Effect of digital elevation models on monitoring slope displacements in open-pit mine by differential interferometry synthetic aperture radar

Displacement monitoring in open-pit mines is one of the important tasks for safe management of mining processes.Differential interferometric synthetic aperture radar(DInSAR),mounted on an artificial satellite,has the potential to be a cost-effective method for monitoring surface displacements over extensive areas,such as open-pit mines.DInSAR requires the ground surface elevation data in the process of its analysis as a digital elevation model(DEM).However,since the topography of the ground surface in open-pit mines changes largely due to excavations,measurement errors can occur due to insufficient information on the elevation of mining areas.In this paper,effect of different elevation models on the accuracy of the displacement monitoring results by DInSAR is investigated at a limestone quarry.In addition,validity of the DInSAR results using an appropriate DEM is examined by comparing them with the results obtained by global positioning system(GPS)monitoring conducted for three years at the same limestone quarry.It is found that the uncertainty of DEMs induces large errors in the displacement monitoring results if the baseline length of the satellites between the master and the slave data is longer than a few hundred meters.Comparing the monitoring results of DInSAR and GPS,the root mean square error(RMSE)of the discrepancy between the two sets of results is less than 10 mm if an appropriate DEM,considering the excavation processes,is used.It is proven that DInSAR can be applied for monitoring the displacements of mine slopes with centimeter-level accuracy.

Regionalization of wheat powdery mildew oversummering in China based on digital elevation

Blumeria graminis f. sp. tritici, the pathogen that causes wheat powdery mildew, is one of the most important diseases affecting wheat production in China, and the oversummering is the key stage of wheat powdery mildew epidemic. The more oversummering regionalization of wheat powdery mildew has played an important role in disease prediction, prevention and control. In this study, we analyzed the correlation between oversummering data of wheat powdery mildew and the meteorological factors over the past years, and determined that temperature was the key meteorological factor influencing oversummering of wheat powdery mildew. The average temperature at which wheat powdery mildew growth was terminated(26.2°C) was used as the threshold temperature to regionalize the oversummering range of wheat powdery mildew. This regionalization was done using the GIS ordinary kriging method combined with the Digital Elevation model(DEM) of China. The results showed that annual probability of oversummering region based on Model 26.2 were consistent with the actual survey of the more summer wheat powdery mildew. Wheat powdery mildew oversummering regions in China mainly cover mountainous or high-altitude areas, and these regions form a narrow north-south oversummering zone. Oversummering regions of wheat powdery mildew is mainly concentrated in the high-altitude wheat growing areas, including northern and southern Yunnan, northwestern Guizhou, northern and southern Sichuan, northern and southern Chongqing, eastern and southern Gansu, southeastern Ningxia, northern and southern Shaanxi, central Shanxi, western Hubei, western Henan, northern and western Hebei, western Liaoning, eastern Tibet, eastern Qinghai, western Xinjiang and other regions of China.

Re-delineating mountainous areas with three topographic parameters in Mainland Southeast Asia using ASTER global digital elevation model data

Tropical mountainous areas not only provide substantial carbon storage and play an important role in global biological diversity, but also provide basic livelihood for a large number of poor ethnic minorities. However, there is no unified and explicit definition for mountainous areas. The local elevation range(LER) is a crucial structural parameter for delineating mountainous areas. However, current LER products are limited by the subjective selection of an optimum statistical window or coarser spatial resolution of topographical data. In this study, we presented an approach using thresholds for three topographic parameters, elevation, slope, and LER, derived from the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model(ASTER GDEM) to redelineate the vast mountainous areas of mainland Southeast Asia(MSEA). The mean change-point analysis method was applied to determine the optimum statistical window of the 1 arc second(approximately 30 m)-resolution GDEM LER. The results showed that: First, the optimum statistical window is 38 × 38 cell units(width × height) in a rectangular neighborhood, or an area of about 1.30 km^2 for calculating GDEM LER in MSEA. Second, the LER of more than 80% of the area ranges from 30 m to 499 m in MSEA. The LERs in the northern and northwestern MSEA are greater than their counterparts in the south and east. Third, the area of the re-delineated mountainous areas was 83.52 × 10~4 km^2, about 38.71% of the total area. Spatially, the mountainous areas are mainly distributed in the north and northeast of MSEA. The re-delineated 30-m resolution map of the mountainous areas will serve as a topographical dataset for monitoring mountainrelated land surface changes in MSEA. The parameter-modified mountain extraction procedure can be expanded to delineate global mountainous areas.

