Search

Exogenous organic carbon (C) inputs and their subsequent microbial and mineral transformation affect the accumulation process of soil organic C (SOC) pool. Nevertheless, knowledge gaps exist on how different long-term forms of crop straw incorporation (direct straw return or pyrolyzed to biochar) modifies SOC composition and stabilization. This study investigated, in a 13-year long-term field experiment, the functional fractions and composition of SOC and the protection of organic C by iron (Fe) oxide minerals in soils amended with straw or biochar. Under the equal C input, SOC accumulation was enhanced with both direct straw return (by 43%) and biochar incorporation (by 85%) compared to non-amended conventional fertilization, but by different pathways. Biochar had greater efficiency in increasing SOC through stable exogenous C inputs and inhibition of soil respiration. Moreover, biochar-amended soils contained 5.0-fold greater SOCs in particulate organic matter (POM) and 1.2-fold more in mineral-associated organic matter (MAOM) relative to conventionally fertilized soils. Comparatively, although the magnitude of the effect was smaller, straw-derived OC was preserved preferentially the most in the MAOM. Straw incorporation increased the soil nutrient content and stimulated the microbial activity, resulting in greater increases in microbial necromass C accumulation in POM and MAOM (by 117% and 43%, respectively) compared to biochar (by 72% and 18%). Moreover, straw incorporation promoted poorly crystalline (Feo) and organically complexed (Fep) Fe oxides accumulation, and both were significantly and positively correlated with MAOM and SOC. The results address the decadal-scale effects of biochar and straw application on the formation of the stable organic C pool in soil, and understanding the causal mechanisms can allow field practices to maximize SOC content. These results are of great implications for better predicting and accurately controlling the response of SOC pools in agroecosystems to future changes and disturbances and for maintaining regional C balance. [ABSTRACT FROM AUTHOR]

Estimating forage yield and nutrient composition using hyperspectral remote sensing is a major challenge. However, there is still a lack of comprehensive research on the optimal wavelength for the analysis of various nutrients in pasture. In this research, conducted in Hailar District, Hulunber City, Inner Mongolia Autonomous Region, China, 126 sets of hyperspectral data were collected, covering a spectral range of 350 to 1800 nanometers. The primary objective was to identify key spectral bands for estimating forage dry matter yield (DMY), nitrogen content (NC), neutral detergent fiber (NDF), and acid detergent fiber (ADF) using principal component analysis (PCA), random forests (RF), and SHapley Additive exPlanations (SHAP) analysis methods, and then the RF and Extra-Trees algorithm (ERT) model was used to predict aboveground biomass (AGB) and nutrient parameters using the optimized spectral bands and vegetation indices. Our approach effectively minimizes redundancy in hyperspectral data by selectively employing crucial spectral bands, thus improving the accuracy of forage nutrient estimation. PCA identified the most variable bands at 400 nm, 520–550 nm, 670–720 nm, and 930–950 nm, reflecting their general spectral significance rather than a link to specific forage nutrients. Further analysis using RF feature importance pinpointed influential bands, predominantly within 930–940 nm and 700–730 nm. SHAP analysis confirmed critical bands for DMY (965 nm, 712 nm, and 1652 nm), NC (1390 nm and 713 nm), ADF (1390 nm and 715–725 nm), and NDF (400 nm, 983 nm, 1350 nm, and 1800 nm). The fitting accuracy for ADF estimated using RF was lower (R2 = 0.58), while the fitting accuracy for other indicators was higher (R2 ≥ 0.59). The performance and prediction accuracy of ERT (R2 = 0.63) were noticeably superior to those of RF. In conclusion, our method effectively identifies influential bands, optimizing forage yield and quality estimation. [ABSTRACT FROM AUTHOR]

