Search

Over the last decades, growing crop production across China has had far-reaching consequences for both the environment and human welfare. One of the emerging questions is "how to meet the growing food demand in China?" In essence, the consensus is that the best way forward would be to increase crop yield rather than further extend the current cropland area. However, assessing progress in crop production is challenging as it is driven by multiple factors. To date, there are no studies to determine how multiple factors affect the crop production increase, considering both intensive farming (using yield and multiple cropping index) and large-scale farming (using mean parcel size and number of parcels). Using the Logarithmic-Mean-Divisia-Index (LMDI) decomposition method combined with statistical data and land cover data (GlobeLand30), we assess the contribution of intensive farming and large-scale farming changes to crop production dynamics at the national and county scale. Despite a negative contribution from MPS (mean parcel size, −25%), national crop production increased due to positive contributions from yield (+77%), MCI (multiple cropping index, +27%), as well as NP (number of parcels, +21%). This allowed China to meet the growing national crop demand. We further find that large differences across regions persist over time. For most counties, the increase in crop production is a consequence of improved yields. However, in the North China Plain, NP is another important factor leading to crop production improvement. On the other hand, regions witnessing a decrease in crop production (e.g., the southeast coastal area of China) were characterized by a remarkable decrease in yield and MCI. Our detailed analyses of crop production provide accurate estimates and therefore can guide policymakers in addressing food security issues. Specifically, besides stabilizing yield and maintaining the total NP, it would be advantageous for crop production to increase the mean parcel size and MCI through land consolidation and financial assistance for land transfer and advanced agricultural infrastructure. [ABSTRACT FROM AUTHOR]

The scarcity of water resources is one of the biggest problems restricting food production. Crop water productivity (CWP) is a practical and quantifiable indicator for characterizing agricultural water use efficiency. Remote sensing technology provides an accurate, cost-effective, and regional method for estimating CWP. However, remote sensing methods for CWP estimation and their application scenarios need to be summarized. In this paper, the CWP and related parameters are clearly defined. Different types of CWP estimation methods and their application at different scales are reviewed. CWP, as the crop yield ratio to actual crop evapotranspiration (ETa), is typically not directly estimated but calculated by estimating crop yield and ETa. Therefore, crop yield and ETa estimation methods are summarized, respectively. ETa can be remotely sensed using surface energy balance residual methods, semi-empirical formula methods, statistical regression methods, and ground instruments. Crop yield can be remotely estimated using data assimilation, statistical regression, and ground instruments. Moreover, the application of these methods at the point, field, and regional scales is further reviewed from previous literature. Finally, the in-situ measurements of CWP are introduced. This review can provide a detailed reference for follow-up studies related to CWP. [ABSTRACT FROM AUTHOR]

The pairwise separability index (SI) has been demonstrated as an effective indicator for capturing crucial phenological differences between two plant species. However, its application to crop types, which have more obvious phenological characteristics than natural vegetation, has received less attention, and extending the pairwise SI to multiple crops for feature selection still remains a challenge. This paper presented two SI extension approaches (\textSI\mathrm{ave} and \textSI\min) to select the optimal spectro-temporal features for multiple crops, and investigated their classification performance using Heilongjiang Province, China, as a study area. Feature interpretability and classification accuracy of different crops were evaluated for the two approaches. The results showed that the \textSI\mathrm{ave} approach generally has relatively high feature interpretability due to its better description of crucial phenological characteristics of different crops. Those crops with high separability are insensitive to the extension approach and have similar classification accuracy for the two approaches, whereas those crops with poor separability show good performance with the \textSI\min method. Due to the higher temporal autocorrelation, the optimal features for crop classification that are selected by the \textSI\mathrm{ave} approach exhibit greater information redundancy across the time domain than those that are selected by the \textSI\min approach, which largely explains the relatively low classification accuracy achieved using the \textSI\mathrm{ave} approach. These comparison results between \textSI\min and \textSI\mathrm{ave} approaches also indicate that time-series images with high temporal resolution do not necessarily produce high classification accuracy, regardless of their ability to describe the seasonal characteristics of crops. [ABSTRACT FROM PUBLISHER]

