Crop discrimination in Northern China with double cropping systems using Fourier analysis of time-series MODIS data

Crop identification is the basis of crop monitoring using remote sensing. Remote sensing the extent and distribution of individual crop types has proven useful to a wide range of users, including policy-makers, farmers, and scientists. Northern China is not merely the political, economic, and cultural centre of China, but also an important base for grain production. Its main grains are wheat, maize, and cotton. By employing the Fourier analysis method, we studied crop planting patterns in the Northern China plain. Then, using time-series EOS-MODIS NDVI data, we extracted the key parameters to discriminate crop types. The results showed that the estimated area and the statistics were correlated well at the county-level. Furthermore, there was little difference between the crop area estimated by the MODIS data and the statistics at province-level. Our study shows that the method we designed is promising for use in regional spatial scale crop mapping in Northern China using the MODIS NDVI time-series.     

Monitoring winter wheat growth in North China by combining a crop model and remote sensing data

Both of crop growth simulation models and remote sensing method have a high potential in crop growth monitoring and yield prediction. However, crop models have limitations in regional application and remote sensing in describing the growth process. Therefore, many researchers try to combine those two approaches for estimating the regional crop yields. In this paper, the WOFOST model was adjusted and regionalized for winter wheat in North China and coupled through the LAI to the SAIL–PROSPECT model in order to simulate soil adjusted vegetation index (SAVI). Using the optimization software (FSEOPT), the crop model was then re-initialized by minimizing the differences between simulated and synthesized SAVI from remote sensing data to monitor winter wheat growth at the potential production level. Initial conditions, which strongly impact phenological development and growth, and which are hardly known at the regional scale (such as emergence date or biomass at turn-green stage), were chosen to be re-initialized. It was shown that re-initializing emergence date by using remote sensing data brought simulated anthesis and maturity date closer to measured values than without remote sensing data. Also the re-initialization of regional biomass weight at turn-green stage led that the spatial distribution of simulated weight of storage organ was more consistent to official yields. This approach has some potential to aid in scaling local simulation of crop phenological development and growth to the regional scale but requires further validation.     

Analysis of cropping pattern changes in Bathinda district,Punjab

A study was conducted in the Bathinda district of Punjab state for mapping the cropping pattern and crop rotation, monitoring long term changes in cropping pattern by using the satellite based remote sensing data along other spatial and non-spatial collateral data. Multi-date IRS LISS I and IRS WiFS sensor data have been used for this study. Cropping pattern maps and crop rotation maps were generated for the years 1988-89 and 1998-99. The present study has shown the increase of cropping intensity significantly, mainly due to increase in rice area. However, crop diversity has decreased mainly due to decline in the area under the minor crops like pearl millet, gram, rapeseed/ mustard. There is increase in area coverage of cotton-wheat and rice-wheat rotation, at the expense of the minor crops.     

Remote Sensing and Geographical Information System for canopy cover mapping

Availability of remote sensing data from earth observation satellites has made it convenient to map and monitor land use/land cover at regional to local scales. A land cover map is very critical for a various planning activities including watershed planning. The spectral and spatial resolutions are major constraints for mapping the crop resources at microlevel. The cropping pattern zones have been mapped using the false color composite, physiography, irrigation and toposheets. The IRS LISS-III data is classified into various categories depending on spectral reflectance from crop canopy and are overlaid on cropping zones map. The re-classified resultant map provides land use/land cover information including dominant cropping systems. The canopy cover is estimated monthly considering the crop calendar for the area.     

The use of MODIS data to derive acreage estimations for larger fields: A case study in the south-western Rostov region of Russia

Recent developments in remote sensing technology, in particular improved spatial and temporal resolution, open new possibilities for estimating crop acreage over larger areas. Remotely sensed data allow in some cases the estimation of crop acreage statistics independently of sub-national survey statistics, which are sometimes biased and incomplete. This work focuses on the use of MODIS data acquired in 2001/2002 over the Rostov Oblast in Russia, by the Azov Sea. The region is characterised by large agricultural fields of around 75 ha on average. This paper presents a methodology to estimate crop acreage using the MODIS 16-day composite NDVI product. Particular emphasis is placed on a good quality crop mask and a good quality validation dataset. In order to have a second dataset which can be used for cross-checking the MODIS classification a Landsat ETM time series for four different dates in the season of 2002 was acquired and classified. We attempted to distinguish five different crop types and achieved satisfactory and good results for winter crops. Three hundred and sixty fields were identified to be suitable for the training and validation of the MODIS classification using a maximum likelihood classification. A novel method based on a pure pixel field sampling is introduced. This novel method is compared with the traditional hard classification of mixed pixels and was found to be superior.     

