Simple steel-tiled houses (SSHs), as lightweight and economical building materials, have been widely utilized in the process of urban development. However, SSH inevitably causes environmental impacts, such as local heat islands. Acquiring spatial information on SSH is crucial for urban planning and understanding the anthropogenic influence on the environment. However, this prevalent artificial structure in urban areas has not received sufficient attention. SSH are often coated in various colors, making them particularly challenging to detect within complex urban backgrounds. Coated steel tile (CST) is an important feature of SSH. Based on Sentinel-2 imagery, this study constructs two intermediate variables (CSI and$\rho $max) to develop an advanced spectral index, termed CST index (CSTI), for detecting SSH in urban areas. The results demonstrate that CSTI outperforms the existing indices across all study areas, with an overall accuracy (OA) exceeding 0.95. Compared with supervised classification methods, the CSTI-based method achieves satisfactory mapping accuracy without the need for training samples. Moreover, CSTI is robust across different seasonal images and other sensors. Finally, in six Chinese cities (Harbin, Shenyang, Qinhuangdao, Zhengzhou, Jinan, and Linyi), SSH shows an aggregation phenomenon in urban centers, with an expansion trend in developing areas over the past decade. In conclusion, this study introduces an effective and robust method for SSH detection, expected to provide strong support for urban land management and ecological assessment.

Accurate delineation of agricultural field boundaries from remote sensing imagery is fundamental for precision agriculture and policy-making. However, existing deep learning methods struggle with this task due to the difficulty of balancing semantic context with fine detail. Their unique imaging features (e.g., intricate linear nature, frequent occlusions) further challenge the detection of continuous boundaries. To address these issues, we propose a High Connectivity and Accuracy Field Boundary Network (HCFNet). HCFNet employs a novel two-stage framework where a lightweight region segmentation task provides explicit semantic constraints to a dedicated boundary delineation task. The boundary task is powered by an enhanced edge network featuring four key innovations: a Linearly Connected Deformable Convolution (LCDConv) to capture fine boundary morphology, a Non-Local Operation (NLO) to resolve local ambiguities using global context, a Continuity Restoration Decoder (CRD) that expands the receptive field and bridge occluded segments, and a Hierarchical Attention Fusion Module (HAFM) for adaptive multi-scale feature fusion. We conducted extensive experiments on diverse datasets, including high-resolution Gaofen-1, Google Earth, UAV imagery, and the global FTW benchmark. HCFNet consistently outperformed state-of-the-art models, achieving superior boundary accuracy (F1 score) and significantly improved connectivity. Our results demonstrate that HCFNet provides a robust and efficient solution for field boundary delineation, with its specialized modules proving particularly valuable in data-scarce scenarios common in practical applications. The code for HCFNet is available at (https://github.com/BNU-zhu/HCFNet).

Accurate and timely information on the early spatial distribution of winter wheat is essential for crop growth monitoring and crop yield prediction. However, the primary challenge in large-scale remote sensing-based early winter wheat mapping lies in the irregular phenology caused by diverse growing environments. Our study aims to explore the suitability of samples in early winter wheat mapping across different geoclimatic zones, with the goal of deriving more accurate early winter wheat distribution maps. Specifically, we chose topography and climate as alternative geoclimatic factors. Using Sentinel-2 imagery as an example, we examined the differences in classification accuracy among samples influenced by geoclimatic zoning. The results show that (1) Training models based on geoclimatic zoning can reduce the omission error due to asynchronous phenology and improve the accuracy of large-scale winter wheat early mapping. (2) Among the different geoclimatic factors, the sample optimization based on both topographical and climate factor zoning had higher accuracy (2~8% higher in OA) and optimization of early identifiable time (about 2~10 days earlier). The training strategy of adjusting the applicable area of the samples according to the geoclimatic factors is of great significance for the accurate early mapping of winter wheat in large-scale.