Crop pattern is a key element in agricultural land systems other than land use and land cover. Crop pattern dynamic changes take place very frequently, but they are not always easily observable, making many difficulties for analysis. As an effective tool for understanding the driver, process and consequence of agricultural land system changes, the spatially-explicit agent-based land change models have successfully been applied in representing human and natural interactions on agricultural landscapes. With the assumption that the crop pattern at a regional level is the aggregation of crop choices at the filed level, we conceptualized an agent-based model to simulate crop pattern dynamics at a regional scale (CroPaDy), which was supposed to represent the frequent but uneasily observed crop pattern changes in agricultural land systems. The conceptualization of CroPaDy model was designed strictly following the standard protocol for agent-based modeling. However, the computational model hinders its application because it needs a grid-based representation and the model itself is complicated with multi objectives, and nested by 3 interactive sub modules. As CroPaDy model can hardly been developed by the common agent-based modeling platforms, such as RePast, NetLogo, and Swarm, we are trying to use another alternative MATLAB to realize an empirical based application in an agricultural region of Northeast China, by taking the advantage of powerful and open-accessed matrix computing ability of MATLAB. We coded the model for the 3 interactive sub modules in steps: 1) Agents generating module. The Monte Carlo method was used to generate the internal factors (family attributes) for each individual agent in the full coverage study region by combining GIS data, statistical data, survey data and the individual based blanket rules. 2) Agent classifying module. The back propagation artificial neural network method was used to automatically classify the generated agents to groups based on the performance of surveyed agents, and the different groups were further linked with discrete probability on making a certain decision. 3) A probabilistic approach was used to determine the decisions of agent in each modeling period. The survey based data was used to support the empirical based application. After coding CroPaDy model in MATLAB with an input of 114 m×114 m grid-based ASCII file (total grid number = 29 799) plus 384 surveyed households randomly distributed on the selected grids, the model can successfully run and output model results for visualization and analysis. The results suggest that the crop areas of maize, rice, soybean, and tobacco are 26 055.9, 3 506.8, 5 192.2, 3 983.9 hm² respectively. Comparing with the local statistic yearbook, the overall accuracy of CroPaDy model can reach as high as 90%. Therefore, it is concluded that not only the conceptual framework of CroPaDy model is able to present the interactions between human and environment in agricultural land systems, but also the computational model can be finely programmed with MATLAB software. The study can further prove that crop pattern dynamics can be modeled by capturing farmer's land use decisions, and CroPaDy model can be applied in other similar regions if the detailed household survey data is available. [ABSTRACT FROM AUTHOR]

Abstract: Land use and its dynamics have attracted much attention from researchers due to their ecological and socio-economic implications. Many studies have used a modelling approach to evaluate land use changes and their effects. Most of these models were designed for the analysis of past, present and future cropland changes at different scales and few have been designed for the study of dynamic changes in sown areas of crops within croplands. This paper presents an integrated modelling approach to simulate dynamically the changes in sown areas for the world''s major crops at a global scale. This approach was based on three core models. A crop choice decision model, the Multinomial Logit model, was used to track and simulate the crop choice decisions made by individual farmers. A crop yield model, the GIS-based Environmental Policy Integrated Climate (EPIC) model, was utilized to estimate yields of different crop types under a given biophysical and management environment, while a crop price model, the International Food Policy and Agricultural Simulation (IFPSIM) model, was employed to assess the price of crops on the international market. Through data exchange, the crop choice decision model was linked with the crop yield and crop price models to allow the study of the dynamic feedback loop between changes in agricultural land use and biophysical and socio-economic driving factors. Sensitivity analysis and empirical validation for the model were conducted after the construction of the model. The model validation indicated the reliability of the model for addressing the complexity of current agricultural land use changes and its capacity for investigating long-term scenarios in the future. Finally, the model was used to simulate future scenarios over a time frame of 30 years with five-year increments, beginning from the year 2000. The simulation results provided insights into potential global cropping patterns, variation in rates and trajectory of changes in sown areas for major crops over the test period. These results can improve understanding of projected land use changes and explain their causes, locations and consequences, and provide support for land use planning and policy making. [Copyright &y& Elsevier]