Disaggregating and mapping crop statistics using hypertemporal remote sensing

Governments compile their agricultural statistics in tabular form by administrative area, which gives no clue to the exact locations where specific crops are actually grown. Such data are poorly suited for early warning and assessment of crop production. 10-Daily satellite image time series of Andalucia, Spain, acquired since 1998 by the SPOT Vegetation Instrument in combination with reported crop area statistics were used to produce the required crop maps. Firstly, the 10-daily (1998–2006) 1-km resolution SPOT-Vegetation NDVI-images were used to stratify the study area in 45 map units through an iterative unsupervised classification process. Each unit represents an NDVI-profile showing changes in vegetation greenness over time which is assumed to relate to the types of land cover and land use present. Secondly, the areas of NDVI-units and the reported cropped areas by municipality were used to disaggregate the crop statistics. Adjusted R-squares were 98.8% for rainfed wheat, 97.5% for rainfed sunflower, and 76.5% for barley. Relating statistical data on areas cropped by municipality with the NDVI-based unit map showed that the selected crops were significantly related to specific NDVI-based map units. Other NDVI-profiles did not relate to the studied crops and represented other types of land use or land cover. The results were validated by using primary field data. These data were collected by the Spanish government from 2001 to 2005 through grid sampling within agricultural areas; each grid (block) contains three 700 m × 700 m segments. The validation showed 68%, 31% and 23% variability explained (adjusted R-squares) between the three produced maps and the thousands of segment data. Mainly variability within the delineated NDVI-units caused relatively low values; the units are internally heterogeneous. Variability between units is properly captured. The maps must accordingly be considered “small scale maps”. These maps can be used to monitor crop performance of specific cropped areas because of using hypertemporal images. Early warning thus becomes more location and crop specific because of using hypertemporal remote sensing.     

基于NDVI时间序列数据的江西省水稻种植制度变化研究

利用Savitzky-Golay滤波对覆盖江西省范围的SPOT VGT NDVI时间序列数据进行平滑处理的基础上,结合坡度数据,通过非监督分类的方法提取了江西省2000、2005和2010年水稻种植范围,并根据NDVI的年内动态变化,从水稻种植范围、水稻生长季起始时间、水稻复种指数和NDVI最大振幅等分析了江西省水稻种植和生长情况,探讨2000~2010年江西省水稻生产的变化。     

Agricultural Area Diversification and Crop Water Demand Analysis: A Remote Sensing and GIS Approach

Large scale adoption of input intensive rice–wheat cropping system in the centrally located Jalandhar district of Indian Punjab has led to over-exploitation of ground water resources, intensive use of chemical fertilizers and deterioration of soil health. To overcome these shortfalls, in the present study, agricultural area diversification plan has been generated from agricultural area and crop rotation maps derived from remote sensing data (IRS P6-AWiFS and RADARSAT ScanSAR) along with few agro-physical parameters in GIS environment. Cropping system indices (area diversity, multiple cropping and cultivated land utilization) were also worked out from remote sensing data .Analysis of remote sensing data (2004–05) revealed that rice and wheat individually remained the dominant crops, occupy 57.8% and 64.9% of total agricultural area (TAA), respectively. Therefore, in the diversified plan, it is suggested that at least 39% of the current 40% TAA under rice–wheat rotation should be replaced by other low water requiring, high value and soil enriching crops, particularly in coarse textured alluvial plain having good quality ground water zones with low annual rainfall(<700 mm). This will reduce water requirement to the tune of 15,660 cm depth while stabilizing the production and profitability by crop area diversification without further degradation of natural resources.     