Accurate cultivated land parcels’ (CLPs) information is essential for precision agriculture. Deep learning methods have shown great potential in CLPs’ delineation but face challenges in detection accuracy, generalization capability, and parcel optimization quality. This study addresses these challenges by developing a high-generalization multitask detection network coupled with a specialized parcel optimization step. Our detection network integrates boundary and region tasks and designs distinct decoders for each task, employing performance-enhancing modules as well as more balanced training strategies to achieve both accurate semantic recognition and fine-grained boundary depiction. To improve the network’s ability to train more generalized models, our study identifies the variations in image hue, landscape surroundings, and boundary granularity as the key factors contributing to generalization degradation and employs color space augmentation (CSA) and attention mechanisms on spatial and hierarchy to enhance the generalization. In addition, the parcel optimization step repairs long-distance boundary breaks and performs object-level fusion of delineated regions and boundaries, resulting in more independent and regular CLP results. Our method was trained and validated on GaoFen-1 images from four diverse regions in China, demonstrating high delineation accuracy. It also maintained stable spatiotemporal generalization across different times and regions. Comprehensive ablation and comparative experiments confirmed the rationale behind our model improvements and demonstrated our method’s effectiveness over existing single-task models (SegNet, modified PSPNet (MPSPNet), DeeplabV3+, U-Net, ResU-Net, and R2U-Net) and recent multitask models (ResUNet-a, BSiNet, and SEANet). The implementation of our method is available athttps://github.com/BNU-zhu/CLPs-delineation.

Timely and accurate spatial distribution information of agricultural plastic greenhouse (APGs) is an important basis for agricultural production. In this study, we used Gao-Fen 1 (GF-1) images as data sources to analyze the applicability of spectral characteristics and related indices and texture extraction algorithms under different seasons for the identification of plastic greenhouses in view of the unique spatial distribution details of plastic greenhouses. The results showed that: 1) the accuracies of APGs in different seasons was close, indicating that the method using texture features can be independent of season; 2) among the texture features obtained in different ways, the mean value of gray level co-occurrence matrix (GLCM) had the best recognition effect on APGs. And the overall accuracy of APGs in different seasons obtained based on the optimal features was above 80%, and the highest OA reached 87.01% in site C. This method is of great significance for the feature selection and accurate mapping of the spatial distribution of APGs.

Fine-scale population distribution data play an essential role in numerous fields, In this paper, we have developed an appropriate method that using a random forest algorithm to down-scale the town-level population distribution to the grid level. the method was supported by remote sensing data and social sensing data to obtain a population distribution map with a spatial resolution of$5\mathrm{m}\times 5\mathrm{m}$. Compared with WorldPop$(100\mathrm{m}\times 100\mathrm{m})$. our method has well accuracy (PearsonR = 0.7985, RMSE = 613.5037, p <0.0001). This method has reference significance for the micro-mapping of urban population distribution.

An algorithm uses the Normalized Difference Vegetation Index (NDVI) time series curve to compute NDVI change rate (CR) for every 8-day over a period (2008-2018) from Moderate-resolution Imaging Spectroradiometer (MODIS) data. The indices of CR variables within the winter wheat growing season were correlated with the end of the season winter wheat yield. A strong correlation (Pearson correlation coefficient = -0.48) was found in day of year 153-161 (filling) and the CR_153-161 was used to build a univariate regression model. Stepwise multiple linear regression was then used and then 16 CR variables were selected to construct the multiple regression model. The two models are good at predicting yield. The prediction results of two models were compared with official agricultural statistics showing that the RMSE = 578.21 kg ha-1/457.38 kg ha-1and MRE=18.54%/14.56% Remote sensing of NDVI-CR, therefore, is a valuable tool for estimating winter wheat yield well in advance of harvest.

Here we present a new index, observed snow probability (OSP), to explore the spatiotemporal characteristics of snow cover. We determine the distribution of OSP values across the Tibetan Plateau (TP) by analyzing Moderate Resolution Imaging Spectroradiometer (MODIS) data from the 2002-2016 time period to investigate the relationship between OSP and two terrain factors, elevation and aspect. Our results suggest that the distribution of OSP across the TP exhibits a large spatiotemporal heterogeneity with high OSPs mainly concentrated along the southern and western edges of the TP, which is strongly linked to the supply of moist air. The distribution of snow cover is also heavily dependent on elevation, with higher OSPs at higher elevations. Aspect is another key factor that significantly influences OSP, with higher OSPs corresponding to the shady aspects and windward slopes.