Leaf area index (LAI) and canopy chlorophyll content (CCC) are important indicators that describe the growth status and nitrogen deficiencies of crops. Several studies have been performed to estimate LAI and CCC using multispectral cameras onboard an unmanned airborne vehicle (UAV) system. However, the impacts of illuminations during UAV flight and problems of how to invert still need more investigation. UAV flights with a multispectral camera were performed under clear (diffuse ratio 0) and cloudy illumination conditions (diffuse ratio 1) over rapeseed, wheat and sunflower (only clear) fields. One-dimension radiative transfer model PROSAIL was run twice to generate a clear-sky model and a cloudy-sky model, respectively. The LAI and CCC of flights under a clear sky were inverted from the clear-sky model, and the flights under cloudy conditions were inverted from both clear-sky and cloudy-sky models to compare the results. Moreover, three Look-Up-Tables (LUT) were built with same input variables but different distributions of LAI. Results showed that LAI from uniform dense LUT had better correspondence with ground measurements for all crops (R2 = 0.51~0.69). The illumination condition had little impact on small to medium LAI (LAI < 5) and CCC. However, the inversion of imageries during cloudy sky conditions from the clear-sky model led to an overestimation of high LAI values. [ABSTRACT FROM AUTHOR]

Information on crop spatial distribution is essential for agricultural monitoring and food security. Classification with remote-sensing time series images is an effective way to obtain crop distribution maps across time and space. Optimal features are the precondition for crop classification and are critical to the accuracy of crop maps. Although several approaches are available for extracting spectral, temporal, and phenological features for crop identification, these methods depend heavily on domain knowledge and human experiences, adding uncertainty to the final crop classification. This study proposed a novel Genetic Programming (GP) approach to learning high-level features from time series images for crop classification to address this issue. We developed a new representation of GP to extend the GP tree's width and depth to dynamically generate either fixed or flexible informative features without requiring domain knowledge. This new GP approach was wrapped with four classifiers, i.e., K-Nearest Neighbor (KNN), Decision Tree (DT), Naive Bayes (NB), and Support Vector Machine (SVM), and was then used for crop classification based on MODIS time series data in Heilongjiang Province, China. The performance of the GP features was compared with the traditional features of vegetation indices (VIs) and the advanced feature learning method Multilayer Perceptron (MLP) to show GP effectiveness. The experiments indicated that high-level features learned by GP improved the classification accuracies, and the accuracies were higher than those using VIs and MLP. GP was more robust and stable for diverse classifiers, different feature numbers, and various training sample sets compared with classification using VI features and the classifier MLP. The proposed GP approach automatically selects valuable features from the original data and uses them to construct high-level features simultaneously. The learned features are explainable, unlike those of a black-box deep learning model. This study demonstrated the outstanding performance of GP for feature learning in crop classification. GP has the potential of becoming a mainstream method to solve complex remote sensing tasks, such as feature transfer learning, image classification, and change detection. [ABSTRACT FROM AUTHOR]

