Open Life Sci. 2018; 13: 533–543 Research Article Jinqiu Zou, Yinlan Huang, Lina Chen, Shi Chen* Remote Sensing-Based Extraction and Analysis of Temporal and Spatial Variations of Winter Wheat Planting Areas in the Henan Province of China https://doi.org/10.1515/biol-2018-0064 Keywords: winter wheat planting (WWP) area; spatial Received April 9, 2018; accepted October 31, 2018 and temporal variations; remote sensing monitoring; Abstract: The aim of this study is to assess the winter planting probability wheat planting (WWP) area in Henan Province and investigate its temporal and spatial variations by using remote sensing (RS) technology. A spectral angle mapper 1 Introduction (SAM) was adopted to identify the WWP area of each district divided by the hierarchical grades of land surface Henan province of China is a large-scale producer of drought index during 2001-2015. The results obtained winter wheat, whose crop yield is closely correlated to show the expediency of monitoring the WWP areas at the planting area and applied planting strategies. In the regional scale via drought regionalization, which view of global climate change, especially, the effect of provides a goodness-of-fit R2 =0.933, a mean relative error global warming on the safety of agricultural production, MRE=49,118 ha, and an overall accuracy up to 90.24%. real-time accurate monitoring of winter wheat planting The major WWP areas in Henan Province were located in (WWP) area data becomes a “must-have” feature of state- Zhoukou, Zhumadian, Shangqiu, Nanyang, and Xinxiang of-the-art agricultural management, which stimulates the prefecture-level cities. Two representative sites are relevant government departments in China to master the mountainous districts, with rich water resources or high monitoring techniques [1]. In this respect, remote sensing urbanization rate, which have a low probability of WWP. (RS) technology possesses a series of advantages, such Both sites exhibited a strongly manifested evolution of as a wide coverage, short data updating cycle, strong WWP areas, which could be attributed to extremely cold objectivity, and low cost, which makes it a lucrative tool weather conditions, crop alternation, the popularization for the accurate, rapid, and convenient extraction of the of new varieties, and fast expansion of built-up areas. WWP area data [2, 3]. The results of this study are instrumental in the analysis Thus, a comprehensive analysis of phenological of crop planting variation characteristics, which should (i.e., related to periodic plant life cycle events influenced be taken into account in the further decision-making by seasonal and interannual variations in climate) process related to the crop planting strategies. characteristics can be considerably enhanced by RS-based monitoring. Since the pioneer application of RS technology to the estimation of wheat area and total yield by the *Corresponding author: Shi Chen, Institute of Agricultural Resources US researchers in 1974 [4], international scholars have and Regional Planning, Chinese Academy of Agricultural Sciences, reported numerous achievements in this domain and Beijing, 100081, China; Bureau of Land Resources Lishui, Lishui, implemented several innovative RS-treatment procedures, 323000, China, E-mail: [email protected] which can be briefly outlined as follows: Jinqiu Zou, Institute of Agricultural Resources and Regional Plan- ning, Chinese Academy of Agricultural Sciences, Beijing, 100081, 1) Visual interpretation and automated computerised China identification. According to the phenological Yinlan Huang, Institute of Agricultural Resources and Regional Pl- characteristics of winter wheat growth and development, anning, Chinese Academy of Agricultural Sciences, Beijing, 100081, RS images are appropriately selected, and the multi- China; Bureau of Land Resources Lishui, Lishui, 323000, China dimensional green map is constructed. Then, the Lina Chen, School of Information Technology, Shangqiu Normal University; Shangqiu, 476000, China winter wheat data are hierarchically and automatically Open Access. © 2018 Jinqiu Zou et al., published by De Gruyter. This work is licensed under the Creative Commons Attribution-NonCommercial- NoDerivs 4.0 License. 