Leaf area index (LAI) is a key indicator of vegetation structure and function, and its products have a wide range of applications in vegetation condition assessment and usually act as important input parameters for ecosystem modeling. Grassland plays an important role in regional climate change and the global carbon cycle and numerous studies have focused on the product-based analysis of grassland vegetation changes. However, the performance of various LAI products and their discrepancies across different grassland types in drylands remain unclear. Therefore, it is critical to assess these products prior to application. We evaluated the accuracy of four commonly used LAI products (GEOV2, GLASS, GLOBMAP, and MODIS) using LAI reference maps based on both bridging and cross-validation approaches. Under different grassland types, the GLASS LAI performed better in meadow steppe (R2 = 0.26, RMSE = 0.41 m2/m2) and typical steppe (R2 = 0.32, RMSE = 0.38 m2/m2); the GEOV2 LAI performed better in desert steppe (R2 = 0.39, RMSE = 0.30 m2/m2). When we assessed their spatial and temporal discrepancies during the period from 2010 to 2019, the four LAI products overall showed a high spatial and temporal consistency across the region. Compared with GLASS LAI, the most consistent to least consistent correlations can be ordered by GEOV2 LAI (R2 = 0.94), MODIS LAI (R2 = 0.92), and GLOBMAP LAI (R2 = 0.87). The largest differences in LAI throughout the year occurred in July for all grassland types. Limited by the location and number of sample plots, we mainly focused on spatial and temporal variations. The spatial heterogeneity of land surface is pervasive, especially in vast grassland areas with rich grassland types, and the results of this study can provide a basis for the application of the product in different grassland types. Furthermore, it is essential to develop highly accurate and reliable satellite-based LAI products focused on grassland from the regional to the global scale according to these popular approaches, which is the next step in our work plan. [ABSTRACT FROM AUTHOR]

Land use and land cover (LULC) change has greatly altered ecosystem carbon storage and exerted an enormous impact on terrestrial carbon cycling. Characterizing its impact on ecosystem carbon storage is critical to balance regional carbon budgets and make land use decisions. However, due to the availability of LULC data and the strong variability in LULC change, uncertainty remains high in quantifying the effect of LULC change on the historical and future carbon stock. Based on four historical LULC maps and one future LULC projection, this study combined the Land Use and Carbon Scenario Simulator (LUCAS) with a process-based CENTURY model to evaluate the historical and future LULC change and its impact on grassland carbon storage from 1991 to 2050 in northern China. Results showed that grassland experienced a drastic decrease of 16.10 × 103 km2 before 2005, while agriculture and barren land increased by 16.91 × 103 km2 and 3.73 × 103 km2, respectively. After that, grassland was projected to increase, agriculture kept steady, and barren land decreased. LULC change has resulted in enormous total ecosystem carbon loss, mainly in agro-pasture areas; the maximum 8.54% of carbon loss happened in 2000, which was primarily attributed to agriculture to grassland, forest to grassland, grassland to agriculture, and grassland to barren. Before 2000, the grassland net biome productivity was projected to be −15.54 Tg C/yr and −2.69 Tg C/yr with and without LULC change. After 2001, the LULC change showed a positive impact on the grassland carbon balance, and the region was projected to be a carbon sink. Ecological projects have made a significant contribution to grassland carbon storage. The paper provides a framework to account for the effects of LULC change on ecosystem carbon and highlights the importance of improving grassland management in balancing the grassland carbon budget, which is helpful to understand the regional carbon budget and better inform local land use strategies. [ABSTRACT FROM AUTHOR]

Leaf area index (LAI) is one of the key biophysical indicators for characterizing the growth and status of vegetation and is also used in modeling earth system processes. Machine learning algorithms (MLAs) such as random forest regression (RFR), artificial neural network regression (ANNR) and support vector regression (SVR) based on satellite data have been widely used for the estimation of LAI. However, the selection of input variables has a great impact on the estimation performance of MLAs. In this study, we aimed to improve the LAI inversion model of Inner Mongolia grassland based on MLAs incorporating empirical knowledge. Firstly, we used the ANNR, SVR and RFR approaches, respectively, to rank the input variables including vegetation indices, climate factors, soil factors and topography factors and found that Normalized Difference Phenology Index (NDPI) contributed the most to LAI estimation. Secondly, we selected four sets of input variables, namely, all variables—A, model selected variables—B, overlapping variables—C and self-defined variables—D, respectively. Subsequently, we built twelve LAI estimation models (RFR-A, RFR-B, RFR-C, etc.) based on three MLAs and four sets of input variables. The evaluation of them showed the RFR produced higher prediction accuracy, followed by ANNR and SVR. Furthermore, the RFR-D presented the highest accuracy in predicting LAI (R2 = 0.55, RMSE = 0.37 m2/m2, MAE = 0.29 m2/m2). Finally, we compared our results with MODIS LAI and GEOV2 LAI products and found that all of them showed a similar spatial distribution of grassland LAI in the four sub-regions covering all grassland types, but our model exhibited larger LAI values in the desert steppe and smaller LAI values in the others. These findings demonstrated that MLAs incorporating empirical knowledge could improve the accuracy of modelling LAI and further study is necessary to reduce the uncertainty in LAI mapping in grassland. [ABSTRACT FROM AUTHOR]