Erosion processes in karst landscapes of the Russian plain northern taiga,based on digital elevation modeling

This paper considers the problems of the potential development of erosion processes in the natural landscapes of northern taiga in the Russian plain. It is considered that in forest ecosystems, erosion processes are slow and are weakly reflected in the terrain. However, the situation changes radically if the vegetation cover integrity is violated, which is inevitable with the modern methods of developing northern territories. Furthermore, global changes in average annual temperatures and the occurrence of karst processes may be the reason behind the development of erosion processes. The authors suggest a method for determining territories with a varying occurrence probability of erosional processes, based on digital elevation modelling. The territory of the Pinezhsky Nature Reserve(Arkhangelsk region) was chosen as the test plot. Direct field studies were previously used to detect exogenous geological processes in this territory. The authors were the first to suggest digital elevation modelling as a method that allows determining the potential danger of erosion in karst landscapes of the northern taiga. The geomorphometric studies resulted in the determination of areas with the greatest and lowest occurrence probability of erosion processes in the Pinezhsky Nature Reserve. It was established that the most significant erosion type was linear erosion, represented by incised river valleys and karst ravines. Sheet erosion is less significant and occurs as sinkholes, local declines, and chasms over the valleys of subterranean rivers.

The J_2 invariant relative configuration of spaceborne SAR interferometer for digital elevation measurement

A 3-craft formation configuration is proposed to perform the digital elevation model （DEM） for the distributed spacebome interferometric synthetic aperture radar （InSAR）, and it is optimized by the modified ant colony algorithm to have the best compatibility with J2 invariant orbits created by differential correction algorithm. The configuration has succeeded in assigning the across-track baseline to vary periodically and with its mean value equal to the optimal baseline determined by the relative height measurement accuracy. The required relationship between crafts＇ magnitudes and phases is formulated for the general case of interferometry measure from non-orthographic and non-lateral view. The J2 invariant configurations created by differential correction algorithm are employed to investigate their compatibility with the required configuration. The colony algorithm is applied to search the optimal configuration holding the near-constant across-track baseline under the J2 perturbation, and the absolute height measurement accuracy is preferable as expected.

The variation in basal channels and basal melt rates of Pine Island Ice Shelf

In recent years,there has been a significant acceleration in the thinning,calving and retreat of the Pine Island Ice Shelf(PIIS).The basal channels,results of enhanced basal melting,have the potential to significantly impact the stability of the PIIS.In this study,we used a variety of remote sensing data,including Landsat,REMA DEM,ICESat-1 and ICESat-2 satellite altimetry observations,and Ice Bridge airborne measurements,to study the spatiotemporal changes in the basal channels from 2003 to 2020 and basal melt rate from 2010 to 2017 of the PIIS under the Eulerian framework.We found that the basal channels are highly developed in the PIIS,with a total length exceeding 450 km.Most of the basal channels are ocean-sourced or groundingline-sourced basal channels,caused by the rapid melting under the ice shelf or near the groundingline.A raised seabed prevented warm water intrusion into the eastern branch of the PIIS,resulting in a lower basal melt rate in that area.In contrast,a deepsea trough facilitates warm seawater into the mainstream and the western branch of the PIIS,resulting in a higher basal melt rate in the main-stream,and the surface elevation changes above the basal channels of the mainstream and western branch are more significant.The El Ni?o event in 2015–2016 possibly slowed down the basal melting of the PIIS by modulating wind field,surface sea temperature and depth seawater temperature.Ocean and atmospheric changes were driven by El Ni?o,which can further explain and confirm the changes in the basal melting of the PIIS.

