Land Cover Types Research Articles

Evaluating the Two-Source Energy Balance Model Using MODIS Data for Estimating Evapotranspiration Time Series on a Regional Scale

Estimating daily continuous evapotranspiration (ET) can significantly enhance the monitoring of crop stress and drought on regional scales, as well as benefit the design of agricultural drought early warning systems. However, there is a need to verify the models’ performance in estimating the spatiotemporal continuity of long-term daily evapotranspiration (ETd) on regional scales due to uncertainties in satellite measurements. In this study, a thermal-based two-surface energy balance (TSEB) model was used concurrently with Terra/Aqua MODIS data and the ERA5 atmospheric reanalysis dataset to calculate the surface energy balance of the soil–canopy–atmosphere continuum and estimate ET at a 1 km spatial resolution from 2000 to 2022. The performance of the model was evaluated using 11 eddy covariance flux towers in various land cover types (i.e., savannas, woody savannas, croplands, evergreen broadleaf forests, and open shrublands), correcting for the energy balance closure (EBC). The Bowen ratio (BR) and residual (RES) methods were used for enforcing the EBC in the EC observations. The modeled ET was evaluated against unclosed ET and closed ET (ETBR and ETRES) under clear-sky and all-sky observations as well as gap-filled data. The results showed that the modeled ET presented a better agreement with closed ET compared to unclosed ET in both Terra and Aqua datasets. Additionally, although the model overestimated ETd across all different land cover types, it successfully captured the spatiotemporal variability in ET. After the gap-filling, the total number of days compared with flux measurements increased substantially, from 13,761 to 19,265 for Terra and from 13,329 to 19,265 for Aqua. The overall mean results including clear-sky and all-sky observations as well as gap-filled data with the Aqua dataset showed the lowest errors with ETRES, by a mean bias error (MBE) of 0.96 mm.day−1, an average mean root square (RMSE) of 1.47 mm.day−1, and a correlation (r) value of 0.51. The equivalent figures for Terra were about 1.06 mm.day−1, 1.60 mm.day−1, and 0.52. Additionally, the result from the gap-filling model indicated small changes compared with the all-sky observations, which demonstrated that the modeling framework remained robust, even with the expanded days. Hence, the presented modeling framework can serve as a pathway for estimating daily remote sensing-based ET on regional scales. Furthermore, in terms of temporal trends, the intra-annual and inter-annual variability in ET can be used as indicators for monitoring crop stress and drought.

Introduction pages, Nova Geodesia 4(4), 2024

Nova Geodesia (https://novageodesia.ro), Issue 4, Volume 4, 2024: The papers published in this issue represent interesting novelties in different topics of geodetic science or related fields. Nova Geodesia publishes significant papers in geodesy and close topics related to cartography, engineering projects and construction, planetology, hydrography, geography, sociodemographic factors, urban administration, planning and environment, landscape, biodiversity, and ecology. Among the exciting articles, we invite readers to find news about: Evaluating the performance of various models for deriving orthometric heights using Global Navigation Satellite System (GNSS) data; Carbon stocks under different land cover types in the Hallaydeghe wildlife reserve, northeastern Ethiopia; Sorghum (Sorghum bicolor) yield and water use efficiency depending on planting techniques and soil mulching in drought-prone areas, Babile District, Ethiopia; Colour selection in Lantana camara L. blooms: A study on butterfly attraction; Assessment of treatment water effects on macrobenthic invertebrates in an urban stream; Influence of geographical provenance, biostimulatory treatments and their interaction on the seed germination of Quercus robur L.

Comparative study of machine learning methods for mapping forest fire areas using Sentinel-1B and 2A imagery