Crop diversity is crucial for producing more food and nutrition in the crowded planet and achieving agricultural sustainable development, and thus it is a hot topic in shaping policies aimed at ensuring food security. Many studies have revealed that enhanced crop diversity can benefit crop productivity. However, research on how to maintain a relatively high crop diversity at regional and national scales remains limited. This study attempts to examine the underlying mechanisms of crop diversity changes in China and eventually answer why China can maintain a high crop diversity from the spatial-temporal perspective. To achieve this end, the county level crop area dataset for the period of 1980–2014 was compiled and used to quantify the spatiotemporal dynamics of crop diversity in China. The result reveals that the China's crop diversity trended upward over past 35 years, evidenced by more than 7 major crops at national level and 4 major crops at county level having undergone massive planting process to maintain a high crop diversity. Spatially, the crop diversity increased in more than two-thirds of the counties, and its hotspots moved gradually to the south-west mountainous area. The natural factor of slope and the social factor of population density contributed to shape the crop diversity pattern in global effects. In contrast, the irrigation degree, elevation of cropland, mean annual temperature and precipitation affected the spatially non-stationary distribution of crop diversity at the local level. On the whole, the maintenance of a higher crop diversity in China not only was limited by natural conditions, but also subject to adopt the multi-cropping systems strategic choice for the country to agricultural conditions. We argued that crop diversity can be an indicator to draw agricultural zoning, and increasing crop diversity should be recognized as a policy tool to implement agricultural sustainable development strategy. [ABSTRACT FROM AUTHOR]

• PSTFS optimizes crop type classifications by fully mining crop phenology information. • PSTFS ranks features by importance and then iteratively prunes redundant features. • We used an SVM classifier to produce a high-quality corn map of Heilongjiang, China. • The optimal features identified by PSTFS are key crop phenological characteristics. • PSTFS can be extended to other land cover types and other dense remote sensing data. Accurate information on crop distribution and its changes is important for food security and environmental management. Although time series analysis is a widely-used and useful tool to characterize the seasonal dynamics of crops, the traditional image stacking approach misses important phenological events. This condition makes it difficult to identify the spectral and temporal features that are potentially important for crop identification, and therefore, makes it difficult to determine the optimal feature inputs for classifying crops with both high accuracy and low computation time. To address this gap, we developed a method to automatically select the spectro-temporal features by mining crop phenology information so as to improve the accuracy of crop classifications. This method of Phenology-based Spectral and Temporal Feature Selection (PSTFS) contains two major components: to identify the features with the highest separability between each pair of classes, and to prune redundant features to retain the best for classification. Using this optimal set of features and support vector machines (SVMs), we generated a high-quality corn cultivation map of China's Heilongjiang Province for 2011. The corn map had accuracies greater than 85% and agreed well with the corn census areas. We also demonstrate the goodness of this method for selecting features with high interpretability: it identified two phenological stages (three leaf and milky mature) that could best separate corn from other land use classes in the region. Our approach indicates the great potential for using the PSTFS method in conjunction with SVM classifiers to accurately map crop types based on satellite time series data. [ABSTRACT FROM AUTHOR]

• The MP-Net model aggregates context information. • The MP-Net model reduces the feature losses. • The MP-Net model efficiently and precisely classifies crops. Accurate crop classification map is of great significance in various fields such as the survey of agricultural resource, the analysis of existing circumstance on land application, the yield estimation of crop and the disaster warning. The methods based on machine learning and deep learning are popularly used in crop classification and recognition of remote sensing images. However, the crop classification task based on neural networks still faces significant challenges due to the spatial and temporal distribution of crops and the inherent characteristics of remote sensing images. Therefore, this study proposes the multi-layer pyramid crop classification network (MP-Net) to solve the above problems. To reduce the feature loss during the crop extraction, the proposed model uses the pyramid pooling module to improve the ability of global information acquisition, and the information concatenation module to retain the upper features. Using the GF-6 and Sentinel-2 satellite data, the proposed model was tested in Erhai Lake Basin and Beian City. Compared with other five deep learning models, such as FCN, SegNet, U-Net, PSPNet and DeepLabv3+, the experimental results indicate that the proposed model achieves the highest accuracy in both study areas. Meanwhile, the proposed model has the advantages of short training time and high efficiency under the same running conditions. Overall, this study is beneficial to improve the efficiency and accuracy of crop classification task in the unbalanced temporal and spatial distribution. It also brings a feasible scheme for crop classification tasks in complex growing areas. The code has been publicly available at https://github.com/Xu-Chang-Hong/MP-Net. [ABSTRACT FROM AUTHOR]

