 m data generated by fusing PlanetScope with Landsat-8, Sentinel-2 and MODIS surface reflectance data. The study area is a 220 × 220 km2 agricultural area in South Dakota, USA, and the study used 3285 PF-SR images from September 1, 2020 to August 31, 2021. The PF-SR images for the surveyed 433 fields were sliced into 10,747 training (70%) and evaluation (30%) non-overlapping time series patches. The training and evaluation patches were from different fields for evaluation data independence. The performance of four deep learning models (i.e., 2D convolutional neural networks (CNN), 3D CNN, CNN-Long short-term memory (LSTM), and attention CNN-LSTM) in tillage practice mapping, as well as their sensitivity to different spatial (i.e., 3 m, 24 m, and 96 m) and temporal resolutions (16-day, 8-day, 4-day, 2-day and 1-day) were examined. Classification accuracy continuously increased with increases in both temporal and spatial resolutions. The optimal models (3D CNN and attention CNN-LSTM) achieved ∼77% accuracy using 2-day or daily 3 m resolution data as opposed to ∼72% accuracy using 16-day 3 m resolution data or daily 24 m resolution data. This study also analyzed the feature importance of different acquisition dates for the two optimal models. The 3D CNN model feature importances were found to agree well with the tillage practice time. High feature importance was associated with observations during the fall and spring tillage period (i.e., fresh tillage signals) whereas the crop peak growing period (i.e., tillage signals weathered and confounded by dense canopy) was characterized by a relatively low feature importance. The work provides valuable insights into the utility of deep learning for tillage mapping and change event time identification based on high resolution imagery.

Estimating and mapping forest age across Canada's forested ecosystems

Forest age is an important variable for assessments of biodiversity and habitat, sustainable forest and land management, as well as forest carbon science and modeling. Tree and stand age are typically measured directly on site, or estimated through visual photo interpretation, with spatially explicit maps of forest age not often produced over large areas. Remote sensing enables the generation of wall-to wall, spatially explicit maps of disturbance events within the satellite record; however, as disturbance is relatively rare on the landscape in a given year, additional means of determining forest age are required. As reviewed herein, the estimation of forest age using optical Earth observation data is challenging due to the limited spectral link to the attribute of interest, especially as forests get older. The temporally dictated multi-method approach to forest age estimation outlined herein acknowledges these limitations, by applying the approach that is best suited to the quality of the information available, depending on the epoch of interest. In this research, we combine three approaches to estimate forest age at a 30-m spatial resolution using Landsat data. The first approach uses change detection protocols to detect disturbance from 1985 to 2019, with time since disturbance used as a proxy for forest age. The second approach uses Landsat surface reflectance composites to identify pixels exhibiting evidence of recovery from a disturbance that occurred within the twenty years prior to 1985, allowing for the extension of forest age estimates to 1965. Finally, given an understanding of the linkage between forest age and canopy height, inverted allometric equations are coupled with maps of forest structure and productivity metrics to model forest age for those pixels that show no evidence of disturbance or recovery to a maximum of 150 years, acknowledging that uncertainty in age estimate increases with increasing age. Combining these three approaches, forest age estimates are made for every treed pixel found within the 650 Mha forested ecosystems of Canada. Nationwide, mean estimated forest age for forests ≤150 years old (representing 94.1% of treed area) was 70 years (standard deviation = 32.1 years). For confidence building, forest age estimates were compared to reported forest age in the National Forest Inventory (NFI) both spatially and aspatially. Nationally, 5.9% of the forested area was estimated to be older than 150 years, while 9.5% of area within in the NFI sample was recorded as older than 150 years. The median estimated forest age for forested pixels ≤150 years old was 68 years while median forest age reported in the NFI was 73 years, with regional variability matching expectations related to disturbance regimes and productivity. Spatially explicit maps of forest age provide important information for understanding forest ecosystems and can be used to inform a wide range of policy, science, and management needs.