The study focuses on identifying fireburning and burnt areas in a large-scale forest fire that occurred in Xintian County, China, in October 2022. To investigate the adaptability of machine learning methods in various scenarios for mapping forest fire areas, this study presents a comparative study on the recognition and mapping accuracy of three machine learning algorithms, namely, Support Vector Machine (SVM), Random Forest (RF), and Neural Network (NN), based on Sentinel-1B and 2A imagery. Initially, three sets of pre-fire, during-fire, and post-fire remote sensing data were preprocessed. Various feature parameters from Sentinel-1B and 2A imagery were combined to identify firerelated land cover types. The experimental results revealed that: (i) During the pre-fire period, the SVM method demonstrated superior accuracy compared to the other two methods. The combination of spectral and Normalized Difference Vegetation Index (NDVI) features achieved an optimal accuracy for identifying forest areas with an overall accuracy (OA) of 93.52%. (ii) In the during-fire period, RF method exhibited higher accuracy compared to the other two methods with peak fire identification accuracy reached by combining spectral and Normalized Burn Ratio (NBR) index features at an OA of 95.43%. (iii) In the post-fire period, SVM demonstrated superior accuracy compared to other methods. The highest accuracy of 94.97% was achieved when combining spectral and radar features from Sentinel-1B imagery, highlighting the effectiveness of using spectral and radar backward scattering coefficients as feature parameters to enhance forest fire recognition accuracy for burnt areas. These findings suggest that appropriate machine learning algorithms should be employed under different conditions to obtain more precise identification of forest fire areas. This study provides technical support and empirical evidence for extracting and mapping forest fire areas while assessing damage caused by fires.

Evolution of landscape patterns and driving forces in rural areas of the Fuchun River Basin

At the optimal analysis granularity and extent, this study reveals the spatiotemporal evolution characteristics of landscape patterns in rural areas of the Fuchun River Basin from 1990 to 2020. It also explores the spatiotemporal differentiation of driving forces using Geodetector and Geographically Weighted Regression. The results indicate that: 1) The optimal analysis granularity for the study area is 30 m × 30 m, and the optimal analysis extent is 200 m. 2) Compared to the 1990s, the past two decades have seen an increased degree of landscape fragmentation, higher isolation, and stronger human disturbance in the study area, with significant differentiation. These changes are most pronounced in the semi-urbanized areas along the river and in the northern regions. 3) Digital Elevation Model and land cover type are the most critical driving factors. Over time, the explanatory power of most driving factors shows an overall increasing trend. 4) There is a significant spatial heterogeneity in the driving forces. The influence of the Digital Elevation Model and slope is stronger in the north and south, weaker in the central region, while the influence of temperature increases from north to south. The impact of land cover type is stronger in the central area and weaker in the periphery. This study provides a reference for the scientific use of land resources and the protection of ecological spaces in rural areas within highly urbanized regions at the basin scale.

Pengaruh Setting kawasan Tamansari terhadap Temperature Humidity Index

Title: The Effect of Environmental Setting of Tamansari on Temperature Humidity Index Geometric shape, land cover type, and density impact increasing global temperature and microclimate comfort. Tamansari was originally a park, garden, and bathhouse for the King and his family. Today, Tamansari is a heritage tourism area surrounded by dense residential areas in the middle of Yogyakarta City. This study aims to determine how the setting of the Tamansari area affects changes in temperature and relative humidity, which impact changes in the Temperature Humidity Index (THI) on the micro-scale of the Tamansari area. The method used is descriptive quantitative analysis, and calculation analysis was performed using the (THI) equation. The study results showed that the THI value in the Tamansari area was uncomfortable, with a THI value of 27.77-32.52°C. Building coverage ratio, population, and human activities affect the THI value. The THI value can also change due to weather factors that affect the physical condition of the area, especially in open areas.

Unlocking the ecohydrological dynamics of vegetation growth's impact on Terrestrial Water Storage trends across the China-Pakistan Economic Corridor

Studying the influence of vegetation dynamics on water storage is fundamental for efficiently managing ecosystems in dryland areas. Therefore, this study was designed to investigate the ecohydrological dynamics of vegetation growth's impact on Terrestrial Water Storage (TWS) trends across the China-Pakistan Economic Corridor (CPEC). Exploring vegetation growth's effect on TWS utilizes remote sensing and statistical methods such as generalized additive model (GAM), random forest, Mann-Kendall test, Sen's slope, and Partial correlation coefficient. Our results revealed a consistent increase in vegetation cover in semi-arid and dry sub-humid regions, especially in croplands, from 1986 to 2020. However, a noticeable finding was that TWS has declined by approximately 39.65 % in the study area. On the other hand, forests have displayed resilience by mitigating the effects of climate change and human activities through hydraulic memory effects. Additionally, soil moisture has decreased in various land cover types, with croplands experiencing the most significant decrease. Similarly, increased vegetation growth in greening drylands significantly negatively impacts terrestrial water storage, with correlations of −0.31 for terrestrial water storage anomaly (TWSA), −0.26 for root zone soil moisture (RZSM), and −0.29 for surface soil moisture (SSM). Meanwhile, evapotranspiration (ET) also had negative correlations with TWSA, RZSM, and SSM, with standardized coefficients of −0.23, −0.16, and −0.11, respectively. These findings exposed the study area's vegetation interaction with land cover and hydrological dynamics. Addressing these hydrological imbalances will help ensure sustainable ecological management in the study area.