Geographical model for precise agriculture monitoring with real-time remote sensing

This paper presents a remote sensing model for crop monitoring that was developed by the authors in a multi-year study. It also presents two experiments conducted for testing a newly developed application. The model combines remote sensing models using mapping of the spatial distribution of vegetation in an agricultural field, with precision agricultural models that maximize the output (yield) while minimizing the input (cost). This combination enables one to operate a monitoring and management process that includes every sub-unit of the field using remote sensing mapping.The model consists of five steps: (1) Preparing information layers that map the crop-affecting elements, e.g. irrigation and topography; (2) Collecting spectral and plant data simultaneously; (3) Processing and analyzing the data in order to prepare vegetation maps; (4) Decision-making in accordance with the above-mentioned maps or with predicted-yield maps; and (5) Quality control.The experiments showed that although the results were not statistically significant, the application of the proposed model enables one to draw recommendations within 45 h, and that remote sensing monitoring results in more benefits than do traditional control methods. The quality control was not ideal, due to the narrow range of the spectrum used in the remote sensing monitoring.     

基于遥感技术的土地利用信息提取及变化

遥感图像是地理信息分析非常重要的数据源。遥感技术和GIS技术被广泛应用于土地利用动态监测。利用1987年和2000年的遥感影像,分析了徐州市14年间土地利用动态变化情况,并运用ArcView实现土地利用的空间分析,对徐州市土地利用状况进行了研究,分析了各土地利用类型变更情况。以耕地为例,利用徐州市的社会经济统计数据,运用多元统计分析方法,基于SPSS软件对耕地变化的主要驱动力和驱动机制进行了分析。     

Crop type mapping using LiDAR,Sentinel-2 and aerial imagery with machine learning algorithms

LiDAR data are becoming increasingly available, which has opened up many new applications. One such application is crop type mapping. Accurate crop type maps are critical for monitoring water use, estimating harvests and in precision agriculture. The traditional approach to obtaining maps of cultivated fields is by manually digitizing the fields from satellite or aerial imagery and then assigning crop type labels to each field - often informed by data collected during ground and aerial surveys. However, manual digitizing and labeling is time-consuming, expensive and subject to human error. Automated remote sensing methods is a cost-effective alternative, with machine learning gaining popularity for classifying crop types. This study evaluated the use of LiDAR data, Sentinel-2 imagery, aerial imagery and machine learning for differentiating five crop types in an intensively cultivated area. Different combinations of the three datasets were evaluated along with ten machine learning. The classification results were interpreted by comparing overall accuracies, kappa, standard deviation and f-score. It was found that LiDAR data successfully differentiated between different crop types, with XGBoost providing the highest overall accuracy of 87.8％. Furthermore, the crop type maps produced using the LiDAR data were in general agreement with those obtained by using Sentinel-2 data, with LiDAR obtaining a mean overall accuracy of 84.3％ and Sentinel-2 a mean overall accuracy of 83.6％. However, the combination of all three datasets proved to be the most effective at differentiating between the crop types, with RF providing the highest overall accuracy of 94.4％. These findings provide a foundation for selecting the appropriate combination of remotely sensed data sources and machine learning algorithms for operational crop type mapping.     

遥感与GIS技术在区域农业地质调查中的应用

较为详细地介绍了遥感与GIS技术在区域生态环境与农业地质研究中的应用。WVS分类（Water, Vegetation and Soil Classification）及地面覆被分类图的编制，为区域性生态环境研究提供了重要的基础资料。在此基础上，结合归一化变换、相关分析与熵的引入，为区域性生态地球化学环境评价提供了定量分析数据。     

Integrated approach of using remote sensing and GIS to study watershed prioritization and productivity

