Précédent|Prochain

A method for reconstructing long-term daily resolution EVIs based on MODIS daily surface reflectance products

  • role： First author第一作者
  • Affiliation：

    International Institute for Earth System Science, Nanjing University, Nanjing 210023, China

    Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology, Nanjing University, Nanjing 210023, China

  • Email：ningw@smail.nju.edu.cn
  • Introduction：王宁，研究方向为多光谱遥感。E-mail： ningw@smail.nju.edu.cn
WANG Ning12,  
  • role： Corresponding author通信作者
  • Affiliation：

    School of instrumentation and optoelectronic engineering, Beihang University, Beijing 100191, China

    International Institute for Earth System Science, Nanjing University, Nanjing 210023, China

  • Email：tianjia@buaa.edu.cn
  • Introduction：田家，研究方向为高光谱遥感理论方法及其定量应用。E-mail： tianjia@buaa.edu.cn
TIAN Jia31*,  
  • Affiliation：

    International Institute for Earth System Science, Nanjing University, Nanjing 210023, China

    Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology, Nanjing University, Nanjing 210023, China

TIAN Qingjiu12

résumé

The Enhanced Vegetation Index (EVI) combines factors such as atmospheric, soil, and saturation conditions and effectively correlates these data with vegetation biomass, leaf area index, and photosynthetically active radiation. Although the performance of the EVI is better than that of the Normalized Difference Vegetation Index (NDVI), the low temporal resolution of EVI products and the presence of cloud cover often result in a large number of missing pixels. In this study, we propose a daily resolution EVI reconstruction method that combines the Maximum Value Composite (MVC) and harmonic analysis of time series (HANTS) algorithms based on MODIS daily surface reflectance products.Given the spectral response differences of varying sensors carried by different satellites, the comparability of the EVIs calculated based on the Terra and Aqua satellites was analyzed prior to conducting the MVC operation. The analysis revealed a strong spatial linear correlation between the two variables, with R2 and RMSE values ranging from 0.9796-0.9935 and 0.0116-0.0297, respectively. The annual mean R2 and RMSE values were 0.9883 and 0.0196, respectively. The fitted parameters a and b had value ranges of 0.9447 to 1.0420 and -0.0065 to -0.0072, respectively, with annual mean values of 0.9910 and 0.0012. Despite spectral differences, the calculated EVIs based on the two satellite datasets exhibit minimal differences and thus are suitable for further processing via the MVC algorithm.This method was applied to reconstruct daily resolution EVI time series data for the North China Plain in 2021. The proposed EVI reconstruction algorithm is effective for large-scale and long-term reconstructions of daily resolution EVI time series data. The reconstructed EVI yields a rich texture, fills in the missing pixels, removes noise from the original EVI data, and follows the changing patterns of various land cover types. The HANTS method offers three advantages over the S-G filtering algorithm. First, compared with the original EVI, the HANTS method better preserved the spatial distribution patterns of the original EVI during reconstruction; by contrast, the S-G algorithm exhibited larger changes in spatial distribution in the reconstructed EVI. Second, the EVI curves reconstructed using the HANTS algorithm are smoother with minimal noise for typical land cover types; by contrast, the EVI curves reconstructed using the S-G algorithm have more local noise and nondifferentiable points, which hinders the extraction of vegetation phenological characteristics. Third, in terms of fidelity evaluation against high-quality reference EVI pixels, the HANTS algorithm demonstrated a strong linear correlation with the reference EVI pixels. The R2 and RMSE values ranged from 0.91 to 0.97 and from 0.017 to 0.032 across the months, with the strongest and weakest correlations occurring in September and June, respectively. By contrast, the S-G algorithm showed a weaker linear correlation with the reference EVI pixels. The R2 and RMSE values ranged from 0.38 to 0.91 and from 0.055 to 0.206 across the months, with the strongest and weakest correlations occurring in May and August, respectively. Overall, the HANTS method consistently outperformed the S-G method in terms of fidelity, with higher R2 values and lower RMSE values across all months. The proposed daily resolution EVI reconstruction method offers new guidelines and technical approaches for generating high-temporal resolution EVI data.