Modeling carbon dynamics from a heterogeneous watershed in the mid-Atlantic USA: A distributed-calibration and independent verification (DCIV) approach

The terrestrial ecosystem plays a vital role in regulating regional and global carbon budgets. Ecosystem models are extensively employed to estimate carbon fluxes across different spatial scales. However, there remains a need to reduce the uncertainties associated with model parameterization and input data. To address these limitations, we assessed a distributed-calibration and independent-verification (DCIV) approach that uses (1) remotely sensed net primary production (NPP) and evapotranspiration (ET) data from the Moderate Resolution Imaging Spectroradiometer (MODIS), (2) multi-site eddy covariance net ecosystem exchange (NEE) data; and (3) field sampling of soil organic carbon (SOC) and aboveground biomass (ABG) data to improve the overall predictability of carbon fluxes for the different land use and land cover (LULC) types at a watershed scale. The DCIV approach was applied to an advanced version of the Soil and Water Assessment Tool (SWAT)-Carbon (or SWAT-C), equipped with Century-based SOC algorithms to simulate carbon dynamics for watersheds with heterogeneous vegetation. The objective of the modeling effort was to assess carbon stocks and fluxes under different land management scenarios for a 3000-acre experimental farm and forest preserve in the northeastern United States. Our study showed that a large SOC stock of at least 100 tons ha−1 is stored under mixed forest, deciduous, shrubland, and floodplain (grass). Our study also showed that converting floodplain (grass) to deciduous forest has the potential to increase CO2 uptake (-NEE) by an order of three magnitude and ABG by 77 %, leading to an increased SOC stock of 23 % after twenty years. Similarly, we found that converting ungrazed grassland to grazed pasture leads to a non-statistically decreasing trend of SOC, especially in the 0–30 cm soil layer. Thus, the methodology used in this study can be applied to improve carbon dynamic prediction from a heterogeneous watershed at a regional scale.

Importance of the interplay between land cover and topography in modeling habitat selection

Anthropogenic land cover change is one of the greatest threats to the persistence of organisms. Therefore, identifying and safeguarding optimal land covers for declining organisms is a major challenge of the Anthropocene. Priority land covers are typically identified through habitat selection studies. These traditionally use land cover and/or topographic features as separate, additive predictors of organismal occurrence, despite the fact that each land cover is superimposed on specific topographic features, such as elevation or slope, which may affect its attractiveness. Here, we tested the importance of considering the largely overlooked interplay between land cover and topography, using as a model a golden eagle Aquila chrysaetos population which is sharply declining due to human-induced land cover alteration. We found that the overlay of a key land cover type (old-growth forest) on underlying terrain known to be favorable for the population (steep slopes at higher elevations than the nest) dramatically increased its attractiveness. Conversely, the matching of the same land cover with unfavorable topography (gentle slopes at lower elevations) deteriorated its attractiveness for the population. Thus, underlying topography acted as the major determinant of land cover suitability for the eagle population. The conservation implications could be profound, because modeling land cover per se could waste conservation resources on low quality sites (old-growth forest on gentle terrain at low elevations) with unlikely benefits for the threatened eagle population. We expect topographic modulators of land cover quality to be more common than previously appreciated in many or most study systems, as numerous organisms inhabiting terrestrial and aquatic environments, regardless of taxonomic group, exhibit selectivity for specific topographic features and specific land cover types. In conclusion, we encourage modelers to take more into account underlying modulators that may drive differences in quality within the same land cover. This would make wildlife-habitat models more realistic, improve their applicability, and enhance the cost-effectiveness of conservation efforts.