Soil data obtained from soil resource inventory, land and climate were derived from the remote sensing satellite data (Landsat TM, bands 1 to 7) and were integrated in GIS environment to obtain the soil erosion loss using USLE model for the watershed area. The priorities of different sub-watershed areas for soil conservation measures were identified. Land productivity index was also used as a measure for land evaluation. Different soil and land attribute maps were generated in GIS, and R,K,LS,C and P factor maps were derived. By integrating these soil erosion map was generated. The mapping units, found not suitable for agriculture production, were delineated and mapped as non-arable land. The area suitable for agricultural production was carved out for imparting the productivity analysis; the land suitable for raising agricultural crops was delineated into different mapping units as productivity ratings good, fair, moderate and poor. The analysis performed using remote sensing and GIS helped to generate the attribute maps with more accuracy and the ability of integrating these in GIS environment provided the ease to get the required kind of analysis. Conventional methods of land evaluation procedures in terms of either soil erosion or productivity are found not comparable with the out put generated by using remote sensing and GIS as the limitations in generating the attribute maps and their integration. The results obtained in this case study show the use of different kinds of data derived from different sources in land evaluation appraisals.     

Utility of satellite data for land evaluation in land use planning for a part of Northern Karnataka

The study reported herein deals with the utility of satellite remote sensing techniques for land evaluation for agricultural land use planning. False colour composite of Landsat imagery in the scale of 1:250,000 was visually interpreted for physiography that formed the base for mapping soil and land resources in the field. The small-scale soil map thus prepared has thirteen map units with association of soil families. Soil and land resource units shown on these small-scale maps were evaluated for their suitability for growing sorghum crop by matching the relevant land qualities against the land requirements for sorghum. The land evaluation carried out for growing sorghum crop in the study area revealed that about 38.6 per cent is highly suitable (S1), 31.5 per cent moderately suitable (S2) and 24.5 per cent marginally suitable (S3). An area of about 5.4 per cent is not suitable, of which 3.0 per cent is currently not suitable (N1) and 2.4 per cent permanently not suitable for growing sorghum crop.     

Yield Forecasting of Spring Maize Using Remote Sensing and Crop Modeling in Faisalabad-Punjab Pakistan

Real time, accurate and reliable estimation of maize yield is valuable to policy makers in decision making. The current study was planned for yield estimation of spring maize using remote sensing and crop modeling. In crop modeling, the CERES-Maize model was calibrated and evaluated with the field experiment data and after calibration and evaluation, this model was used to forecast maize yield. A Field survey of 64 farm was also conducted in Faisalabad to collect data on initial field conditions and crop management data. These data were used to forecast maize yield using crop model at farmers’ field. While in remote sensing, peak season Landsat 8 images were classified for landcover classification using machine learning algorithm. After classification, time series normalized difference vegetation index (NDVI) and land surface temperature (LST) of the surveyed 64 farms were calculated. Principle component analysis were run to correlate the indicators with maize yield. The selected LSTs and NDVIs were used to develop yield forecasting equations using least absolute shrinkage and selection operator (LASSO) regression. Calibrated and evaluated results of CERES-Maize showed the mean absolute % error (MAPE) of 0.35–6.71% for all recorded variables. In remote sensing all machine learning algorithms showed the accuracy greater the 90%, however support vector machine (SVM-radial basis) showed the higher accuracy of 97%, that was used for classification of maize area. The accuracy of area estimated through SVM-radial basis was 91%, when validated with crop reporting service. Yield forecasting results of crop model were precise with RMSE of 255 kg ha^?1, while remote sensing showed the RMSE of 397 kg ha^?1. Overall strength of relationship between estimated and actual grain yields were good with R² of 0.94 in both techniques. For regional yield forecasting remote sensing could be used due greater advantages of less input dataset and if focus is to assess specific stress, and interaction of plant genetics to soil and environmental conditions than crop model is very useful tool.     