Multiple cropping is a widespread agricultural practice that can partly reflect cropland use intensity. Previous studies have largely focused on harvest frequencies within an annual (e.g., multiple cropping frequency [MCF]) or multiannual basis. However, the complexity of cropland use intensity, which combines different cropping and fallow strategies, has not been fully investigated. Moreover, such information is particularly lacking at the regional scale. This study proposed a new perspective of multiple cropping system (MCS), and developed a method for mapping stable MCSs for a prolonged time. We selected an agricultural region (Poyang Lake Plain, Jiangxi Province, China) and proposed four stable MCSs in terms of the number of crops (C) grown in a given number of years (Y) defined by the length of crop rotation: Continuous single-cropping system (1C/1Y), continuous double-cropping system (2C/1Y), and two alternating systems (3C/2Y and 5C/3Y). To map MCSs, we first obtained the remote-sensing-observed annual MCF from 2001 to 2018. Second, total harvests, frequency, and continuity were proposed based on multi-year MCF information to classify each pixel as one of the four proposed MCSs. Total harvests were defined as the total number of harvests over the observation period. Frequency was defined as the total number of rotation units over the observation period for a given MCS. We defined continuity as the largest number of consecutive years under a given MCS. Finally, we determined the final type of an MCS using decision rules considering the three indicators. The results showed that the overall accuracy of the MCF maps were > 90%, and the overall accuracy of MCS map was 87.8%. We found that 84.4% of croplands could be characterized as one of the four MCSs: 2C/1Y accounted for the largest area (38.1%), followed by the two alternating systems (together they accounted for 48.6% of the total), and finally 1C/1Y. The results of our study suggest that multiple cropping practices should be characterized from the cropping system perspective. The MCS mapping method has great potential for large-scale crop rotation system mapping, and the MCS mapping results could help explore the combined effects of different cropping systems and further provide strategies for optimizing the spatially adapted practices in agricultural land systems. [Display omitted] • The complexity of cropland use intensity that combines different cropping and fallow strategies has not been fully studied. • Multiple cropping systems (MCSs) depict the fluctuating changes in multiple cropping frequency across years. • We developed a method for mapping MCSs through a case study and using decision rules and indicators. • Almost half of the MCSs alternate single-cropping and double-cropping regularly rather than keeping constant. • Multiple cropping practices should be characterized from the cropping system perspective. [ABSTRACT FROM AUTHOR]

To predict regional-scale winter wheat yield, we developed a crop model and data assimilation framework that assimilated leaf area index (LAI) derived from Landsat TM and MODIS data into the WOFOST crop growth model. We measured LAI during seven phenological phases in two agricultural cities in China’s Hebei Province. To reduce cloud contamination, we applied Savitzky–Golay (S–G) filtering to the MODIS LAI products to obtain a filtered LAI. We then regressed field-measured LAI on Landsat TM vegetation indices to derive multi-temporal TM LAIs. We developed a nonlinear method to adjust LAI by accounting for the scale mismatch between the remotely sensed data and the model’s state variables. The TM LAI and scale-adjusted LAI datasets were assimilated into the WOFOST model to allow evaluation of the yield estimation accuracy. We constructed a four-dimensional variational data assimilation (4DVar) cost function to account for the observations and model errors during key phenological stages. We used the shuffled complex evolution–University of Arizona algorithm to minimize the 4DVar cost function between the remotely sensed and modeled LAI and to optimize two important WOFOST parameters. Finally, we simulated winter wheat yield in a 1-km grid for cells with at least 50% of their area occupied by winter wheat using the optimized WOFOST, and aggregated the results at a regional scale. The scale adjustment substantially improved the accuracy of regional wheat yield predictions ( R 2 = 0.48; RMSE = 151.92 kg ha −1 ) compared with the unassimilated results ( R 2 = 0.23; RMSE = 373.6 kg ha −1 ) and the TM LAI results ( R 2 = 0.27; RMSE = 191.6 kg ha −1 ). Thus, the assimilation performance depends strongly on the LAI retrieval accuracy and the scaling correction. Our research provides a scheme to employ remotely sensed data, ground-measured data, and a crop growth model to improve regional crop yield estimates. [ABSTRACT FROM AUTHOR]