Remote sensing of fluorescent humification levels and its potential environmental linkages in lakes across China

The pollution or eutrophication affected by dissolved organic matter (DOM) composition and sources of inland waters had attracted concerns from the public and government in China. Combined with remote sensing techniques, the fluorescent DOM (FDOM) parameters accounted for the important part of optical constituent as chromophoric dissolved organic matter (CDOM) was a useful tool to trace relative DOM sources and assess the trophic states for large-scale regions comprehensively and timely. Here, the objective of this research is to calibrate and validate a general model based on Landsat 8 OLI product embedded in Google Earth Engine (GEE) for deriving humification index (HIX) based on EEMs in lakes across China. The Landsat surface reflectance was matched with 1150 pairs fieldtrip samples and the nine sensitive spectral variables with good correlation with HIX were selected as the inputs in machine learning methods. The calibration of XGBoost model (R2 = 0.86, RMSE = 0.29) outperformed other models. Our results indicated that the entire dataset of HIX has a strong association with Landsat reflectance, yielding low root mean square error between measured and predicted HIX (R2 = 0.81, RMSE = 0.42) for lakes in China. Finally, the optimal XGBoost model was used to calculate the spatial distribution of HIX of 2015 and 2020 in typical lakes selected from the Report on the State of the Ecology and Environment in China. The significant decreasing of HIX from 2015 to 2020 with trophic states showed positive control of humification level of lakes based on the published document of Action plan for prevention and control of water pollution in 2015 of China. The calibrated model would greatly facilitate FDOM monitoring in lakes, and provide indicators for relative DOM sources to evaluate the impact of water protection measures or human disturbance effect from Covid-19 lockdown, and offer the government supervision to improve the water quality management for lake ecosystems.

Quantifying Mangrove Extent Using a Combination of Optical and Radar Images in a Wetland Complex, Western Region, Ghana

The classification of mangrove forests in tropical coastal zones, based only on passive remote sensing methods, is hampered by mangrove complexities, topographical considerations, and cloud cover effects, among others. This paper reports on a study that combines optical and radar data to address the challenges of distinguishing mangrove stands in cloud-prone regions. The Google Earth Engine geospatial processing platform was used to extract multiple scenes of Landsat surface reflectance Tier 1 and synthetic aperture radar (C-band and L-band). The images were enhanced by creating a feature that removes clouds from the optical data and using speckle filters to remove noise from the radar data. The random forest algorithm proved to be a robust and accurate machine learning approach for mangrove classification and assessment. Classification was evaluated using three scenarios: classification of optical data only, classification of radar data only, and combination of optical and radar data. Our results revealed that the scenario that combines optical and radar data performed better. Further analysis showed that about 16.9% and 21% of mangrove and other vegetation/wetland cover were lost between 2009 and 2019. Whereas water body and bare land/built-up areas increased by 7% and 45%, respectively. Accuracy was evaluated based on the three scenarios. The overall accuracy of the 2019 classification was 98.9% (kappa coefficient = 0.979), 84.6% (kappa coefficient = 0.718), and 99.1% (kappa coefficient = 0.984), for classification of optical data only, classification of radar data only, and combination of optical and radar data, respectively. This study has revealed the potential to map mangroves correctly, enabling on-site conservation practices in the climate change environment.

Permanganate Index Variations and Factors in Hongze Lake from Landsat-8 Images Based on Machine Learning

The permanganate index (CODMn), defined as a comprehensive index to measure the degree of surface water pollution by organic matter and reducing inorganic matter, plays an important role in indicating water pollution and evaluating aquatic ecological health. However, remote sensing monitoring of water quality is presently focused mainly on phytoplankton, suspended particulate matter, and yellow substance, while there is still great uncertainty in the retrieval of CODMn. In this study, the Landsat-8 surface reflectance data set from Google Earth Engine and in situ CODMn measurements were matched. The support vector regression (SVR) machine learning model was calibrated using the matchups. With the SVR model, this study estimates the CODMn in Hongze Lake, presents the historical spatiotemporal CODMn distributions, and discusses the affecting factors of the change trend of the CODMn in Hongze Lake. The results showed that the SVR model adequately estimated CODMn, with a sum squared error of 1.49 mg2/L2, a coefficient of determination (R2) of 0.95, and a root mean square error of 0.15 mg/L. CODMn in Hongze Lake was high in general and showed a decreasing trend in the past decade. Huai River, Xinsu River, and Huaihongxin River were still the main sources of oxygen-consuming pollutants in Hongze Lake. The wetland natural reserve near Yugou had a significant effect on reducing CODMn. This study provides not only a scientific reference for the management of CODMn in Hongze Lake, but also a feasible scheme for remote sensing monitoring of CODMn in inland water.