arcPycor:an open-source automated GIS tool to co-register elevation datasets

Subtraction of elevation datasets(e.g.digital elevation models(DEMs)and non-continuous elevation points)acquired at different times is a useful method to monitor landform surface change.Due to heavy post-processing of these elevation datasets,multi-source errors are introduced into the resulting elevation data products.To improve the estimation of elevation change,co-registration of elevation datasets is a prerequisite.This paper presents an open-source automated GIS tool(arc Pycor)for co-registering elevation datasets.arc Pycor is coded in Python 2.7 and is run via Arc GIS for Desktop.The performances of arc Pycor have been evaluated using a series of experiments.In benchmark tests,the resolved co-registration vectors of arc Pycor are compared to the predefined shift vectors obtained by artificially misaligning the slave DEMs from the master elevation datasets.Results show that arc Pycor is able to co-register DEMs with relative high accuracy and can well align slave DEMs to non-continuous elevation points,which indicates its robustness in co-registering of elevation datasets.arc Pycor is also able to co-register multi-sourced DEMs of different resolutions in mountain areas.

Morphodynamics and sediment connectivity index in an unmanaged,debris-flow prone catchment:a through time perspective

Torrential processes are among the main actors responsible for sediment production and mobility in mountain catchments.For this reason,the understanding of preferential pathways for sediment routing has become a priority in hazard assessment and mitigation.In this context,the sediment Connectivity Index(IC)enables to analyse the existing linkage between sediment sources and the selected target(channel network or catchment outlet).The IC is a grid-based index that allows fast computation of sediment connectivity based on landscape information derived from a single Digital Terrain Model(DTM).The index computation is based on the log-ratio between an upslope and a downslope component,including information about drainage area,slope,terrain roughness,and distance to the analysis target(e.g.outlet).The output is a map that highlights the degree of structural connectivity of sediment pathways over analysed catchments.Until now,these maps are however rarely used to help defining debris-flow hazard maps,notably due to a lack of guidelines to interpret the IC spatial distribution.This paper proposes an exploitation procedure along profiles to extract more information from the analysis of mapped IC values.The methodology relies on the analysis of the IC and its component variables along the main channel profile,integrated with information about sediment budgeting derived from Difference of DEMs(DoD).The study of connectivity was applied in the unmanaged sub-catchment(without torrent control works)of the Rio Soial(Autonomous Province of Trento–NE Italy)to understanding the geomorphic evolution of the area after five debris flows(in ten years)and the related changes of sediment connectivity.Using a recent DTM as validation,we demonstrated how an IC analysis over the older DTM can help predicting geomorphic changes and associated hazards.The results show an IC aptitude to capture geomorphic trajectories,anticipate debris flow deposits in a specific channel location,and depict preferential routing pathways.

The morphological changes of basal channels based on multisource remote sensing data at the Pine Island Ice Shelf

The basal channel is a detailed morphological feature of the ice shelf caused by uneven basal melting.This kind of specifically morphology is widely distributed in polar ice shelves.It is an important research object of sea-ice interaction and plays a vital role in studying the relationship between the ice sheet/ice shelf and global warming.In this paper,high-resolution remote sensing image and ice penetration data were combined to extract the basal channel of the Pine Island Ice Shelf.The depth variation of Pine Island Ice Shelf in the recent 20 years was analyzed and discussed by using ICESat-1,ICESat-2,and IceBridge data.Combined with relevant marine meteorological elements(sea surface temperature,surface melting days,circumpolar deep water and wind)to analyze the basal channel changes,the redistribution of ocean heat is considered to be the most important factor affecting the evolution and development of the basal channel.

