Mapping paddy rice in Jiangsu Province, China, based on phenological parameters and a decision tree model

doi:10.1007/s11707-018-0723-y

Front. Earth Sci.

2019, Vol. 13

Issue (1) : 111-123 https://doi.org/10.1007/s11707-018-0723-y

RESEARCH ARTICLE

Mapping paddy rice in Jiangsu Province, China, based on phenological parameters and a decision tree model

Jianhong LIU^1,²(

), Le LI³, Xin HUANG², Yongmei LIU^1,², Tongsheng LI^1,²

¹. Shaanxi Key Laboratory of Earth Surface System and Environmental Carrying Capacity, Northwest University, Xi’an 710127, China
². College of Urban and Environmental Science, Northwest University, Xi’an 710127, China
³. Guangdong Research Center of Smart Homeland Engineering, South China Normal University, Guangzhou 510631, China

Download: PDF(7386 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Timely and accurate mapping of rice planting areas is crucial under China’s current cropping structure. This study proposes a new paddy rice mapping method by combining phenological parameters and a decision tree model. Six phenological parameters were developed to identify paddy rice areas based on the analysis of the Moderate Resolution Imaging Spectroradiometer (MODIS) Enhanced Vegetation Index (EVI) time series and the Land Surface Water Index (LSWI) time series. The six phenological parameters considered the performance of different land cover types during specific phenological phases (EVI₁ and EVI₂), one-half of or the entire rice growing cycle (LSWI₁ and LSWI₂), and the shape of the LSWI time series (Kurtosis_LSWI and Skewness_LSWI). A hierarchical decision tree model was designed to classify paddy rice areas according to the potential separability of different land cover types in paired phenological parameter spaces. Results showed that the decision tree model was more sensitive to LSWI₁, LSWI₂, and Skewness_LSWI than the other phenological parameters. A paddy rice map of Jiangsu Province for 2015 was generated with an optimal threshold set of (0.4, 0.42, 9, 19, 1.5, –1.7, 0.0) with a total accuracy of 93.9%. The MODIS-derived paddy rice map generally agreed with the paddy land fraction map from the National Land Cover Dataset project, but there were regional discrepancies because of their different definitions of land use and the inability of MODIS to map paddy rice at a fragmental level. The MODIS-derived paddy rice map showed high correlation (R²=0.85) with county-level agricultural statistics. The results of this study indicate that the phenological parameter-based paddy rice mapping algorithm could be applied at larger spatial scales.

Keywords phenological parameter paddy rice MODIS EVI LSWI decision tree

Corresponding Author(s): Jianhong LIU

Just Accepted Date: 12 September 2018 Online First Date: 13 November 2018 Issue Date: 25 January 2019

Cite this article:

Jianhong LIU,Le LI,Xin HUANG, et al. Mapping paddy rice in Jiangsu Province, China, based on phenological parameters and a decision tree model[J]. Front. Earth Sci., 2019, 13(1): 111-123.

URL:

https://academic.hep.com.cn/fesci/EN/10.1007/s11707-018-0723-y
https://academic.hep.com.cn/fesci/EN/Y2019/V13/I1/111

Fig.1 Location of the study area and the training samples, with Landsat-8 OLI images based on false-color composites of the red, near-infrared, and blue bands.

Tab.1 Landsat-8 OLI images used in this study

Fig.2 MODIS EVI time series (a) and the LSWI time series (b) of the main land cover types in the study area and (c) the LSWI time series curves of the only four land cover types that were inseparable using the EVI time series. The gray areas in (a) and (c) indicate the period of EVI/LSWI data used to compute the corresponding parameters. For example, EVI₁ was the minimum EVI value from June 10 to June 26.