• Knowledge-based methods for mapping cropping intensity dynamics are highly demanded. • A dynamic Bayesian network for change analysis model is proposed for this purpose. • The model is a crop-phenology-knowledge-based model for automatic detection. • The model is a semantic-information-based model for detecting the trend. • The model can also extract the interest points and detect cropping intensity anomaly. The demand for agricultural products continues to increase while there is little room for cropland expansion. Agricultural intensification on existing croplands may provide a promising solution to increase agricultural production and alleviate the human-land conflict. To achieve this, the spatio-temporal dynamics of cropping intensity are essential information for agricultural production. However, existing methods are usually based on machine learning and are highly reliant on training samples. There are relatively few studies on automatic change detection of cropping intensity. In this study, a knowledge-based dynamic Bayesian network model for change analysis (DBN-CA) was developed for automatic detection of cropping intensity dynamics as well as the overall trends. Four statuses of cropping intensity dynamics were designed by referencing crop phenology to delineate the temporal dynamics of cropping intensity, and the overall trends from 2001 to 2019 were extracted based on their combinations. The conclusions were as follows: (1) the DBN-CA model based on crop phenology knowledge has a remarkable ability to effectively detect cropping intensity dynamics as well as its overall trends. (2) The cropping intensity in Hubei Province has shown a significant decreasing trend during the early 21st century. [ABSTRACT FROM AUTHOR]

Reliable crop type maps are vital for agricultural monitoring, ensuring food security, and environmental sustainability assessments. Coarse-resolution imagery such as MODIS are widely used for crop type mapping due to their short revisit cycles, which is advantageous for detecting the seasonal dynamics of different crop types. However, the inherently low spatial resolution may restrict their utility for mapping crop types in regions with heterogeneous agricultural landscapes. Agricultural statistics, which provide crop acreage information at different spatial and temporal scales, have the potential to improve crop type mapping from remote sensing. Yet, previous studies have often used agricultural statistics as reference data to evaluate the accuracy of satellite-derived crop type maps but have rarely utilized them to improve crop type distribution mapping. The utility of integrating agricultural statistics with satellite images to produce high-accuracy crop type maps is rarely explored. This study presents a methodology for mapping sub-pixel crop type distributions via the integration of MODIS time series and agricultural statistics. We tested our approach in Heilongjiang Province, which has the highest agricultural production in China. First, we used an optimized random forest regression (RF-r) model with training samples derived from high spatial resolution images (i.e. SPOT and Landsat) to predict the sub-pixel crop type distributions from MODIS time series. To optimize the RF-r model, an 8-day MODIS time series of five vegetation indices in 2011 were used as the candidate independent variables, and a backward feature elimination strategy was implemented to select the best variables for model prediction. Second, we developed an Iterative Area Gap Spatial Allocation (IAGSA) method to spatially reconcile the discrepancies between the crop acreage estimated from MODIS-based maps and the agricultural statistics. We found that the MODIS-derived crop fractions agreed with those derived from the high-resolution images, with R2 > 0.75 for all crop types, yet there was a clear discrepancy between the crop acreage estimated from MODIS and agricultural statistics. The sub-pixel crop type maps adjusted by IAGSA were not only consistent with the agricultural statistics for crop acreage, but also retained the spatial distribution patterns of the original MODIS-derived crop fraction. Our results suggest the advantages of integrating coarse-resolution images and agricultural statistics to map sub-pixel crop type distributions, and to provide consistent estimation of crop acreage. The presented methodology has the potential to map large-scale crop type extent across regions in a cost-effective way. • Integrating MODIS images and agricultural statistics for crop type mapping. • Combing random forest with backward elimination strategy to select optimal features. • IAGSA reconciles the discrepancy between MODIS-based crop area and statistical data. • It is tested in Heilongjiang Province and generated high-quality crop fraction maps. • This method can be extended to other crop types and other coarse resolution images. [ABSTRACT FROM AUTHOR]