Land Use Land Cover Dynamics of Oba Hills Forest Reserve, Nigeria, Employing Multispectral Imagery and GIS

Land use Land cover (LULC) has undergone progressive changes worldwide over the years. However, there is limited information available about these changes in Oba Hills Forest Reserve, Nigeria. The existing spatial analysis of the forest excluded important land use classes like settlements. Therefore, this study aimed at assessing the dynamics of LULC in Oba Hills Forest Reserve between 1987 and 2019. Images from Landsat 5, Landsat 7, and Landsat 8 for the years 1987, 2001, 2013, and 2019 were obtained and subjected to preprocessing and classification using the maximum likelihood algorithm, change detection, and Normalized Differential Vegetation Index (NDVI). The coordinates of specific benchmark locations and other points were acquired for ground-truthing and developing Digital Elevation Model (DEM). Three distinct LULC classes were identified: forest, bare land (including open spaces, agriculture, rocks, and grasslands), and built-up areas. The forest cover in the reserve gradually decreased from 56% in 1987 to 47% in 2019, resulting in a total area loss of 455.4 hectares. Correspondingly, the other LULC classes experienced exponential expansion. Bare land increased from 44% in 1987 to 52% in 2019, while the built-up area expanded by 57.28 hectares. These changes are attributed to prevalent anthropogenic activities such as agriculture, grazing, logging, firewood collection, and population growth within the catchment area. The declining NDVI values in the forest reserve, from 0.52 to 0.44 within the years of assessment, further substantiated the substantial loss of forest cover. The DEM and topographical map highlighted notable steep slopes and elevations of up to over 550 m above sea level (asl) within the reserve, which have implications for forest growth and dynamics. In conclusion, this study reveals extensive rates of forest cover changes into bare land, primarily for agriculture, and settlements, and offers further recommendations to reverse the trend.

Multi-scale Decomposition of Co-seismic Deformation from High Resolution DEMs:a Case Study of the 2004 Mid-Niigata Earthquake

Decomposing co-seismic deformation is an immediate need for researchers who are interested in earthquake inversion analysis and geo-hazard mapping. However, conventional InSAR or digital elevation models （DEMs） imagery analyses only provide the displacement in the Line-of-Sight （LOS） direction or elevation changes. The 2004 Mid-Niigata earthquake in Japan provides lessons on how to decompose co-seismic deformation from two sets of DEMs. If three adjacent points undergo a rigid-body-translation movement, their co-seismic deformation can be decomposed by solving simultaneous equations. Although this method has been successfully used to discuss tectonic deformations, the algorithm needed improvement and a more rigorous algorithm, including a new definition of nominal plane, DEMs comparability improvement and matrix condition check is provided. Even with these procedures, the obtained decomposed displacement often showed remarkable scatter prompting the use of the moving average method, which was used to determine both tectonic and localized displacement characteristics. A cut-off window and a pair of band-pass windows were selected according to the regional geology and construction activities to ease the tectonic and localized displacement calculations, respectively. The displacement field of the tectonic scale shows two major clusters of large lateral components, and coincidently major visible landslides were found mostly within them. The localized displacement helps to reveal hidden landslides in the target area. As far as the Kizawa hamlet is concerned, the obtained vectors show down-slope movements, which are consistent with the observed traces of dislocations that were found in the Kizawa tunnel and irrigation wells. The method proposed has great potential to be applied to understanding post-earthquake rehabilitation in other areas.