农业遥感研究应用进展与展望

得益于中国自主遥感卫星、无人机遥感和物联网等技术的发展,中国农业遥感研究与应用在过去20年取得了显著进步,中国农业遥感信息获取呈现出天地网一体化的趋势;农业定量遥感在关键参数遥感反演技术方法与应用方面取得进展;作物面积、长势、产量、灾害遥感监测的理论与技术方法取得突破,农业遥感技术应用领域不断拓展。本文从农业遥感信息获取、农业定量遥感、农业灾害遥感、作物遥感识别与制图、作物长势遥感监测与产量预测、农业土地资源遥感等方面对中国农业遥感科研与应用进行了总结综述。     

Application of remote sensing and GIS technology to the study of desertification of arable lands in North Shaanxi,China

The policy of the Chinese government concerning the horizontal expansion of the cultivated land through the reclamation of desert soils result in a total increase of 665. 985 km2 during the period 1987–1999 in North Shaanxi. This increase is less than the loss in arable land by urbanization. The accelerated rate of change in agricultural areas calls for more rapid surveys of urbanization and loss of arable land. Remote sensing has a number of advantages over ground-based methods for such surveys. The multi-scale concept of remote sensing data help us study the problem in four towns. Several maps were produced to analyze the situation of urban coverage in different times. The evaluation of the status, rate and risk of urbanization are based on an accepted average of urban increase as 2% of population growth per year.     

Application of Remote Sensing and GIS Technology to the Study of Desertification of Arable Lands in North Shaanxi, China

The policy of the Chinese government concerning the horizontal expansion of the cultivated land through the reclamation of desert soils result in a total increase of 665. 985 km^2 during the period 1987-1999 in North Shaanxi. This increase is less than the loss in arable land by urbanization. The accelerated rate of change in agricultural areas calls for more rapid surveys of urbanization and loss of arable land. Remote sensing has a number of advantages over ground-based methods for such surveys. The multi-scale concept of remote sensing data help us study the problem in four towns. Several maps were produced to analyze the situation of urban coverage in different times. The evaluation of the status, rate and risk of urbanization are based on an accepted average of urban increase as 2% of population growth per year.     

多时相MODIS影像水田信息提取研究

水稻种植及其分布信息是土地覆被变化、作物估产、甲烷排放、粮食安全和水资源管理分析的重要数据源。基于遥感的水田利用监测中,通常采用时序NDVI植被指数法和影像分类法分别进行AVHRR和TM影像的水田信息获取。针对8天合成MODIS陆地表面反射比数据的特点和水稻生长特征,选取水稻种植前的休耕期、秧苗移植期、秧苗生长期和成熟期等多时相MODIS地表反射率影像数据,通过归一化植被指数、增强植被指数及利用对土壤湿度和植被水分含量较敏感的短波红外波段计算得到的陆表水指数进行水田信息获取。将提取结果与基于ETM＋影像的国土资源调查水田数据,通过网格化计算处理并进行对比分析,结果表明,利用MODIS影像的8天合成地表反射率数据,进行区域甚至全国的水田利用监测是可行的。     

MODIS植被指数的美国玉米单产遥感估测

针对中国开展的国外农作物产量遥感估测大多依靠中低分辨率耕地信息、省级(州级)或国家级作物产量统计数据的现状,本文以美国玉米为例,探讨利用多年中高分辨率作物分布信息、时序遥感植被指数和县级作物产量统计数据开展国外重点地区作物单产遥感估测技术研究,以期进一步提高中国对国外农作物产量监测精度和精细化水平。首先,利用美国农业部国家农业统计局(NASS/USDA)生产的作物分布数据(CDL)获得多个年份玉米空间分布图,并对相应年份250 m分辨率16天合成的MODIS-NDVI时序数据进行掩膜处理,统计获得每年各县域内玉米主要生育期NDVI均值;其次,以各州为估产区,以多年县级玉米统计单产和县域内玉米主要生育期NDVI均值为基础,建立各州玉米主要生育期NDVI与玉米单产间关系模型;然后,通过主要生育期玉米单产和玉米植被指数间拟合程度,筛选确定各州玉米最佳估产期和最佳估产模型。最终,利用最佳估产模型实现美国各州玉米单产估测和全国玉米单产推算。其中,建模数据覆盖时间为2007年—2010年,验证数据为2011年。结果表明,应用最佳估产模型的2011年美国各州玉米单产估测相对误差在-4.16%—4.92%,均方根误差在148.75—820.93 kg/ha,各州估测结果计算获得全国玉米单产的相对误差仅为2.12%,均方根误差为285.57 kg/ha。可见,本研究的作物单产遥感估测技术方法具有一定可行性,可准确估测全球重点地区作物单产信息。