mots-clés

MODIS;vegetation index;EVI;MVC;HANTS;daily resolution;North China Plain

References

  1. 1.
    Bai Y. 2021. Analysis of vegetation dynamics in the Qinling-Daba Mountains region from MODIS time series data. Ecological Indicators, 129: 108029
  2. 2.
    Beck P S A, Atzberger C, Høgda K A, Johansen B and Skidmore A K. 2006. Improved monitoring of vegetation dynamics at very high latitudes: a new method using MODIS NDVI. Remote Sensing of Environment, 100(3): 321-334
  3. 3.
    Chen J, Jönsson P, Tamura M, Gu Z H, Matsushita B and Eklundh L. 2004. A simple method for reconstructing a high-quality NDVI time-series data set based on the Savitzky–Golay filter. Remote sensing of Environment, 91(3/4): 332-344
  4. 4.
    Chen S S, Chen X Z, Chen W Q, Su Y X and Li D. 2011. A simple retrieval method of land surface temperature from AMSR-E passive microwave data —A case study over Southern China during the strong snow disaster of 2008. International Journal of Applied Earth Observation and Geoinformation, 13(1): 140-151
  5. 5.
    Chen Y F, Zhou Y M and Zhao X Y. 2020. PM2.5 over North China based on MODIS AOD and effect of meteorological elements during 2003-2015. Frontiers of Environmental Science and Engineering, 14(2): 23
  6. 6.
    Holben B N. 1986. Characteristics of maximum-value composite images from temporal AVHRR data. International Journal of Remote Sensing, 7(11): 1417-1434
  7. 7.
    Jonsson P and Eklundh L. 2002. Seasonality extraction by function fitting to time-series of satellite sensor data. IEEE Transactions on Geoscience and Remote Sensing, 40(8): 1824-1832
  8. 8.
    Kong D D, Zhang Y Q, Gu X H and Wang D G. 2019. A robust method for reconstructing global MODIS EVI time series on the Google Earth Engine. ISPRS Journal of Photogrammetry and Remote Sensing, 155: 13-24
  9. 9.
    Li Y H, Wang J X and Li Y. 2016. Decomposition of MERSI multispectral mixed pixels by EVI time series. Journal of Remote Sensing (in Chinese), 20(3): 459-467
  10. 10.
    Lu L L, Wang C Z, Guo H D and Li Q T. 2014. Detecting winter wheat phenology with SPOT-VEGETATION data in the North China Plain. Geocarto International, 29(3): 244-255
  11. 11.
    Motohka T, Nasahara K N, Murakami K and Nagai S. 2011. Evaluation of sub-pixel cloud noises on MODIS daily spectral indices based on in situ measurements. Remote Sensing, 3(8): 1644-1662
  12. 12.
    Niu F, Li C C and Mao J T. 2006. A synergic application of TERRA and AQUA on retrieving aerosol optical depth over land. Journal of Remote Sensing (in Chinese), 10(4): 456-462
  13. 13.
    Pan Y Z, Li L, Zhang J S, Liang S L and Hou D. 2011. Crop area estimation based on MODIS-EVI time series according to distinct characteristics of key phenology phases: a case study of winter wheat area estimation in small-scale area. Journal of Remote Sensing (in Chinese), 15(3): 578-594
  14. 14.
    Pepin N, Deng H J, Zhang H B, Zhang F, Kang S C and Yao T D. 2019. An examination of temperature trends at high elevations across the Tibetan Plateau: the use of MODIS LST to understand patterns of elevation-dependent warming. Journal of Geophysical Research: Atmospheres, 124(11): 5738-5756
  15. 15.
    Ran Y L, Chen Y P, Chen F, Zhang H L and Jia X B. 2021. May-July NDVI variation for the middle Qinling Mountains over the past 194 years indicated by tree rings of Pinus tabuliformis. Chinese Journal of Applied Ecology, 32(10): 3661-3670
  16. 16.
    Rittger K, Raleigh M S, Dozier J, Hill A F, Lutz J A and Painter T H. 2020. Canopy adjustment and improved cloud detection for remotely sensed snow cover mapping. Water Resources Research, 56(6): e2019WR024914
  17. 17.
    Roerink G J, Menenti M and Verhoef W. 2000. Reconstructing cloudfree NDVI composites using Fourier analysis of time series. International Journal of Remote Sensing, 21(9): 1911-1917