Integration of Landsat-8 OLI/TIRS and Sentinel-1A PolSAR for analyzing land surface temperature and its anomalies linked to ENSO in Surakarta, Indonesia

The increase in urban buildings leads to degraded vegetated areas, resulting in higher surface radiation and air temperatures. The rise of land surface temperature (LST) is also influenced by land cover changes and global climate related to the ENSO (El Nino-Southern Oscillation) phenomena. This study is the first to combine active and passive sensors to analyze LST anomalies linked to ENSO-related land cover change on a time series approach. Conducted from August 2018 to 2023 in an urban area dominated by buildings, we used Python programming to extract LST from the Landsat-8 OLI/TIRS passive sensor with a Mono-window algorithm. Meanwhile, the land cover classification was performed by Sentinel-1A active sensor imagery using polarimetric decomposition with unsupervised Wishart. The LST and land cover results were equalized to 30 m spatial resolution for regression and anomaly analysis based on reported ENSO phenomena. The results revealed that land cover type significantly affected LST variation during the study period, proven by the significance value of each land cover type being less than 0.05 and showing a positive correlation. However, the correlation is low, meaning that land cover change is not the dominant factor causing LST change. The low correlation caused by El Nino and La Nina, contributed more to the change in LST during the study period. The integrated method can overcome the weakness of passive sensors in penetrating clouds, contribute to a broader knowledge of the factors causing LST changes, and provide effective early mitigation strategies against the threat of future climate change crises.

Impact of wildfires on the suitability habitat of the lowland tapir in a native savanna landscape of northern South America

Fire plays a crucial role in shaping multiple ecological processes in native savannas. Still, the natural fire regime in the Neotropical savanna of South America is being altered by climate variability, resulting in reduced precipitation and increased mean temperatures. In recent years, an increase in the duration, intensity, frequency, and extent of wildfires in South American savannas has been reported. However, knowledge of the ecology of wildfires in the savanna landscapes and their impact on the habitats of large mammals is limited or non-existent. This study investigates the impact of wildfires on the suitability habitat for Tapirus terrestris in a native high plains native savanna landscape in the department of Vichada in eastern Colombia. We used the MaxEnt algorithm to model the ecological niche of the lowland tapir and then classify land covers into low, medium, and high probability of species occurrence. We then assessed the representativeness of the occurrence of wildfires and the frequency of detection of burned areas within the land covers classified into the suitability model of lowland tapir. Our results indicate that wildfires significantly affect high-suitability habitats such as forest and shrubland covers. The effects of fire on these land cover types would reduce the resource supply for lowland tapir and cause the displacement of individuals to new areas which could increase the exposure to threats such as hunting and pathogens from domestic animals. This research underscores the need for targeted conservation strategies to mitigate wildfire impacts and preserve critical habitats for vulnerable species.

Implementing Parameterizations of Sub‐Pixel Terrain Long‐Wave Radiative Effect for FY‐4A/AGRI Infrared Data Assimilation Over Land Based on the Community Radiative Transfer Model

AbstractThe sub‐pixel terrain long‐wave radiative effect (STLRE) plays an important role in the satellite surface‐sensitive infrared brightness temperature (TB) simulation over rugged areas, however, it is absent in current radiative transfer models. This study incorporates a STLRE scheme in the Community Radiative Transfer Model (CRTM) for the FY‐4A/Advanced Geosynchronous Radiation Imager (AGRI). The total surface downwelling long‐wave radiance (SDLR) over rugged areas at sub‐pixel is composed of the radiance from the overlying atmosphere under plane‐parallel radiation condition and the radiance emitted from surrounding terrains, weighted by the sky view factor (SKV). Based on the homogenous assumption, a STLRE scheme is established through constructing terrain correction factors at AGRI pixel scale. The ExpCTL/ExpTOPO experiments with the CRTM adopting the plane‐parallel scheme/STLRE scheme are conducted to evaluate the simulations of AGRI clear‐sky TB over the rugged areas. Results show that considering the STLRE can significantly reduce the orography dependent simulation biases in the plane‐parallel scheme. The reductions of biases for the AGRI channel 11–13 regionally averaged over the rugged areas are 64.2%, 72.7% and 67.6%, respectively. Stable correction performance is found at different times of the day. Meanwhile, the corrections of SDLR are more obvious in winter than in summer, and in barren condition than in the other land cover types. This is theoretically attributed to the land‐air temperature difference and the land surface emissivity, respectively. This pilot study improves the observation operator for infrared TB data assimilation over complex terrain and exhibits broad promotion and application value.