The rapid and accurate acquisition of rice growth variables using unmanned aerial system (UAS) is useful for assessing rice growth and variable fertilization in precision agriculture. In this study, rice plant height (PH), leaf area index (LAI), aboveground biomass (AGB), and nitrogen nutrient index (NNI) were obtained for different growth periods in field experiments with different nitrogen (N) treatments from 2019–2020. Known spectral indices derived from the visible and NIR images and key rice growth variables measured in the field at different growth periods were used to build a prediction model using the random forest (RF) algorithm. The results showed that the different N fertilizer applications resulted in significant differences in rice growth variables; the correlation coefficients of PH and LAI with visible-near infrared (V-NIR) images at different growth periods were larger than those with visible (V) images while the reverse was true for AGB and NNI. RF models for estimating key rice growth variables were established using V-NIR images and V images, and the results were validated with an R2 value greater than 0.8 for all growth stages. The accuracy of the RF model established from V images was slightly higher than that established from V-NIR images. The RF models were further tested using V images from 2019: R2 values of 0.75, 0.75, 0.72, and 0.68 and RMSE values of 11.68, 1.58, 3.74, and 0.13 were achieved for PH, LAI, AGB, and NNI, respectively, demonstrating that RGB UAS achieved the same performance as multispectral UAS for monitoring rice growth. [ABSTRACT FROM AUTHOR]

The leaf area index (LAI) is a key vegetation canopy structure parameter and is closely associated with vegetation photosynthesis, transpiration, and energy balance. Developing a landscape-scale LAI dataset with a high temporal resolution (daily) is essential for capturing rapidly changing vegetation structure at field scales and supporting regional biophysical modeling efforts. In this study, two daily 30 m LAI time series from 2014 to 2016 over a meadow steppe site in northern China were generated using a spatial and temporal adaptive reflectance fusion model (STARFM) combined with an LAI retrieval radiative transfer model (PROSAIL). Gap-filled Landsat 7, Landsat 8 and Sentinel-2A surface reflectance (SR) images were used to generate fine-resolution LAI maps with the PROSAIL look-up table method. Two daily 500 m moderate-resolution imaging spectroradiometer (MODIS) LAI product-the existing MCD15A3H LAI product and one was generated from the MCD43A4 SR product and the PROSAIL model, were used to provide temporally continuous LAI variations. The STARFM model was then used to fuse the fine-resolution LAI maps with the two 500 m LAI products separately to generate two daily 30 m LAI time series. Both results were assessed for three types of pasture (mowed pasture, grazing pasture, and fenced pasture) using ground measurements from 2014-2015. The results showed that the PROSAIL-generated LAI maps all exhibited a high accuracy, and the root mean squared errors (RMSEs) for the Landsat 7 LAI and Landsat 8 LAI compared to the ground-measured LAI were 0.33 and 0.28 respectively. The Landsat LAI maps also showed good agreement and similar spatial patterns with the Sentinel-2A LAI with mean differences between ± 0.5. The MCD43A4_PROSPECT LAI product exhibited similar seasonal variability to the ground measurements and to the Landsat and Sentinel-2A LAIs, and these data are also smoother and contain fewer noisy points than the gap-filled MCD15A3H LAI product. Compared to the ground measurements, the daily 30 m LAI time series fused from the fine-resolution LAI maps and PROSPECT generated MODIS LAI product demonstrated better performance with an RMSE of 0.44 and a mean absolute error (MAE) of 0.34, which is an improvement from the LAI time series fused from the fine-resolution LAI maps and the existing MCD15A3H LAI product (RMSE of 0.56 and MAE of 0.42). The latter dataset also exhibited abnormal temporal fluctuations, which may have been caused by the interpolation method. The results also demonstrated the very good performance of the STARFM model in grazing and mowed pasture with homogeneous surfaces compared to fenced pasture with smaller patch sizes. The Sentinel-2A data offers increased landscape vegetation observation frequency and provides temporal information about canopy changes that occur between Landsat overpass dates. The scheme developed in this study can be used as a reference for regional vegetation dynamic studies and can be applied to larger areas to improve grassland modeling efforts. [ABSTRACT FROM AUTHOR]

