光谱学与光谱分析 |
|
|
|
|
|
Modeling and Predicting of MODIS Leaf Area Index Time Series Based on a Hybrid SARIMA and BP Neural Network Method |
JIANG Chun-lei1,2, ZHANG Shu-qing1*, ZHANG Ce3, LI Hua-peng1, DING Xiao-hui1,2 |
1. Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun 130102, China2. University of Chinese Academy of Sciences, Beijing 100049, China3. Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, UK |
|
|
Abstract The modeling and predicting of vegetation Leaf area index (LAI) is an important component of land surface model and assimilation of remote sensing data. The MODIS LAI product (i.e. MOD15A2) is one of the most widely used LAI data sources. However, the time series of MODIS LAI contains some data of low quality. For example, because of the influence of the cloud, aerosol, etc., the MODIS LAI presents the characteristics of the discontinuous in time and space. In fact, the time series of MODIS LAI include both linear and nonlinear components, which cannot be accurately modeled and predicted by either linear method or nonlinear method alone. In this paper, the original LAI time series data were first smoothed with Savitzky-Golay (SG) filtration and linear interpolation; SARIMA, BP neural network and a hybrid method of SARIMA-BP neural network were then used for modeling and predicting MODIS LAI time series. The SARIMA-BP neural network combined both SARIMA and BP neural network, which could model the linear and the nonlinear component of MODIS LAI time series respectively. That is, the final result of SARIMA-BP neural network was the sum of results of the two methods. Experiments showed that the time series of MODIS LAI that were smoothed with the SG filtration and linear interpolation were more smooth than original time series, with a determination coefficient up to 0.981, closer to 1 than that of SARIMA (0.941) and BP neural network (0.884); the correlation coefficient between SARIMA-BP neural network and the observation is 0.991, higher than that of between SARIMA (0.971) or BP neural network (0.942) SARIMA and the observation. Thus, it can be concluded that, the proposed SARIMA-BP neural network method can better adapt to the LAI time series, and it outperforms the SARIMA and BP neural network methods.
|
Received: 2015-11-25
Accepted: 2016-03-27
|
|
Corresponding Authors:
ZHANG Shu-qing
E-mail: zhangshuqing@iga.ac.cn
|
|
[1] Xiao Z Q, Liang S L, Wang J D, et al. Remote Sensing of Environment, 2011, 115(1): 97. [2] Jin H A, Wang J D, Xiao Z Q, et al. Geoscience and Remote Sensing Symposium (IGARSS), 2010 IEEE International. IEEE, 2010. [3] LI Xi-jia, XIAO Zhi-qiang, WANG Jin-di, et al(李喜佳,肖志强,王锦地,等). Journal of Remote Sensing(遥感学报), 2013,18(1): 27. [4] Savoy P, Mackay D S. Agricultural and Forest Meteorology, 2015, 200: 46. [5] Dickinson R E, Tian Y H, Liu Q, et al. Journal of Geophysical Research: Atmospheres (1984—2012), 2008, 113(D16). [6] Jiang B, Liang S L, Wang J D, et al. Remote Sensing of Environment, 2010, 114(7): 1432. [7] Gao F, Morisette J T, Wolfe R E, et al. IEEE Geoscience and Remote Sensing Letters, 2008, 5(1): 60. [8] Zhang G P. Neurocomputing, 2003, 50: 159. [9] CUI Ri-xian, LIU Ya-dong, FU Jin-dong(崔日鲜, 刘亚东, 付金东). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2015,35(9): 46. [10] ZHOU Chun-yan, HUA Deng-xin, LE Jing, et al(周春艳,华灯鑫,乐 静). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2015, 35(9): 53. [11] Box G E, Jenkins G M, Reinsel G C. Time Series Analysis: Forecasting and Control. John Wiley & Sons, 2011. [12] Cheng C, Cheng X S, Dai N, et al. Computers in Biology and Medicine, 2015, 66: 103. [13] Liu Y K, Xie F, Xie C L, et al. Annals of Nuclear Energy, 2015, 85: 566. [14] Fu L M. Neural Networks in Computer Intelligence. Tata: McGraw-Hill Education, 2003. [15] Khashei M, Bijari M. Applied Soft Computing, 2011, 11(2): 2664. [16] Faruk D . Engineering Applications of Artificial Intelligence, 2010, 23(4): 586. |