  18. 18.
    Savitzky A and Golay M J E. 1964. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry, 36(8): 1627-1639
  19. 19.
    Son N T, Chen C F, Chen C R, Minh V Q and Trung N H. 2014. A comparative analysis of multitemporal MODIS EVI and NDVI data for large-scale rice yield estimation. Agricultural and Forest Meteorology, 197: 52-64
  20. 20.
    Stillinger T, Roberts D A, Collar N M and Dozier J. 2019. Cloud masking for Landsat 8 and MODIS Terra over snow-covered terrain: error analysis and spectral similarity between snow and cloud. Water Resources Research, 55(7): 6169-6184
  21. 21.
    Tao J B, Mishra D R, Cotten D L, O’Connell J, Leclerc M, Nahrawi H B, Zhang G S and Pahari R. 2018. A comparison between the MODIS Product (MOD17A2) and a tide-robust empirical GPP model evaluated in a georgia wetland. Remote Sensing, 10(11): 1831
  22. 22.
    Tarpley J D, Schneider S R and Money R L. 1984. Global vegetation indices from the NOAA-7 meteorological satellite. Journal of Climate and Applied Meteorology, 23(3): 491-494
  23. 23.
    Vijith H and Dodge-Wan D. 2020. Applicability of MODIS land cover and Enhanced Vegetation Index (EVI) for the assessment of spatial and temporal changes in strength of vegetation in tropical rainforest region of Borneo. Remote Sensing Applications: Society and Environment, 18: 100311
  24. 24.
    Wang D D, Chen Y H, Hu L Q, Voogt J A, Gastellu-Etchegorry J P and Krayenhoff E S. 2021. Modeling the angular effect of MODIS LST in urban areas: a case study of Toulouse, France. Remote Sensing of Environment, 257: 112361
  25. 25.
    Wang X, Li X B, Tan M H and Xin L J. 2015. Remote sensing monitoring of changes in winter wheat area in North China Plain from 2001 to 2011. Transactions of the Chinese Society of Agricultural Engineering, 31(8): 190-199
  26. 26.
    Wang X Y, Han C, Ouyang Z Q, Chen S Y, Guo H, Wang J and Hao X H. 2022. Cloud–snow confusion with MODIS snow products in boreal forest regions. Remote Sensing, 14(6): 1372
  27. 27.
    Wardlow B D, Egbert S L and Kastens J H. 2007. Analysis of time-series MODIS 250 m vegetation index data for crop classification in the U.S. Central Great Plains. Remote Sensing of Environment, 108(3): 290-310
  28. 28.
    Wilson A M, Parmentier B and Jetz W. 2014. Systematic land cover bias in Collection 5 MODIS cloud mask and derived products —A global overview. Remote Sensing of Environment, 141: 149-154
  29. 29.
    Xie F and Fan H. 2021. Deriving drought indices from MODIS vegetation indices (NDVI/EVI) and Land Surface Temperature (LST): is data reconstruction necessary?. International Journal of Applied Earth Observation and Geoinformation, 101: 102352
  30. 30.
    Xin J F, Yu Z R and Driessen P M. 2001. Monitoring phenological key stages of winter wheat with NOAA NDVI data. Journal of Remote Sensing, 5(6): 442-447
  31. 31.
    Yang Y J, Tao B, Ren W, Zourarakis D P, Masri B E, Sun Z G and Tian Q J. 2019. An improved approach considering intraclass variability for mapping winter wheat using multitemporal MODIS EVI images. Remote Sensing, 11(10): 1191
  32. 32.
    Zhang H and Ren Z Y. 2014. Comparison and application analysis of several NDVI time-series reconstruction methods. Scientia Agricultura Sinica, 47(15): 2998-3008
  33. 33.
    Zhang X, Li R, Yue Y M, Liu B and Liu H X. 2010. Improved algorithm for reconstructing vegetation index image time series based on Fourier Harmonic Analysis. Journal of Remote Sensing (in Chinese), 14(3): 437-447
  34. 34.
    Zhu H, Hu Y, Sun F, Wang Q and Ma X Y. 2022. Evaluation of reconstruction effect of NDVI time series based on QA weight: a case study of MODIS NDVI in the Yangtze River Basin. Scientia Geographica Sinica, 42(11): 2019-2027

Lire l'article complet

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website