534 J. Zou, et al. extracted via the pattern recognition, which is then factors, WWP in Henan underwent substantial changes. corrected through the artificial visual interpretation [5, The data were gathered using Moderate Resolution 6] with either supervised or unsupervised classification Imaging Spectroradiometer (MODIS) instruments located options [7, 8]. aboard the Terra and Aqua satellites. The aim of this study 2) Options of pixel-based extraction (for areas with a is to monitor WWP area in Henan and assess the planting wide planting range of winter wheat but a simple probability, reflecting variations in the planting structure. planting structure) and sub-pixel one (for areas with The present study is anticipated to be instrumental a complex terrain, scattered fields, and diverse crops). in getting a deeper insight into the driving forces of Here spectral unmixing is used, which is a quantitative winter wheat spatial and temporal variations, whose analysis procedure used to recognize constituent account should promote further agricultural sustainable ground cover materials (or endmembers) and obtain development and food security. their mixing proportions (or abundances) from a mixed pixel [9]. The extraction accuracy of planting area depends on the selection of appropriate endmembers 2 Research Area in the pixel unmixing [10, 11]. Henan province of China is located in the middle and 3) Plot-based extraction. The complexity of mixed pixel lower reaches of the Yellow River (110°21′E-116°39′E, and the differences between the spectra of different 31°23′N-36°22′N). Regarding terrain, it is high in the east objects on the ground can be reduced with the use of but low in the west, while regarding climate, it is in a plot data, thus improving the extraction accuracy and transition between temperate and subtropical monsoon operating stability to a certain degree [12]. climates. Henan is characterized by a moderate climate, 4) Extraction of planting area is based on the RS monitoring four distinct seasons, abundant sunshine and ample model. According to the variation rules of the vegetation rainfall, with an average temperature of 12℃-16℃, an index in different key phases of the winter wheat growth, average sunshine duration of 1740-2310 h. Average annual the spectral thresholds for distinguishing different precipitation in Henan ranges from 500 to 900 mm but objects are adjusted, the RS-based monitoring model of steadily decreases from south to north. An overwhelming winter wheat is established, and moreover, in order to majority of lands possesses deep soil layers, nutrient- enhance the extraction accuracy, some auxiliary data rich and fertile soils, abundant heat, light and water are introduced, and effective classification rules are resources. All of these favorable conditions have laid constructed using a decision-making tree [13, 14]. solid foundations for agricultural development in Henan. 5) Extraction is based on textual features: the winter According to Henan Economy and Social Development wheat spectral, phonological, and textural features are Statistical Bulletin in 2015, the grain planting area in Henan taken into account for further enhancement of winter amounted to 10286150 hm2, with a share of winter wheat wheat identification precision [15]. being 53.14% or 5465660 hm2, which strongly suggests that the winter wheat crop is an integral component of Conclusively, the above achievements and procedures Henan’s agricultural production (Figure 1). have expanded the application range of RS technology in the monitoring of WWP area. MODIS has advantages such as short revisit period, 3 Data source and processing wide coverage and characterization of phenology [16]. It has become an important data source for monitoring 3.1 Data source WWP area [17]. It can give full play to the temporal and spectral characteristics of winter wheat. The method Global MOD13Q1 data are provided by National Aeronautics of monitoring WWP area using MODIS data is mainly and Space Administration (NASA) every 16 days at a 250- to master the variation of phenological curve in winter meter spatial resolution as a gridded level-3 product in the wheat growth period. According to the variation of winter Sinusoidal projection. MODIS medium-resolution imaging wheat growth phenology curve and the difference of other data show a series of advantages mainly including a wide crops, the WWP area is extracted [18,19] using the shape field of view, large revisiting period and free sharing, and characteristics of winter wheat growth curve. are lucrative for the large-area RS monitoring of WWP. In recent years, due to the effects of both natural In this study, MODIS 13Q1 data were downloaded from factors (such as crop alternation, the adjustment of the official website of NASA with temporal and spatial planting structure, and farmland occupation) and human Remote Sensing-Based Extraction and Analysis of Temporal and Spatial Variations... 535 Figure 1. Overview of farmlands in Henan resolutions of 16 days and 250m×250m, respectively. The ETM+ (Enhanced Thematic Mapper Plus), and HJ-1 in time dimension of the data was from 2000 to 2015. Then, 2010 [21, 22]. Noteworthy is the data masking frequently using MODIS Reproduction Tools (MRT) software, data applied to the data fields,