Leaf area index (LAI) is one of the most important parameters for determining grassland canopy conditions. LAI controls numerous biological and physical processes in grassland ecosystems. Remote-sensing techniques are effective for estimating grassland LAI at a regional scale. Comparison of LAI inversion methods based on remote sensing is significant for accurate estimation of LAI in particular areas. In this study, we developed and compared two inversion models to estimate the LAI of a temperate meadow steppe in Hulunbuir, Inner Mongolia, China, based on HJ-1 satellite data and field-measured LAI data. LAI was measured from early June to late August in 2013, obtained from 326 sampling data. The back propagation (BP) neural network method proved better than the statistical regression model for estimating grassland LAI, the accuracy of the former being 82.8%. We then explored the spatio-temporal distribution in LAI ofStipa baicalensisRoshev. in the meadow steppe of Hulunbuir, including cut, grazed, and fenced plots. The LAI in the cut and grazed plots reflected the growth variations inS. baicalensisRoshev. However, because of the obvious litter layer, the LAI in the fenced plots was underestimated. [ABSTRACT FROM PUBLISHER]

The 8.0-magnitude Wenchuan earthquake in 2008 damaged the ecology of northwestern Sichuan, China. This study assessed ecological changes within a few years of the earthquake through satellite observations of vegetation dynamics in the earthquake area. As an ecological indicator, the fractional vegetation cover was extracted using the Normalized Difference Vegetation Index based on multi-year Landsat images and was validated using airborne images. We found that the entire mountainous disaster area had recovered by 68.45 % 3 years after the 2008 earthquake. After rapid recovery of vegetation in 2009, the recovery process slowed. The vegetation recovery rate (VRR) in the area heavily damaged by landslides was slightly lower but nearly that of the entire disaster area. In addition, because of differences in the proportions of soil and rock in the damaged areas, recovery of vegetation in the southwest portion of the study area was slower than in the northeast areas. Topographic analysis of vegetation recovery patterns indicated that damage to vegetation was closely related to slope, while recovery of vegetation was more sensitive to elevation. The landscape analysis showed that the recovery rate (65.21 %) of the excellent vegetation cover type was slower than the overall VRR. This study suggests that vegetation recovery is a slow ecological process and that ecological restoration should be implemented in mountainous regions affected by the earthquake. [ABSTRACT FROM AUTHOR]

Rapid quantification of soil organic matter (SOM) is a great challenge for the health assessment and fertility management of agricultural soil. Laser-induced breakdown spectroscopy (LIBS) with appropriate modeling algorithms is an alternative tool for this measurement. However, the current calibration strategy limits the prediction performance of the LIBS technique. In this study, 563 soil samples from Hetao Irrigation District in China were collected; the LIBS spectra of the soils were recorded in the wavenumber range of 288–950 nm with a resolution of 0.116 nm; a self-adaptive partial least squares regression model (SAM–PLSR) was employed to explore optimal model parameters for SOM prediction; and calibration parameters including sample selection for the calibration database, sample numbers and sample location sites were optimized. The results showed that the sample capacity around 60–80, rather than all of the samples in the soil library database, was selected for calibration from a spectral similarity re-ordered database regarding unknown samples; the model produced excellent predictions, with R2 = 0.92, RPD = 3.53 and RMSEP = 1.03 g kg−1. Both the soil variances of the target property and the spectra similarity of the soil background were the key factors for the calibration model, and the small sample set led to poor predictions due to the low variances of the target property, while negative effects were observed for the large sample set due to strong interferences from the soil background. Therefore, the specific unknown sample depended strategy, i.e., self-adaptive modelling, could be applied for fast SOM sensing using LIBS for soils in varied scales with improved robustness and accuracy. [ABSTRACT FROM AUTHOR]

Short episodes of low-temperature stress during reproductive stages can cause significant crop yield losses, but our understanding of the dynamics of extreme cold events and their impact on rice growth and yield in the past and present climate remains limited. In this study, by analyzing historical climate, phenology and yield component data, the spatial and temporal variability of cold stress during the rice heading and flowering stages and its impact on rice growth and yield in China was characterized. The results showed that cold stress was unevenly distributed throughout the study region, with the most severe events observed in the Yunnan Plateau with altitudes higher than 1800 m. With the increasing temperature, a significant decreasing trend in cold stress was observed across most of the three ecoregions after the 1970s. However, the phenological-shift effects with the prolonged growing period during the heading and flowering stages have slowed down the cold stress decreasing trend and led to an underestimation of the magnitude of cold stress events. Meanwhile, cold stress during heading and flowering will still be a potential threat to rice production. The cold stress-induced yield loss is related to both the intensification of extreme cold stress and the contribution of related components to yield in the three regions. [ABSTRACT FROM AUTHOR]