Raccoon density estimation from camera traps for raccoon rabies management

AbstractDensity estimation for unmarked animals is particularly challenging, yet density estimates are often necessary for effective wildlife management. Raccoons (Procyon lotor) are the primary terrestrial wildlife reservoir for Lyssavirus rabies within the United States. The raccoon rabies variant (RRVV) is actively managed at landscape scales using oral rabies vaccination (ORV) within the eastern United States. To effectively manage RRVV, it is important to know the density of raccoons to appropriately scale the density of ORV baits distributed on the landscape. We compared methods to estimate raccoon densities from camera‐trap data versus more intensive capture‐mark‐recapture (CMR) estimates across 2 land cover types (upland pine and bottomland hardwood) in the southeastern United States during 2019 and 2020. We evaluated the effect of alternative camera configurations and durations of camera trapping on density estimates and used an N‐mixture model to estimate raccoon densities, including covariates on abundance and detection. We further compared different methods of scaling camera‐based counts, with the maximum number of raccoons seen on any given image within a day best explaining density. Camera‐trap density estimates were moderately correlated with CMR estimates (r = 0.56). However, densities from camera‐trap data were more reliable when classifying category of density as an index used to inform management (83% correct when compared to CMR estimates), although the densities in our study fell into the 2 lowest density classes only. Using more cameras reduced bias and uncertainty around density estimates; however, if ≤6 camera traps were used at a site, a line transect approach proved less biased than a grid design. Camera trapping should be conducted for at least 3 weeks for more accurate estimates of raccoon population density in our study area (<5% bias). We show that camera‐trap data can be used to assign raccoon densities to management‐relevant density index bins, but more studies are needed to ensure reliability across a greater range of environmental conditions and raccoon densities.

The spatiotemporal distribution patterns and impact factors of bird species richness: A case study of urban built-up areas in Beijing, China

This study aims to investigate the distribution patterns of bird species in urban built-up areas and their impact factors from multiple dimensions, including spatiotemporal distribution, environmental changes, and anthropogenic disturbances, as well as overall distribution and local hotspots. Leveraging geographic distribution data of 383 bird species from 70 families and 21 orders, the study recorded dynamic changes in bird distribution within urban built-up areas characterized by highly heterogeneous core areas and suburban areas. It examined species distribution across different seasons and land cover types, evaluated population fluctuations based on migratory behaviors, and assessed the relative abundance of bird families and species in hotspot areas. Additionally, this study employed three tree-based machine learning algorithms—Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boosting (XGBoost)—to investigate the influence of environmental factors on bird species distribution within urban built-up areas. The findings showed that, temporally, the number of observed bird species in the study area peaked in May and September, while the lowest numbers of species (54.83 %) and individuals (5.45 %) were recorded during the high-temperature period from June to August. Spatially, (1) woodlands, including stable mature forests, unstable juvenile forests, and sparse vegetation, recorded the highest frequencies of bird observations (2,053 times), bird species (369 species), and bird individuals (38,623 individuals); (2) comprehensive parks, where bird species demonstrated higher adaptability to anthropogenic disturbances, experienced a more significant decline in species richness compared to country parks; and (3) the number of bird species in hotspot areas located in the core areas, which developed earlier, has decreased annually, while the number of species in suburban areas, which developed later, has increased annually. In terms of impact factors, water area (WA), point of interest kernel density (DPOI), relative humidity (RH), green space area (GSA), and nighttime light pollution (NL) were identified as the five most important environmental factors affecting bird species richness. These results suggest that although birds exhibit some adaptability to environmental changes and anthropogenic disturbances, species distribution remains significantly impacted. This research aims to provide a planning framework for future urban development that promotes biodiversity and adaptability, thereby bolstering the resilience of urban ecosystems.