GIS tools for preliminary debris-flow assessment at regional scale

The assessment of the areas endangered by debris flows is a major issue in the context of mountain watershed management. Depending on the scale of analysis, different methods are required for the assessment of the areas exposed to debris flows.While 2-D numerical models are advised for detailed mapping of inundation areas on individual alluvial fans, preliminary recognition of hazard areas at the regional scale can be adequately performed by less data-demanding methods, which enable priority ranking of channels and alluvial fans at risk by debris flows. This contribution focuses on a simple and fast procedure that has been implemented for regionalscale identification of debris-flow prone channels and prioritization of the related alluvial fans. The methodology is based on the analysis of morphometric parameters derived from Digital Elevation Models(DEMs). Potential initiation sites of debris flows are identified as the DEM cells that exceed a threshold of slope-dependent contributing area. Channel reaches corresponding to debris flows propagation, deceleration and stopping conditions are derived from thresholds of local slope. An analysis of longitudinal profiles is used for the computation of the runout distance of debris flows. Information on erosion-resistant bedrock channels and sediment availability surveyed in the field are taken into account in the applications. A set of software tools was developed and made available(https://github.com/Hydrogeomorphology Tools) to facilitate the application of the procedure. This approach, which has been extensively validated by means of field checks, has been extensively applied in the eastern Italian Alps. This contribution discusses potential and limitations of the method in the frame of the management of small mountain watersheds.

The Importance of Digital Elevation Model Selection in Flood Simulation and a Proposed Method to Reduce DEM Errors:A Case Study in Shanghai

Digital Elevation Models(DEMs)play a critical role in hydrologic and hydraulic modeling.Flood inundation mapping is highly dependent on the accuracy of DEMs.Various vertical differences exist among open access DEMs as they use various observation satellites and algorithms.The problem is particularly acute in small,flat coastal cities.Thus,it is necessary to assess the differences of the input of DEMs in flood simulation and to reduce anomalous errors of DEMs.In this study,we first conducted urban flood simulation in the Huangpu River Basin in Shanghai by using the LISFLOOD-FP hydrodynamic model and six open-access DEMs(SRTM,MERIT,CoastalDEM,GDEM,NASADEM,and AW3D30),and analyzed the differences in the results of the flood inundation simulations.Then,we processed the DEMs by using two statistically based methods and compared the results with those using the original DEMs.The results show that:(1)the flood inundation mappings using the six original DEMs are significantly different under the same simulation conditions—this indicates that only using a single DEM dataset may lead to bias of flood mapping and is not adequate for high confidence analysis of exposure and flood management;and(2)the accuracy of a DEM corrected by the Dixon criterion for predicting inundation extent is improved,in addition to reducing errors in extreme water depths—this indicates that the corrected datasets have some performance improvement in the accuracy of flood simulation.A freely available,accurate,high-resolution DEM is needed to support robust flood mapping.Flood-related researchers,practitioners,and other stakeholders should pay attention to the uncertainty caused by DEM quality.

Hydrological daily rainfall-runoff simulation with BTOPMC model and comparison with Xin'anjiang model

A grid-based distributed hydrological model, the Block-wise use of TOPMODEL （BTOPMC）, which was developed from the original TOPMODEL, was used for hydrological daily rainfall-runoff simulation. In the BTOPMC model, the runoff is explicitly calculated on a cell-by-cell basis, and the Muskingum-Cunge flow concentration method is used. In order to test the model＇s applicability, the BTOPMC model and the Xin＇anjiang model were applied to the simulation of a humid watershed and a semi-humid to semi-arid watershed in China. The model parameters were optimized with the Shuffle Complex Evolution （SCE-UA） method. Results show that both models can effectively simulate the daily hydrograph in humid watersheds, but that the BTOPMC model performs poorly in semi-humid to semi-arid watersheds. The excess-infiltration mechanism should be incorporated into the BTOPMC model to broaden the model＇s applicability.

Calculation of watershed flow concentration based on the grid drop concept

The grid drop concept is introduced and used to develop a micromechanism-based methodology for calculating watershed flow concentration. The flow path and distance traveled by a grid drop to the outlet of the watershed are obtained using a digital elevation model （DEM）. Regarding the slope as an uneven carpet through which the grid drop passes, a formula for overland flow velocity differing from Manning＇s formula for stream flow as welt as Darcy＇s formula for pore flow is proposed. Compared with the commonly used unit hydrograph and isochronal methods, this new methodology has outstanding advantages in that it considers the influences of the slope velocity field and the heterogeneity of spatial distribution of rainfall on the flow concentration process, and includes only one parameter that needs to be calibrated. This method can also be effectively applied to the prediction of hydrologic processes in un-gauged basins.