Heterogenous controls on lake color and trends across the high-elevation U.S. Rocky Mountain region

Global change may contribute to ecological changes in high-elevation lakes and reservoirs, but a lack of data makes it difficult to evaluate spatiotemporal patterns. Remote sensing imagery can provide more complete records to evaluate whether consistent changes across a broad geographic region are occurring. We used Landsat surface reflectance data to evaluate spatial patterns of contemporary lake color (2010–2020) in 940 lakes in the U.S. Rocky Mountains, a historically understudied area for lake water quality. Intuitively, we found that most of the lakes in the region are blue (66%) and were found in steep-sided watersheds (>22.5°) or alternatively were relatively deep (>4.5 m) with mean annual air temperature (MAAT) <4.5°C. Most green/brown lakes were found in relatively shallow sloped watersheds with MAAT ⩾4.5°C. We extended the analysis of contemporary lake color to evaluate changes in color from 1984 to 2020 for a subset of lakes with the most complete time series (n = 527). We found limited evidence of lakes shifting from blue to green states, but rather, 55% of the lakes had no trend in lake color. Surprisingly, where lake color was changing, 32% of lakes were trending toward bluer wavelengths, and only 13% shifted toward greener wavelengths. Lakes and reservoirs with the most substantial shifts toward blue wavelengths tended to be in urbanized, human population centers at relatively lower elevations. In contrast, lakes that shifted to greener wavelengths did not relate clearly to any lake or landscape features that we evaluated, though declining winter precipitation and warming summer and fall temperatures may play a role in some systems. Collectively, these results suggest that the interactions between local landscape factors and broader climatic changes can result in heterogeneous, context-dependent changes in lake color.

Quantifying uncertainty in high resolution biophysical variable retrieval with machine learning

The estimation of biophysical variables is at the core of remote sensing science, allowing a close monitoring of crops and forests. Deriving temporally resolved and spatially explicit maps of parameters of interest has been the subject of intense research. However, deriving products from optical sensors is typically hampered by cloud contamination and the trade-off between spatial and temporal resolutions. In this work we rely on the HIghly Scalable Temporal Adaptive Reflectance Fusion Model (HISTARFM) algorithm to generate long gap-free time series of Landsat surface reflectance data by fusing MODIS and Landsat reflectances. An artificial neural network is trained on PROSAIL inversion to predict monthly biophysical variables at 30 m spatial resolution with associated, realistic uncertainty bars. We emphasize the need for a more thorough analysis of uncertainty, and propose a general and scalable approach to combine both epistemic and aleatoric uncertainties by exploiting Monte Carlo (MC) dropout techniques from the trained artificial network and the propagation of HISTARFM uncertainties through the model, respectively. A model recalibration was performed in order to provide reliable uncertainties. We provide new high resolution products of several key variables to quantify the terrestrial biosphere: Leaf Area Index (LAI), Fraction of Absorbed Photosynthetically Active Radiation (FAPAR), Canopy Water Content (CWC) and Fractional Vegetation Cover (FVC) are at 30 m Landsat spatial resolution and over large continental areas. Two study areas are considered: the large heterogeneous but moderately cloud covered contiguous United States, and the homogeneous but largely cloud covered Amazonia. The produced vegetation products largely agree with the test dataset (R = 0.90, RMSE = 0.80 m2/m2 and ME = 0.12 m2/m2 for LAI, and R = 0.98, RMSE = 0.07 and ME = 0.01 for FAPAR) providing low error and high accuracy. Additionally, the validation considers a thorough comparison with operational and largely validated medium resolution products, such as the Moderate-Resolution Imaging Spectroradiometer (MODIS) and Copernicus Global Land Service. Our products presented a good agreement and consistency with both MODIS (R = 0.84 and R = 0.85 for LAI and FAPAR, respectively) and Copernicus (R = 0.92 and R = 0.91 for LAI and FAPAR, respectively). To foster a wider adoption and reproducibility of the methodology we provide an application in GEE and source code at:https://github.com/IPL-UV/ee_BioNet/