In China, the traditional method of determining actual grain yield is expensive, labor-intensive and inefficient. Using satellite imaging data, this work designed a remote-sensing-aided grain yield sampling and measuring method by studying the corn produced in Shunyi, Beijing, in 2015, based on counting plants and kernels. The aim of the research is to transform the traditional practice of field surveying, harvesting and weighing, into counting plants and kernels, culminating in building a regression model based on the relationship of the calculated sample yield and the corresponding normalized vegetation difference index (NDVI) for realizing the spatial distribution of the crop yield. The results show that the yield monitoring results based on counting plants and kernels and the validation error of NDVI regression model are all within a reasonable range. The scheme designed in this research is practically feasible, because it is simple and easy to operate; therefore, it is worthy of promotion and further work.

Cropland is one of highly variable land cover on earth surface, which is greatly significant for grain production and food security. In this work we developed a strategy of temporal transfer with fully convolutional neural networks for cropland mapping using high resolution remote sensing imagery. The results showed cropland mapping with historical label dataset performed well with 90.05% overall accuracy, while the accuracy slightly increased by 1.15% after fine-tuning with a small new dataset. This means it is a practical proposal to monitor cropland efficiently and accurately using historical sample dataset.

Monitoring crop areas is a key issue in remote sensing studies. A Crop Proportion Phenology Index (CPPI) model has previously been developed for estimation of winter wheat areas. Here we test the CPPI model in different areas using remote sensing data for varied kernel functions, including linear regression (LR), Artificial Neural Network (ANN), and Support Vector Regression (SVR). The differences of the model performances among different kernel functions were found to be small for areas with simple planting structure. For areas where multiple crop types have similar phenology cycles, the non-linear model of ANN was found to perform the best. This study indicates that the CPPI model can be applied to map winter wheat distribution in areas with complex planting structures, thus it holds promises for estimating fractional areas of winter wheat areas over large geographic areas.

The traditional flood risk analysis methods based on probability and statistics have limitations in terms of availability of samples. Information diffusion is a risk analysis method for fuzzy analysis of data, which is used to make up for the lack of sample information. In this work, we established a flood disaster index, and assessed the flood risk situation in 18 cities in Henan province, China. The results showed that Zhumadian city has the highest probability of flood risk among all types of risk degrees, and some other cities, such as Jiaozuo, face high risk in case of mild floods, but low risk in case of severe floods. Generally, the northwestern areas of Henan face lower risk than the southeastern areas. Our research would provide useful decision-making information for the disaster and emergency management departments. It will also prove useful for the insurance companies in determining the agricultural disaster insurance rates.

Flood is the most frequent disaster in the world, which can do harm to agriculture and threat to food security. Using kernel based supervised classifier to execute change detection for multi-temporal remote sensing data is a common method for flood disaster monitoring and assessment, and kernel Fisher's discrimination analysis (KFDA) is one of them. Choosing training sample by visual interpretation is an important step, but difficult and wasting time, for the reason that a great amount of the flooded pixels are heterogeneous. In this study, we proposed an automatic sample extraction method, finding pixels in relative homogeneous areas by multiresolution segmentation and zonal standard deviation calculating, and then assigning sample class via clustering or linear discrimination of some specific index. The results showed that overall accuracy could reach 91.57% and the Kappa coefficient was 0.8316. The method we proposed was proved to be efficient.

An accurate estimation method of a population's spatial distribution is useful and necessary for disaster risk analysis. In this study, we proposed a synthesis method using NPP/VIIRS data and land-cover data and use it to estimate the population density in the Henan Province of China, taking into account the differences between urban and rural areas. The results showed that the population density is from 0–16047 people/0.25 km2; high values are mainly distributed in the big cities, while low values are distributed in the rural areas, agreeing with the actual distribution in the study area. The mean estimation error is 26.27% in our study, which is lower than that of previous studies (44.98% and 34.67%), showing that our method has advantages for regional or small-scale risk analysis.

To take advantage of conventional hard land use/cover change detection method (HLUCD) and soft land use/cover change detection method (SLUCD), we develop a soft and hard land use/cover change detection method (SHLUCD) for crop distribution mapping. Two HJ-1/CCD images, acquired on 6 October 2011 (T1) and 16 April 2012 (T2) which represented the period of sowing and jointing respectively, were utilized by SHLUCD to extract wheat area in study area. The results show that the crop distribution derived from the SHLUCD reflects reality more accurately than that from HLUCD and SLUCD. Crops distribution mapping derived from SHLUCD give lowest RMSE and bias and the highest R2than that from other two methods in all window size. Wheat distribution in typical area and mixed pixels zone could be identified by land use/cover change status and land change scope respectively through SHLUCD. Moreover, the theory and methods employed in developing the SHLUCD provide a new way for crop distribution mapping based on change detection technique.