Global sensitivity analysis (SA) has become an efficient way to identify the most influential parameters on model results. However, the effects of cultivar variation and specific-stage variations of climate conditions on model outputs still remain unclear. In this study, 30 indica hybrid rice cultivars were simulated in the CERES-Rice model; then the Sobol' method was used to perform a global SA on 16 investigated parameters for three model outputs (anthesis day, maturity day, and yield). In addition, we also compared the differences in the sensitivity results under four specific-stage variations (vegetative phase, panicle-formation phase, ripening phase, and the whole growth season) of climate conditions. The results indicated that (1) parameter Tavg, G4, and P2O are the most influential parameters for all model outputs across cultivars during the whole growth season; (2) under the vegetative-phase variation of climate parameters; the variability of model outputs is mainly controlled by parameter P2O and Tavg; (3) under the panicle-formation-phase or ripening-phase variation of climate parameters, parameter P2O was the dominant variable for all model outputs; (4) parameter PORM had a considerable effect (the total sensitivity index, S T i ; S T i > 0.05 ) on yield regardless of the various specific-stage variations of the climate parameters. Findings obtained from this study will contribute to understanding the comprehensive effects of crop parameters on model outputs under different cultivars and specific-stage variations of climate conditions. [ABSTRACT FROM AUTHOR]

The accurate estimation of grain yield in rice breeding is crucial for breeders to screen and select qualified cultivars. In this study, a low-cost unmanned aerial vehicle (UAV) platform mounted with an RGB camera was carried out to capture high-spatial resolution images of rice canopy in rice breeding. The random forest (RF) regression techniques were used to establish yield models by using (1) only color vegetation indices (VIs), (2) only phenological data, and (3) fusion of VIs and phenological data as inputs, respectively. Then, the performances of RF models were compared with the manual observation and CERES-Rice model. The results indicated that the RF model using VIs only performed poorly for estimating yield; the optimized RF model that combined the use of phenological data and color VIs performed much better, which demonstrated that the phenological data significantly improved the model performance. Furthermore, the yield estimation accuracy of 21 rice cultivars that were continuously planted over three years in the optimal RF model had no significant difference (p > 0.05) with that of the CERES-Rice model. These findings demonstrate that the RF model, by combining phenological data and color Vis, is a potential and cost-effective way to estimate yield in rice breeding. [ABSTRACT FROM AUTHOR]

Accurate and timely detection of phenology at plot scale in rice breeding trails is crucial for understanding the heterogeneity of varieties and guiding field management. Traditionally, remote sensing studies of phenology detection have heavily relied on the time-series vegetation index (VI) data. However, the methodology based on time-series VI data was often limited by the temporal resolution. In this study, three types of ensemble models including hard voting (majority voting), soft voting (weighted majority voting) and model stacking, were proposed to identify the principal phenological stages of rice based on unmanned aerial vehicle (UAV) RGB imagery. These ensemble models combined RGB-VIs, color space (e.g., RGB and HSV) and textures derived from UAV-RGB imagery, and five machine learning algorithms (random forest; k-nearest neighbors; Gaussian naïve Bayes; support vector machine and logistic regression) as base models to estimate phenological stages in rice breeding. The phenological estimation models were trained on the dataset of late-maturity cultivars and tested independently on the dataset of early-medium-maturity cultivars. The results indicated that all ensemble models outperform individual machine learning models in all datasets. The soft voting strategy provided the best performance for identifying phenology with the overall accuracy of 90% and 93%, and the mean F1-scores of 0.79 and 0.81, respectively, in calibration and validation datasets, which meant that the overall accuracy and mean F1-scores improved by 5% and 7%, respectively, in comparison with those of the best individual model (GNB), tested in this study. Therefore, the ensemble models demonstrated great potential in improving the accuracy of phenology detection in rice breeding. [ABSTRACT FROM AUTHOR]