[1] |
XU Li-peng1, 2, GE Liang-quan1*, DENG Xiao-qin2, CHEN Li2, ZHAO Qiang2, LI Bin2, WANG Liang2. Research of Carborne γ-Ray Energe Spectrum Radiation Dose Rate Based on FFT-BP Network Model[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(02): 590-594. |
[2] |
Lü Meng1, HU Ying-tian1, GAO Ya1, WANG Xiao-ping2*. Novel Spectral COD Measurement Method Based on Identification of Water Samples[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(12): 3797-3802. |
[3] |
LI Yue-sheng1, LU Wei-ye1, ZHAO Jing-bo2, 3, FENG Guo-xing1, WEI Dong-ming2, LU Ji-dong2, 3, YAO Shun-chun2, 3*, LU Zhi-min2, 3. Detection of Caloric Value of Coal Using Laser-Induced Breakdown Spectroscopy Combined with BP Neural Networks[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(08): 2575-2579. |
[4] |
YANG Liu1,2, FENG Zhong-ke1*, YUE De-peng1, SUN Jin-hua3. Forest Stock Volume Estimation Model Using Textural and Topographic Factors of Landsat8 OLI[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(07): 2140-2145. |
[5] |
SONG Kun-lin, ZHANG Chu, PENG Ji-yu, YE Lan-han, LIU Fei*, HE Yong. Determination of Caffeine Content in Coffee Beans Based on Laser Induced Breakdown Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(07): 2199-2204. |
[6] |
WANG Fang1,2, ZHU Han1, LI Yun-peng1, LIU Yu-fang1, 2* . Combined Transmission Laser Spectrum of Core-Offset Fiber and BP Neural Network for Temperature Sensing Research[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(11): 3732-3736. |
[7] |
LI Jun-feng1, WANG Yue-le1, HU Sheng2, HE Hui-ling2* . The Comparison of Spectral Classification Based on DBN, BP Neural Network and SVM[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(10): 3261-3264. |
[8] |
CHEN Zheng-guang, LI Xin, FAN Xue-jia . Method for the Discrimination of the Variety of Potatoes with Vis/NIR Spectroscopy [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(08): 2474-2478. |
[9] |
YU Lei1, 2, HONG Yong-sheng1, 2, ZHOU Yong1, 2*, ZHU Qiang1, 2. Inversion of Soil Organic Matter Content Using Hyperspectral Data Based on Continuous Wavelet Transformation[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(05): 1428-1433. |
[10] |
WANG Shu-tao, CHEN Dong-ying*, WANG Xing-long, WEI Meng, WANG Zhi-fang . A New Method for the Determination of Potassium Sorbate Combining Fluorescence Spectra Method with PSO-BP Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(12): 3549-3554. |
[11] |
CUI Ri-xian1, LIU Ya-dong1, FU Jin-dong2* . Estimation of Winter Wheat Biomass Using Visible Spectral and BP Based Artificial Neural Networks [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(09): 2596-2601. |
[12] |
ZHOU Chun-yan1, 2, HUA Deng-xin1*, LE Jing1, WAN Wen-bo1, JIANG Peng1,MAO Jian-dong2. Research on Modeling Method for Chlorophyll Content Fine Measurement Based on Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(09): 2629-2633. |
[13] |
MA Xiao2, YUAN Hong-fu1*, SONG Chun-feng1, HU Ai-qin1, LI Xiao-yu1, ZHAO Zhong2, LI Xiu-qin3, GUO Zhen3, ZHU Zhi-qiang1. Research on Rapid Discrimination of Edible Oil by ATR Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(07): 1879-1884. |
[14] |
WANG Shu-tao, CHEN Dong-ying, HOU Pei-guo*, WANG Xing-long, WANG Zhi-fang, WEI Meng . Determination of the Sodium Methylparaben Content Based on Spectrum Fluorescence Spectral Technology and GA-BP Neural Network [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(06): 1606-1610. |
[15] |
TIAN Xi1, 2, HE Shao-lan2, 3, Lü Qiang2, 3, YI Shi-lai2, 3, XIE Rang-jin2, 3, ZHENG Yong-qiang2, 3, LIAO Qiu-hong1, 2, DENG Lie2, 3* . Determination of Photosynthetic Pigments in Citrus Leaves Based on Hyperspectral Images Datas[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2014, 34(09): 2506-2512. |
|
|
|
|