Triple Collocation-Based Uncertainty Analysis and Data Fusion of Multi-Source Evapotranspiration Data Across China

Accurate estimation of evapotranspiration (ET) is critical for understanding land-atmospheric interactions. Despite the advancement in ET measurement, a single ET estimate still suffers from inherent uncertainties. Data fusion provides a viable option for improving ET estimation by leveraging the strengths of individual ET products, especially the triple collocation (TC) method, which has a prominent advantage in not relying on the availability of “ground truth” data. In this work, we proposed a framework for uncertainty analysis and data fusion based on the extended TC (ETC) and multiple TC (MTC) variants. Three different sources of ET products, i.e., the Global Land Evaporation and Amsterdam Model (GLEAM), the fifth generation of European Reanalysis-Land (ERA5-Land), and the complementary relationship model (CR), were selected as the TC triplet. The analyses were conducted based on different climate zones and land cover types across China. Results show that ETC presents outstanding performance as most areas conform to the zero-error correlations assumption, while nearly half of the areas violate this assumption when using MTC. In addition, the ETC method derives a lower root mean square error (RMSE) and higher correlation coefficient (Corr) than the MTC one over most climate zones and land cover types. Among the ET products, GLEAM performs the best, while CR performs the worst. The merged ET estimates from both ETC and MTC methods are generally superior to the original triplets at the site scale. The findings indicate that the TC-based method could be a reliable tool for uncertainty analysis and data fusion.

Robust vegetation detection using RGB colour composites and isoclust classification of the Landsat TM image

The paper presents the application of ArcGIS for environmental modelling of the landscapes in northern Iceland (17.00°W–23.00°W, 64.30°N–67.00°N). The aim was to explore the vegetation distribution by NDVI and ISOCLUST classification of the land cover types. Data include the Landsat TM image. Freely available satellite remote sensing data from the Landsat mission have been processed by GIS to deliver information on land cover types from image classification and NDVI vegetation index. Landsat products provide geospatial data on regional scale with moderate temporal (weekly) and spatial (30–10 m) resolution, making them useful for environmental monitoring and landscape studies. The tools include the ArcGIS software used for raster processing. Data processing was performed in the three steps: 1) comparative analysis of the visualized sixteen band combinations to assess the distinguishability of vegetation and other land cover types in colour composites; 2) computed NDVI standardized vegetation index; 3) unsupervised classification of the Landsat TM by the ISOCLUST algorithm. Large glaciers Hofsjökull and Langjökull were detected on various colour composites, and the visibility of the water/land borders is assessed (Blöndulón lake), agricultural areas near the Varmahlíð, vegetated areas around the Akrahreppur municipality. Computing the NDVI and using ISOCLUST by ArcGIS software enabled to distinguish various land cover types and map landscapes in the study area. The computed NDVI shown the presence and condition of vegetation, that is, a relative biomass in the area of northern Iceland. The NDVI was used based on the contrast of the two channels from a multispectral Landsat TM raster data.

Multidimensional analysis of NDVI dynamics in response to climate and land use/land cover change in Northwest Algeria

Abstract Climate change has emerged as a major concern at both regional and global scales in recent decades. Northwestern Algeria is particularly vulnerable as a semi-arid zone, where changes in climate and land cover (LC) will have a significant impact on vegetation in the long term. This study analyses, through a multidimensional approach, the influence of climate change and LC on vegetation dynamics. Hierarchical partitioning (HP) analysis was conducted to determine the most influential climatic variables (precipitation, temperature) on the dynamics of NDVI. The results show that the annual NDVI shows a fluctuating spatial trend between decrease and increase in different regions. Trends in seasonal NDVI are spatially varied and less uniform. Variations in precipitation are stable, while temperatures show clear and consistent significant increases across the region, with a general tendency to increase (p<0.01) in spring and summer. In mountainous areas, NDVI shows an increasing trend both annually and seasonally. The correlation (r²) between NDVI, temperature and precipitation (0.75–1.0) over the different seasons reveals significant seasonal and regional variability. LC transition patterns also influence spatio-temporal trends in vegetation cover. They reveal that the rate of change of NDVI varies between LC types and regions, with resilience in forests and grasslands. These variations have significant implications for vegetation dynamics, as observed by NDVI.

The effect of land use land cover types on MODIS land surface temperature in Ghana

The effect of land use land cover types on MODIS land surface temperature in Ghana

Comparative Validation and Misclassification Diagnosis of 30-Meter Land Cover Datasets in China