Fig.3 Scatterplots of training samples in different feature spaces. Figs. 3(a), 3(b), and 3(c) show all the training samples scattered in the EVI₁-EVI₂, LSWI₁-LSWI₂, and Kurtosis_LSWI-Skewness_LSWI parameter spaces, respectively. The gray regions in Fig. 3(a) show the potential separation of paddy rice with the condition that EVI₁<0.4 and EVI₂>0.4. Fig. 3(d) shows the distribution of the training samples in the gray regions of Fig. 3(a) in the LSWI₁-LSWI₂ parameter space. The gray regions in Fig. 3(d) indicate the potential separation of paddy rice under the conditions that 9<LSWI₁<19 and LSWI₂>1.5. Fig. 3(e) shows the distribution of the training samples in the gray regions of Fig. 3(b) in the Kurtosis_LSWI-Skewness_LSWI parameter space. The gray regions in Fig. 3(e) indicate the potential separation of paddy rice under the conditions that Kurtosis_LSWI>?1.7 and Skewness_LSWI<0.

Fig.4 Flowchart of the decision tree classification.

Tab.2 Overall accuracy (OA) of each test

Fig.5 The 500-m-MODIS derived paddy rice map (a) and the 1-km-NLCD-derived paddy land fraction map (b) of Jiangsu Province in 2015. The NLCD-derived paddy land fraction map was generated by resampling the paddy land class into a 1-km resolution. Regions A?D are the major inconsistent regions between them.

Fig.6 Detailed comparison of Region A. (a) CBERS-04 image (RGB composited, R: red band, G: near-infrared band, B: blue band) on October 20, 2015. (b) Paddy rice from SVM classification overlaid on the CBERS-04 image. (c) Rice planting area derived from MODIS overlaid on the CBERS-04 image. (d) NLCD-derived 1-km paddy land fraction map overlaid on the CBERS-04 image.

Fig.7 Detailed comparisons for Regions B–D. The left column shows Landsat-8 OLI images (RGB composited, R: red band, G: near-infrared band, B: blue band) acquired in 2015 for each Region. The middle column shows the MODIS-derived rice planting area overlaid on the Landsat-8 OLI images. The right column shows the NLCD-derived 1-km paddy land fraction map overlaid on the Landsat-8 OLI images.

Fig.8 Comparisons of the paddy rice area estimated from MODIS (a) and from NLCD (b) to the reported statistics at the county level, respectively. The mapped MODIS rice pixels were assumed to be 100% fractional cover and the paddy field area from the NLCD was calculated from actual fractional cover at 1-km resolution.