Accurate and timely information regarding the location and area of major crop types has significant economic, food, policy, and environmental implications. Both hard and soft classification methods are used throughout the growing season to generate cropland distribution maps using multiple remotely sensed data. Hard classification models (HCMs) yield good results in large homogeneous areas where pure pixels are dominant, but they fail in fragmented areas where mixed pixels are dominant. Conversely, soft classification models (SCMs) are thought to have greater accuracy in fragmented areas than in regions with pure pixels. To take advantage of both methods, we develop a hard and SCM (HSCM) based on existing HCMs and SCMs, and test it using data from simulated images as well as actual satellite data from southeast Beijing, China. The model assessment was performed using three statistical metrics at scales ranging from 1$\times$1 to 10$\times$10 pixels. The results reveal that the HSCM has the highest classification accuracy and produces more reasonable cropland distribution maps than those produced by either HCMs or SCMs. Moreover, the theory and methods employed in developing the HSCM provide a unifying framework for mapping land cover types, and they can be applied to different HCMs and SCMs beyond those currently in use.

Time-series data of normalized difference vegetation index (NDVI), derived from satellite sensors, can be used to support land-cover change detection and phenological interpretations, but further analysis and applications are hindered by residual noise in the data. As an alternative to a number of existing algorithms developed to compensate for such noise, we develop a simple but computationally efficient method (which we call the changing-weight filter method) to reconstruct a high-quality NDVI time series. The new algorithm consists of two major procedures: (1) detecting the local maximum/minimum points in a growth cycle along an NDVI temporal profile based on a mathematical morphology algorithm and a rule-based decision process and (2) filtering an NDVI time series with a three-point changing-weight filter. This method is tested at 470 test points for 55 vegetation types and a test region in China using a 250-m 16-day Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI product. Comparing our results to those of three other well-known methods-asymmetric Gaussian function fitting, double logistic function fitting, and Savitzky-Golay filtering-the new method has many of the advantages of existing methods, while in some cases, the changing-weight filter method more effectively preserves the curve shape as well as the timing and the amplitude of the local maxima/minima in the NDVI time series for a broad range of phenologies. Moreover, the response of the filtering algorithm is relatively insensitive to the exact values of its design parameters, making the new method more flexible and effective in adjusting to fit a variety of classes of NDVI time series.

With the significantly improved data availability in remote sensing technology, mid-resolution images have become the primary data source for crop sown area measurement in large scale. However, it is still difficult to solve the problems of spectrum heterogeneity in one field and spectra similarity between fields. This paper developed mixed field decomposition method and tested the method in an urban agriculture region with complex plant structure through several steps: first, distinguishing the mixed parcels by calculating the coefficient of variation of multi-temporal TM image within the parcels; then, operating multivariate regression model and mixed field decomposition model based on support vector machine (SVM) to estimate the sown area of winter wheat in the mixed parcels with different sample size. Results show that the mixed field decomposition of SVM has a higher accuracy than the multivariate regression model both in amount and position.

The objective of this study is to measure the area of the winter wheat based on the multi-temporal middle resolution remote sensing images and the database of farmland parcel, and then compare the classification accuracy under the different amount of information. The study firstly analyzed the different crop samples in the support of field data, and then classified the images by the three different ways of NDVI threshold segmentation. The classification accuracies were compared in the three conditions of information amount: single-temporal image in November 19 in 2007, three-temporal image and three-temporal image with database of farmland parcel in Beijing in 2006. The results indicate that single-temporal image reveals the low classification accuracy. The multi-temporal images efficiently distinguish wheat from others plants. Moreover, the accuracy could be further improved if the database of farmland parcel was used to rule out the non-cultivated land.

Traditional methods always use NDVI to extract rice from remote sensing images, for example TM. However, little research focusses on introducing a moisture index into rice extraction. This paper introduced NDII and NDVI respectively to extract rice based on a TM image of a rice and ginkgo mixed area, and compared their ability in distinguishing rice and ginkgo from two aspects: position accuracy and regional accuracy. Research results show that: (1) NDII has the higher ability to distinguish rice and ginkgo; (2) rice extraction results from NDII are both better than results from NDVI, whether position accuracy or whole accuracy.