Land cover maps with high accuracy are essential for environmental protection and climate change research. The 30-meter-resolution maps, with their better resolution and longer historical records, are extensively utilized to assess changes in land cover and their effects on carbon storage, land–atmosphere energy balance, and water cycle processes. However, current data products use different classification methods, resulting in significant classification inconsistency and triggering serious disagreements among related studies. Here, we compared four mainstream land cover products in China, namely GLC_FCS30, CLCD, Globeland30, and CNLUCC. The result shows that only 50.34% of the classification results were consistent across the four datasets. The differences between pairs of datasets ranged from 21.10% to 37.53%. Importantly, most inconsistency occurs in transitional zones among land cover types sensitive to climate change and human activities. Based on the accuracy evaluation, CLCD is the most accurate land cover product, with an overall accuracy reaching 86.98 ± 0.76%, followed by CNLUCC (81.38 ± 0.87%) and GLC_FCS30 (77.83 ± 0.80%). Globeland30 had the lowest accuracy (75.24 ± 0.91%), primarily due to misclassification between croplands and forests. Misclassification diagnoses revealed that vegetation-related spectral confusion among land cover types contributed significantly to misclassifications, followed by slope, cloud cover, and landscape fragmentation, which affected satellite observation angles, data availability, and mixed pixels. Automated classification methods using the random forest algorithm can perform better than those that depend on traditional human–machine interactive interpretation or object-based approaches. However, their classification accuracy depends more on selecting training samples and feature variables.

Understanding ecosystem services of detailed forest and wetland types using remote sensing and deep learning techniques in Northern China

Spanning both temperate and sub-frigid zones, Northeast China boasts typical boreal forests and abundant wetland resources. Because of these attributes, the region is critically significant for global climate regulation, carbon sequestration, and biodiversity preservation. While existing research explores the ecosystem service (ESs) functions of different land cover types, a thoroughly in-depth investigation into the ESs of detailed forest and wetland types is essential. This study addresses this deficiency by combining remote sensing and deep learning techniques, employing a lightweight convolutional neural network (CNN) model and a decision tree for the large-scale classification of forests and wetlands. The ESs of various forest and wetland types—encompassing habitat quality, carbon stock, and soil retention—were assessed during two periods (2008 and 2018) in Heilongjiang Province. Key factors determinants of ESs were identified using the Geodetector tool. The results indicated an overall accuracy of 0.77 in 2008 and 0.78 in 2018 for forest type classification, and 0.88 in 2008 and 0.87 in 2018 for wetland type classification. In particular, the transition from mixed broadleaf forests to mixed coniferous-broadleaf forests dominated changes from 2008 to 2018, probably due to natural succession. Among forest types, Mongolian oak forests exhibited the highest carbon stock and soil retention capacity owing to their rapid growth and deep root systems. Mixed broadleaf forests exhibited superior habitat quality, suggesting minimal disturbance. Habitat quality, carbon stock, and soil retention were found to be significantly influenced by human activity, atmospheric quality, and topographic factors, respectively. By leveraging remote sensing and deep learning methodologies, this study offers a comprehensive analysis of forests and wetlands, elucidating the nuanced ecosystem roles of specific forest and wetland types.

Integral recovery of an urban watershed through the implementation of nature-based solutions

Unplanned urban expansion in Latin America has profoundly disturbed the natural hydrological processes of watersheds, particularly in tropical regions with extreme weather conditions. This has resulted in substantial runoff volumes, necessitating extensive engineering interventions to avoid damage. Urbanization has led not only to flooding problems, but also to a lack of recreational spaces for residents and an ecological degradation, negatively affecting the watersheds health. This study shows a methodology for creating scenarios making use of available sites for Nature-based-Solutions (NbS). Taking account of existing land cover, the methodology harnesses NbS multifunctionality to achieve hydrological, ecological and social benefits counteracting the threats to the health of a 23 km2 urban tropical watershed located in Costa Rica. In these scenarios, suitable public roads in residential areas were used as sites for implementing permeable pavement, bioretention cells, infiltration trenches and street planters, while appropriate flat roofs and car parks were converted into green roofs and permeable pavements in industrial areas. In addition, several green public spaces were transformed into multifunctional storage areas. Hydrological and hydraulic modeling was used to evaluate the performance of each scenario, while the increase in green area of each scenario was used as an indicator of the ecological and social benefits. The concurrent implementation of the residential, industrial and green space scenarios provides an area available for NbS of 165.3 ha (7.2% of the watershed area), potentially leading to a decrease of up to 63.7% in peak flow and up to 46.2% in runoff volume for a specific rainfall event. As haphazard urban growth is a common feature of most urban watersheds in Latin America, this methodology, based on the use of different land cover types to create scenarios for NbS implementation, is transferable to and adaptable for other tropical urban watersheds.