1	BBouman (2009). How much water does rice use? Rice Today, (1): 15
2	B ABradley, J F Mustard (2008). Comparison of phenology trends by land cover class: a case study in the Great Basin, USA. Glob Change Biol, 14(2): 334–346 https://doi.org/10.1111/j.1365-2486.2007.01479.x
3	JDong, X Xiao (2016). Evolution of regional to global paddy rice mapping methods: a review. ISPRS J Photogramm Remote Sens, 119: 214–227 https://doi.org/10.1016/j.isprsjprs.2016.05.010
4	W SDvorak (2012). Water use in plantations of eucalypts and pines: a discussion paper from a tree breeding perspective. Int Rev, 14(1): 110–119 https://doi.org/10.1505/146554812799973118
5	M AFriedl, C E Brodley (1997). Decision tree classification of land cover from remotely sensed data. Remote Sens Environ, 61(3): 399–409 https://doi.org/10.1016/S0034-4257(97)00049-7
6	M AFriedl, D Sulla-Menashe, BTan, ASchneider, NRamankutty, ASibley, XHuang (2010). MODIS Collection 5 global land cover: algorithm refinements and characterization of new datasets. Remote Sens Environ, 114(1): 168–182 https://doi.org/10.1016/j.rse.2009.08.016
7	SFrolking, J Qiu, SBoles, XXiao, J Liu, YZhuang, CLi, X Qin (2002). Combining remote sensing and ground census data to develop new maps of the distribution of rice agriculture in China. Global Biogeochem Cycles, 16(4): 1091 https://doi.org/10.1029/2001GB001425
8	AHuete, K Didan, TMiura, E PRodriguez, XGao, L G Ferreira (2002). Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sens Environ, 83(1‒2): 195–213 https://doi.org/10.1016/S0034-4257(02)00096-2
9	J RJensen (2004). Introductory Digital Image Processing: A Remote Sensing Perspective (3nd ed). Englewood Cliffs: Prentice Hall
10	J SKimball, K C McDonald, S W Running, S E Frolking (2004). Satellite radar remote sensing of seasonal growing seasons for boreal and subalpine evergreen forests. Remote Sens Environ, 90(2): 243–258 https://doi.org/10.1016/j.rse.2004.01.002
11	M PLabus, G A Nielsen, R L Lawrence, R Engel, D SLong (2002). Wheat yield estimates using multi-temporal NDVI satellite imagery. Int J Remote Sens, 23(20): 4169–4180 https://doi.org/10.1080/01431160110107653
12	JLiu, W Kuang, ZZhang, XXu, Y Qin, JNing, WZhou, S Zhang, RLi, CYan, S Wu, XShi, NJiang, DYu, X Pan, WChi (2014a). Spatiotemporal characteristics, patterns, and causes of land-use changes in China since the late 1980s. J Geogr Sci, 24(2): 195–210 https://doi.org/10.1007/s11442-014-1082-6
13	JLiu, M Liu, HTian, DZhuang, ZZhang, WZhang, XTang, X Deng (2005). Spatial and temporal patterns of China’s cropland during 1990–2000: an analysis based on Landsat TM data. Remote Sens Environ, 98(4): 442–456 https://doi.org/10.1016/j.rse.2005.08.012
14	JLiu, Y Pan, XZhu, WZhu (2014b). Using phenological metrics and the multiple classifier fusion method to map land cover types. J Appl Remote Sens, 8(1): 083691 https://doi.org/10.1117/1.JRS.8.083691
15	JLiu, W Zhu, XCui (2012). A Shape-matching Cropping Index (CI) mapping method to determine agricultural cropland intensities in China using MODIS time-series data. Photogramm Eng Remote Sensing, 78(8): 829–837 https://doi.org/10.14358/PERS.78.8.829
16	T RLoveland, A S Belward (1997). The IGBP-DIS global 1 km land cover data set DISCover: first results. Int J Remote Sens, 18(15): 3289–3295 https://doi.org/10.1080/014311697217099
17	H DMatthews, K Caldeira (2007). Transient climate-carbon simulations of planetary geoengineering. Proc Natl Acad Sci USA, 104(24): 9949–9954 https://doi.org/10.1073/pnas.0700419104
18	J ROtukei, T Blaschke (20 10). Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms. Int J Appl Earth Obs Geoinf, 12(Supplement 1): S27–S31 https://doi.org/10.1016/j.jag.2009.11.002
19	MOzdogan, C E Woodcock (2006). Resolution dependent errors in remote sensing of cultivated areas. Remote Sens Environ, 103(2): 203–217 https://doi.org/10.1016/j.rse.2006.04.004
20	MPal, P M Mather (2003). An assessment of the effectiveness of decision tree methods for land cover classification. Remote Sens Environ, 86(4): 554–565 https://doi.org/10.1016/S0034-4257(03)00132-9