Reliable in-situ sample data is essential to run the remotely sensed (RS) model and to check the RS result. Existing in-situ data collection methods are ruler measurement by manual work or tracking the parcel by high-accuracy Global Position System (GPS). The former is time-consuming and lacks spatial information, while the latter is costly and the GPS signal is not stable. On account of this, we have developed a new approach named Line-track to get reliable in-situ sample data in the paper published in the 2006 IGARSS Conference. This paper simulates plenty of in-situ samples from the QuickBird data. The percentage of wheat in every sample varies from 0 to 100%. This paper sets up an accuracy assessment model of Line-track, which analyzes the variation of the accuracy of Line-track under different conditions, such as the actual percentage of wheat in a sample, and the distance of two sample lines. The analysis of Line-track demonstrates that the Line-track is not suitable for the wheat sample of low percentage (from 0 to 20%). The average accuracy of the samples, the percentage of which is from 20% to 50%, is more than 97% when the distance of two sample lines is less than eight meters. The average accuracy of the sample (from 50% to 70%) is more than 97%, when the distance of two sample lines is less than sixteen meters, the average accuracy of the sample of high percentage (from 70% to 100%) is about 97%, and the standard deviation is less than 2%, when the distance of two sample lines is less than twenty-five meters. Such results suggest that Line-track is reliable to obtain accurate in-situ data due to less dependence on time and cost. The development of the Line- track will strongly support the collection of in-situ sample in the application of RS technology.

Modeling of population density based on GIS and RS has a promising prospect in China. In this article, an improved model of population density based on geographical information system (GIS) and remote sensing (RS) technology was used to simulate the population density distribution of four counties of Shandong Province, China. Results showed that: population density of the study area is in the range of 0 - 19214.76 persons/km2. As a whole, population density in urban area is higher than that in rural area.

This paper has put forward an innovated CA model to sub-pixel mapping. It has the advantage of easily constructing the rules of CA and the potential for controlling the spatial scale in sub-pixel mapping freely. The neural networks have played an important role in the whole work.

In order to replenish the current few researches on degraded grassland management, this paper develops an ecological management index (EMI) of degraded grassland, which integrates the information of degradation degree, risk degree and easily-restoration degree by spatial analysis of remote sensing (RS) and geography information system (GIS), in the context of ecological risk assessment. Specially, it estimates EMI of degraded grassland at Xilin River Basin, Inner Mongolia, China and proposes relative measures to restore the degraded grasslands at different EMI level. The case study demonstrates that the EMI can synthesize the degradation level, hazard and the restoration benefit of degraded grassland, which is significant for the optimum allocation of resources during the restoration of degraded grassland. In addition, the EMI is universal to the assessment of degraded grassland in most regions in that it can integrate most concerned information.

Reliable in-situ sample data is essential to run the remotely sensed (RS) model and to check the RS result. Existing in-situ data collection methods are ruler measurement by manual work or tracking the parcel by high-accuracy Global Position System (GPS). The former is time-consuming and lacks spatial information, while the latter is costly and the GPS signal is not stable. On account of this, the paper develops a new approach named line-track to get reliable in-situ sample data. The main idea can be described as follows: first, we locate the sample by high-accuracy GPS; then we lay sample lines at certain intervals both horizontally and vertically and obtain the percentage of certain objective on each sample line. In the end we calculate the percentage of objective in the sample. This idea is similar to the line transect in ecology, which is used to estimate wildlife densities. By computer modeling, this paper sets up an accuracy assessing model of line-track, and analyzes the variation of the accuracy of line-track under different conditions, such as distribution of objective, the actual percentage of objective in a sample, and the distance of two sample lines. The modeling of line-track demonstrates that the total accuracy of covering of certain objective at the middle percentage, that is from 15% to 85%, is about 97% when the distance of two sample lines is less than twelve meters, and it would be about 95% at high or low percentage when the distance is less than twenty meters. Such results suggest that line-track is reliable to obtain accurate in-situ data due to less dependence on time and cost. The development of the line-track will strongly support the collection of in-situ sample in the application of RS technology.