Estimating plant nitrogen concentration (PNC) has been conducted using vegetation indices (VIs) from UAV-based imagery, but color features have been rarely considered as additional variables. In this study, the VIs and color moments (color feature) were calculated from UAV-based RGB images, then partial least square regression (PLSR) and random forest regression (RF) models were established to estimate PNC through fusing VIs and color moments. The results demonstrated that the fusion of VIs and color moments as inputs yielded higher accuracies of PNC estimation compared to VIs or color moments as input; the RF models based on the combination of VIs and color moments (R2 ranging from 0.69 to 0.91 and NRMSE ranging from 0.07 to 0.13) showed similar performances to the PLSR models (R2 ranging from 0.68 to 0.87 and NRMSE ranging from 0.10 to 0.29); Among the top five important variables in the RF models, there was at least one variable which belonged to the color moments in different datasets, indicating the significant contribution of color moments in improving PNC estimation accuracy. This revealed the great potential of combination of RGB-VIs and color moments for the estimation of rice PNC. [ABSTRACT FROM AUTHOR]

Rice yield is not only influenced by factors of varieties and managements, but also by environmental factors. In this study, agronomic trait data of rice and climate data in eastern China were collected, and rice yields were predicted using a variety of algorithms, including the non-linear tool of feed-forward backpropagation neural networks (FFBN) and the linear model of partial least squares regression (PLSR). The results showed that both the agronomic traits and the climate data were significantly related with rice yield. The PLSR model showed that covariates occurred among the parameters, and modifications should be considered for climate data-based modelling. The FFBN model demonstrated better prediction performance than that of PLSR, in which the relation coefficient (R2) and root mean square error (RMSE) were 0.611 vs. 0.374 and 0.578 vs. 0.865 ton/ha using climate data, respectively; and 0.742 vs. 0.689 and 0.556 vs. 0.608 using agronomic trait data, respectively. When using fused data the R2 and RMSE improved to 0.843 vs. 0.746 and 0.440 vs. 0.549, respectively. The optimum architecture of the FFBN consisted of one hidden layer with 29 neurons. Therefore, the FFBN algorithm is an effective option for the prediction of rice yield in complex systems of rice production. [ABSTRACT FROM AUTHOR]

Accurately estimating grassland carbon stocks is important in assessing grassland productivity and the global carbon balance. This study used the regression kriging (RK) method to estimate grassland carbon stocks in Northeast China based on Landsat8 operational land imager (OLI) images and five remote sensing variables. The normalized difference vegetation index (NDVI), the wide dynamic range vegetation index (WDRVI), the chlorophyll index (CI), Band6 and Band7 were used to build the RK models separately and to explore their capabilities for modeling spatial distributions of grassland carbon stocks. To explore the different model performances for typical grassland and meadow grassland, the models were validated separately using the typical steppe, meadow steppe or all-steppe ground measurements based on leave-one-out crossvalidation (LOOCV). When the results were validated against typical steppe samples, the Band6 model showed the best performance (coefficient of determination (R2) = 0.46, mean average error (MAE) = 8.47%, and root mean square error (RMSE) = 10.34 gC/m2) via the linear regression (LR) method, while for the RK method, the NDVI model showed the best performance (R2 = 0.63, MAE = 7.04 gC/m2, and RMSE = 8.51 gC/m2), which were much higher than the values of the best LR model. When the results were validated against the meadow steppe samples, the CI model achieved the best estimation accuracy, and the accuracy of the RK method (R2 = 0.72, MAE = 8.09 gC/m2, and RMSE = 9.89 gC/m2) was higher than that of the LR method (R2 = 0.70, MAE = 8.99 gC/m2, and RMSE = 10.69 gC/m2). Upon combining the results of the most accurate models of the typical steppe and meadow steppe, the RK method reaches the highest model accuracy of R2 = 0.69, MAE = 7.40 gC/m2, and RMSE = 9.01 gC/m2, while the LR method reaches the highest model accuracy of R2 = 0.53, MAE = 9.20 gC/m2, and RMSE = 11.10 gC/m2. The results showed an improved performance of the RK method compared to the LR method, and the improvement in the accuracy of the model is mainly attributed to the enhancement of the estimation accuracy of the typical steppe. In the study region, the carbon stocks showed an increasing trend from west to east, the total amount of grassland carbon stock was 79.77 × 104 Mg C, and the mean carbon stock density was 47.44 gC/m2. The density decreased in the order of temperate meadow steppe, lowland meadow steppe, temperate typical steppe, and sandy steppe. The methodology proposed in this study is particularly beneficial for carbon stock estimates at the regional scale, especially for countries such as China with many grassland types. [ABSTRACT FROM AUTHOR]