21	YPan, L Li, JZhang, SLiang, XZhu, D Sulla-Menashe (2012). Winter wheat area estimation from MODIS-EVI time series data using the crop proportion phenology index. Remote Sens Environ, 119(3): 232–242 https://doi.org/10.1016/j.rse.2011.10.011
22	DPeng, A R Huete, J Huang, FWang, HSun (2011). Detection and estimation of mixed paddy rice cropping patterns with MODIS data. Int J Appl Earth Obs Geoinf, 13(1): 13–23 https://doi.org/10.1016/j.jag.2010.06.001
23	JPongratz, D Lobell, LCao, KCaldeira (2012). Crop yields in a geoengineered climate. Nat Clim Chang, 2(2): 101–105 https://doi.org/10.1038/nclimate1373
24	A BPotgieter, AApan, G Hammer, PDunn (2010). Early-season crop area estimates for winter crops in NE Australia using MODIS. ISPRS J Photogramm Remote Sens, 65(4): 380–387 https://doi.org/10.1016/j.isprsjprs.2010.04.004
25	A BPotgieter, KLawson, A RHuete (2013). Determining crop acreage estimates for specific winter crops using shape attributes from sequential MODIS imagery. Int J Appl Earth Obs Geoinf, 23(8): 254–263 https://doi.org/10.1016/j.jag.2012.09.009
26	M JPringle, R J Denham, R Devadas (2012). Identification of cropping activity in central and southern Queensland, Australia, with the aid of MODIS MOD13Q1 imagery. Int J Appl Earth Obs Geoinf, 19(1): 276–285 https://doi.org/10.1016/j.jag.2012.05.015
27	YQin, X Xiao, JDong, YZhou, Z Zhu, GZhang, GDu, C Jin, WKou, JWang, X Li (2015). Mapping paddy rice planting area in cold temperate climate region through analysis of time series Landsat 8 (OLI), Landsat 7 (ETM+) and MODIS imagery. ISPRS J Photogramm Remote Sens, 105: 220–233 https://doi.org/10.1016/j.isprsjprs.2015.04.008
28	BQiu, W Li, ZTang, CChen, W Qi (2015). Mapping paddy rice areas based on vegetation phenology and surface moisture conditions. Ecol Indic, 56: 79–86 https://doi.org/10.1016/j.ecolind.2015.03.039
29	ARobock, L Oman, G LStenchikov (2008). Regional climate responses to geoengineering with tropical and arctic SO2 injections. J Geophys Res, 113: D16101 https://doi.org/10.1029/2008JD010050
30	TSakamoto, C Van Phung, AKotera, K DNguyen, MYokozawa (2009). Analysis of rapid expansion of inland aquaculture and triple rice-cropping areas in a coastal area of the Vietnamese Mekong Delta using MODIS time-series imagery. Landsc Urban Plan, 92(1): 34–46 https://doi.org/10.1016/j.landurbplan.2009.02.002
31	MStory, R G Congalton (1986). Accuracy assessment: a user’s perspective. Photogramm Eng Remote Sensing, 52(3): 397–399
32	T RTooke, N C Coops, N R Goodwin, J A Voogt (2009). Extracting urban vegetation characteristics using spectral mixture analysis and decision tree classifications. Remote Sens Environ, 113(2): 398–407 https://doi.org/10.1016/j.rse.2008.10.005
33	B DWardlow, S L Egbert (2008). Large-area crop mapping using time-series MODIS 250m NDVI data: an assessment for the U.S. Central Great Plains. Remote Sens Environ, 112(3): 1096–1116 https://doi.org/10.1016/j.rse.2007.07.019
34	XXiao, S Boles, SFrolking, CLi, J Y Babu, W Salas, BMoore III (2006). Mapping paddy rice agriculture in South and Southeast Asia using multi-temporal MODIS images. Remote Sens Environ, 100(1): 95–113 https://doi.org/10.1016/j.rse.2005.10.004
35	XXiao, S Boles, JLiu, DZhuang, SFrolking, CLi, W Salas, BMoore III (2005). Mapping paddy rice agriculture in southern China using multi-temporal MODIS images. Remote Sens Environ, 95(4): 480–492 https://doi.org/10.1016/j.rse.2004.12.009
36	GZhang, X Xiao, JDong, WKou, C Jin, YQin, YZhou, J Wang, M AMenarguez, CBiradar (2015). Mapping paddy rice planting areas through time series analysis of MODIS land surface temperature and vegetation index data. ISPRS J Photogramm Remote Sens, 106: 157–171 https://doi.org/10.1016/j.isprsjprs.2015.05.011
37	WZhu, Y Pan, HHe, LWang, M Mou, JLiu (2012). A changing-weight filter method for reconstructing a high-quality NDVI time series to preserve the integrity of vegetation phenology. IEEE Trans Geosci Remote Sens, 50(4): 1085–1094 https://doi.org/10.1109/TGRS.2011.2166965
38	JZou, Y Huang, XZheng, YWang (2007). Quantifying direct N2O emissions in paddy fields during rice growing season in mainland China: dependence on water regime. Atmos Environ, 41(37): 8030–8042 https://doi.org/10.1016/j.atmosenv.2007.06.049