In this paper, a new object-based change detection approach is developed. The approach consists of three steps: (1) producing multi-scale objects from multi-temporal remote sensing images by combining the spectrum, texture and context information; (2) extracting potential change object by the comparison of the attributes of shape, structure, texture, etc. of each object; (3) determining the changed object and detecting urban expansion area with the help of in-situ investigation. When the object-based approach was applied to the urban expansion detection in Haidian District, Beijing, China with the support of two Landsat Thematic Mapper (TM) data in 1997 and 2004, the satisfactory results were obtained. The overall accuracy is about 80.3%, Kappa about 0.607, which are more accurate than postclassification change detection. The newly developed objectbased change detection approach possesses the advantage of its reduction to error accumulation of image classification of individual date and its independence to the radiometric correction to some extent. Keywordschange detection; object-orient; similarity; remote sensing; texture; land use/cover

High spatial remote sensing images have a promising prospect in land use and cover change study. In this article, we combined spectral, textural and structure information to classify IKONOS image by using SVM. Results indicated the method proposed by us could solve fragmented problem brought by single source data classification and attain higher accuracy

Maximum light use efficiency (emax) is a key parameter for the estimation of net primary productivity (NPP) driven with remote sensing data. There are still many divergences about its value for each vegetation type. In this paper, the emax for some typical vegetation types in China was simulated using a modified least squares function based on NOAA/AVHRR remote sensing data and field observed NPP data. The vegetation classification accuracy was also introduced to the process. The results showed that the simulated values of emax in this study were greater than the value used in CASA model, and less than the values simulated with BIOME-BGC model. This was consistent with some other studies and indicated that the simulated values of emax were reliable and stable.

Grassland deterioration due to climate variability and human disturbance in arid and semi-arid areas is becoming a serious environmental problem in China. Establishing grassland protection area to control the overgrazing is regarded as one of the effective measure to protect grassland. The paper presented a new method that integrates cellular automata (CA) model, geographical information system and remote sensing to zone grassland protection area with the case study in Xilingol "typical steppe" grassland in Inner Mongolia autonomous regions of China. The basic idea of the method is to extract "seed points" of the grassland protection area by using remote sensing techniques and geographical information system (GIS) at first, then simulate the grassland protection area by CA model. Since the method tries to satisfy the zoning requirement of the grassland protection area and utilizes the advantages of remote sensing techniques and CA models, satisfactory results have been produced for governmental officials and planners with manpower saved

Modeling land use scenario changes and its potential impact on the ecosystem structure and functioning in typical region are helpful to understand the reciprocal mechanism between land use system and ecosystem. A land use scenario dynamics model (LUSD) by the integration of system dynamics (SD) model and cellular automata (CA) model is developed with land use scenario changes in China in next 50 years simulated in this paper. The basic idea of LUSD is to model the land use scenario demand by SD model at the national/regional scales at first, then allocate the land use pattern at the local scale with the consideration of land use suitability, inheritance and neighborhood effect by CA model to satisfy the balance of land use demand and supply. The application of LUSD in China suggests that the model have the ability to reflect the complex behaviors of land use system at different scales to some extent and be a useful tool to assess the potential impact of land use system on ecosystem.

Statistical interpolation of the temperature for the missing points is one of the most popular approaches for generating high spatial resolution data sets. However, many interpolation methods used by previous studies are purely mathematic ways, without geographical significance being considered. In the present study the authors interpolate the monthly and annual mean temperature using 726-station observations in China, utilizing improved methods by taking into account geographical factors such as latitude, longitude, altitude. In addition, a smart distance-searching technique is adopted, which helps select the optimum stations on which the guess values at missing points are generated. Results show that the methods used here have evident advantages over the previous approaches. The mean absolute error of ordinary inverse-distance-squared (IDS) technique is in the range of 1.44-1.63degC, on average 1.52degC. The improved method yields a mean of 0.53-0.92degC, on average 0.69degC. Errors have been reduced as much as 50%

Some vegetation primary production models have been developed in recent years as research issues related to food security and biotic response to climate warming have become more compelling. An estimation model of net primary productivity (NPP), based on geographic information system (GIS) and remote sensing (RS) technology, is presented. The model, driven with ground meteorological data and remote sensing data, moves beyond simple correlative models to a more mechanistic basis and avoids the need for a full suite of eco-physiological process algorithms that require explicit parameterization. Therefore, it is relatively easier to acquire data. Application and validation of this model in Inner Mongolia, China, was conducted. After the validation with observed data and the comparison with other NPP models, the results showed that the predicted NPP was in good agreement with field measurement, and the remote sensing method can more actually reflect the forest NPP than Chikugo model. These results illustrated the utility of the model for terrestrial primary production over regional scales