HUBEX强化观测期雷达测雨在水文过程模拟中的应用(英)

On the basis of Digital Elevation Model data, the raster flow vectors, watershed delineation, and spatial topological relationship are generated by the Martz and Garbrecht method for the upper area of Huangnizhuang station in the Shihe Catchment with 805 km2 of area, an intensified observation field for the HUBEX/GAME Project. Then, the Xin’anjiang Model is applied for runoff production in each grid element where rain data measured by radar at Fuyang station is utilized as the input of the hydrological model. The elements are connected by flow vectors to the outlet of the drainage catchment where runoff is routed by the Muskingum method from each grid element to the outlet according to the length between each grid and the outlet. The Nash-Sutcliffe model efficiency coefficient is 92.41% from 31 May to 3 August 1998, and 85.64%, 86.62%, 92.57%, and 83.91%, respectively for the 1st, 2nd, 3rd, and 4th flood events during the whole computational period. As compared with the case where rain-gauge data are used in simulating the hourly hydrograph at Huangnizhuang station in the Shihe Catchment, the index of model efficiency improvement is positive, ranging from 27.56% to 69.39%. This justifies the claim that radar-measured data are superior to rain-gauge data as inputs to hydrological modeling. As a result, the grid-based hydrological model provides a good platform for runoff computation when radar-measured rain data with highly spatiotemporal resolution are taken as the input of the hydrological model.

Characteristics and influencing factors of proximity distance index on the northern Shaanxi Loess Plateau in China

Topographic feature points and lines are the framework of topography,and their spatial distance relationship is an breakthrough in the study of topographical geometry,internal structure and development level.Proximity distance(PD)is an indicator to describe the distance between the gully source point(GSP)and the watershed boundary.In the upstream catchment area,PDs can be expressed by the streamline proximity distance(SPD),as well as by the horizontal proximity distance(HPD)and the vertical proximity distance(VPD)in the horizontal and vertical dimensions,respectively.The series of indicators(e.g.,SPD,HPD and VPD)are important for quantifying the geomorphological development process of a loess basin because of the headward erosion of loess gullies.In this study,the digital elevation model data with 5 m resolution and a digital topographic analysis method are used for the statistical analyses of the SPD,VPD and HPD in 50 sample areas of 6 geomorphic types in the Loess Plateau of northern Shaanxi.The spatial characteristics and the influencing factors are also analysed.Results show that:1)Central tendencies for the HPDs and the VPDs for the whole study area and the six typical loess landforms are evident.2)Spatial patterns of the HPDs and the VPDs exhibit evident trends and zonal distributions over the whole study area.3)The HPDs have a strong positive correlation with gully density(GD)and hypsometric integral.The VPDs also correlates with GD to an extent.Vegetation cover,mean annual precipitation and loess thickness have stronger effects on the VPD than on the HPD.

Deriving the slope-mean shielded astronomical solar radiation spectrum and slope-mean possible sunshine duration spectrum over the Loess Plateau

Solar radiation is often shielded by terrain relief, especially in mountainous areas, before reaching the surface of the Earth. The objective of this paper is to study the spatial structures of the shielded astronomical solar radiation(SASR) and the possible sunshine duration(PSD) over the Loess Plateau. To this end, we chose six test areas representing different landforms over the Loess Plateau and the software package of Matlab was used as the main computing platform. In each test area, 5-m-resolution digital elevation model established from 1:10,000 scale topographic maps was used to compute the corresponding slope, SASR and PSD. Then, we defined the concepts of the slope-mean SASR spectrum and the slope-mean PSD spectrum, and proposed a method to extract them from the computed slope, SASR and PSD over rectangular analysis windows. Using this method, we found both spectrums in a year or in a season for each of the four seasons in the six test areas. Each spectrum was found only when the area of the corresponding rectangular analysis window was greater than the corresponding stable area of the spectrum. The values of the two spectrums decreased when the slope increased.Furthermore, the values of the stable areas of the spectrums in a year or in a season were positively correlated with the variable coefficients of the slope or the profile curvature. The values of the stable areas of the two spectrums in a year or in a season may represent the minimum value of test areas for corresponding future research on the spatial structures of the SASR or PSD. All the findings herein suggest that the spatial structures of the PSD and the SASR are caused by the interactions between solar radiation and terrain relief and that the method for extracting either spectrum is effective for detecting their spatial structures. This study may deepen our understanding of the spatial structure of solar radiation and help us further explore the distribution of solar energy in mountainous regions.