[1]	Jianhong LIU, Clement ATZBERGER, Xin HUANG, Kejian SHEN, Yongmei LIU, Lei WANG. *Modeling grass yields in Qinghai Province, China, based on MODIS NDVI* data—an empirical comparison**[J]. Front. Earth Sci., 2020, 14(2): 413-429.
[2]	Mohammad Saeid MIRAKHORLO, Majid RAHIMZADEGAN. Evaluating estimated sediment delivery by Revised Universal Soil Loss Equation (RUSLE) and Sediment Delivery Distributed (SEDD) in the Talar Watershed, Iran[J]. Front. Earth Sci., 2020, 14(1): 50-62.
[3]	Xujun HU, Huiyuan ZHANG, Haiguang HAO, Danyang FENG, Haiyan LIU, Qiang ZHANG. Understanding the relationships between poverty alleviation and ecosystem conservation: empirical evidence from western China[J]. Front. Earth Sci., 2020, 14(1): 209-220.
[4]	Wanben WU, Wei WANG, Michael E. Meadows, Xinfeng YAO, Wei PENG. Cloud-based typhoon-derived paddy rice flooding and lodging detection using multi-temporal Sentinel-1&2[J]. Front. Earth Sci., 2019, 13(4): 682-694.
[5]	Yongfeng WANG, Zhaohui XUE, Jun CHEN, Guangzhou CHEN. Spatio-temporal analysis of phenology in Yangtze River Delta based on MODIS NDVI time series from 2001 to 2015[J]. Front. Earth Sci., 2019, 13(1): 92-110.
[6]	Xiaojuan HUANG, Mingguo MA, Xufeng WANG, Xuguang TANG, Hong YANG. The uncertainty analysis of the MODIS GPP product in global maize croplands[J]. Front. Earth Sci., 2018, 12(4): 739-749.
[7]	Donal O’Leary III, Dorothy Hall, Michael Medler, Aquila Flower. Quantifying the early snowmelt event of 2015 in the Cascade Mountains, USA by developing and validating MODIS-based snowmelt timing maps[J]. Front. Earth Sci., 2018, 12(4): 693-710.
[8]	David BRAND,Chathurika WIJEWARDANA,Wei GAO,K. Raja REDDY. Interactive effects of carbon dioxide, low temperature, and ultraviolet-B radiation on cotton seedling root and shoot morphology and growth[J]. Front. Earth Sci., 2016, 10(4): 607-620.
[9]	Helmut SCHAEBEN,Georg SEMMLER. The quest for conditional independence in prospectivity modeling: weights-of-evidence, boost weights-of-evidence, and logistic regression[J]. Front. Earth Sci., 2016, 10(3): 389-408.
[10]	Babula JENA. Satellite remote sensing of the island mass effect on the Sub-Antarctic Kerguelen Plateau, Southern Ocean[J]. Front. Earth Sci., 2016, 10(3): 479-486.
[11]	Qing XU,Hongyuan ZHANG,Yongcun CHENG. *Multi-sensor monitoring of Ulva prolifera* blooms in the Yellow Sea using different methods**[J]. Front. Earth Sci., 2016, 10(2): 378-388.
[12]	Weichun MA,Liguo ZHOU,Hao ZHANG,Yan ZHANG,Xiaoyan DAI. Air temperature field distribution estimations over a Chinese mega-city using MODIS land surface temperature data: the case of Shanghai[J]. Front. Earth Sci., 2016, 10(1): 38-48.
[13]	Yue LIU,Qiuming CHENG,Qinglin XIA,Xinqing WANG. The use of evidential belief functions for mineral potential mapping in the Nanling belt, South China[J]. Front. Earth Sci., 2015, 9(2): 342-354.
[14]	Youzhi AN,Wei GAO,Zhiqiang GAO,Chaoshun LIU,Runhe SHI. Trend analysis for evaluating the consistency of Terra MODIS and SPOT VGT NDVI time series products in China[J]. Front. Earth Sci., 2015, 9(1): 125-136.
[15]	Tianyong ZHAI,Qing ZHAO,Wei GAO,Runhe SHI,Weining XIANG,Hung-lung Allen HUANG,Chao ZHANG. Analysis of spatio-temporal variability of aerosol optical depth with empirical orthogonal functions in the Changjiang River Delta, China[J]. Front. Earth Sci., 2015, 9(1): 1-12.

Viewed

Full text

Abstract

Cited

Shared

Discussed