Distributed modeling of direct solar radiation on rugged terrain of the Yellow River Basin

Due to the influences of local topographical factors and terrain inter-shielding, calculation of direct solar radiation （DSR） quantity of rugged terrain is very complex. Based on digital elevation model （DEM） data and meteorological observations, a distributed model for calculating DSR over rugged terrain is developed. This model gives an all-sided consideration on factors influencing th a resolution of 1 km × 1 km for thDSR. Using the developed model, normals of annual DSR quantity wie Yellow River Basin was generated, with DEM data as the general characterization of terrain. Characteristics of DSR quantity influenced by geographic and topographic factors over rugged terrain were analyzed thoroughly. Results suggest that： influenced by local topographic factors, i.e. azimuth, slope and so on, and annual DSR quantity over mountainous area has a clear spatial difference; annual DSR quantity of sunny slope （or southern slope） of mountains is obviously larger than that of shady slope （or northern slope）. The calculated DSR quantity of the Yellow River Basin is provided in the same way as other kinds of spatial information and can be employed as basic geographic data for relevant studies as well.

Machine learning and geostatistical approaches for estimating aboveground biomass in Chinese subtropical forests

Background:Aboveground biomass(AGB)is a fundamental indicator of forest ecosystem productivity and health and hence plays an essential role in evaluating forest carbon reserves and supporting the development of targeted forest management plans.Methods:Here,we proposed a random forest/co-kriging framework that integrates the strengths of machine learning and geostatistical approaches to improve the mapping accuracies of AGB in northern Guangdong Province of China.We used Landsat time-series observations,Advanced Land Observing Satellite(ALOS)Phased Array L-band Synthetic Aperture Radar(PALSAR)data,and National Forest Inventory(NFI)plot measurements,to generate the forest AGB maps at three time points(1992,2002 and 2010)showing the spatio-temporal dynamics of AGB in the subtropical forests in Guangdong,China.Results:The proposed model was capable of mapping forest AGB using spectral,textural,topographical variables and the radar backscatter coefficients in an effective and reliable manner.The root mean square error of the plotlevel AGB validation was between 15.62 and 53.78 t∙ha^(−1),the mean absolute error ranged from 6.54 to 32.32 t∙ha^(−1),the bias ranged from−2.14 to 1.07 t∙ha^(−1),and the relative improvement over the random forest algorithm was between 3.8%and 17.7%.The largest coefficient of determination(0.81)and the smallest mean absolute error(6.54 t∙ha^(−1)were observed in the 1992 AGB map.The spectral saturation effect was minimized by adding the PALSAR data to the modeling variable set in 2010.By adding elevation as a covariable,the co-kriging outperformed the ordinary kriging method for the prediction of the AGB residuals,because co-kriging resulted in better interpolation results in the valleys and plains of the study area.Conclusions:Validation of the three AGB maps with an independent dataset indicated that the random forest/cokriging performed best for AGB prediction,followed by random forest coupled with ordinary kriging(random forest/ordinary kriging),and the random forest model.The proposed random forest/co-kriging framework provides an accurate and reliable method for AGB mapping in subtropical forest regions with complex topography.The resulting AGB maps are suitable for the targeted development of forest management actions to promote carbon sequestration and sustainable forest management in the context of climate change.