Evaluation of Vegetation Index-Based Curve Fitting Models for Accurate Classification of Salt Marsh Vegetation Using Sentinel-2 Time-Series

The successful launch of the Sentinel-2 constellation satellite, along with advanced cloud detection algorithms, has enabled the generation of continuous time series at high spatial and temporal resolutions, which is in turn expected to enable the classification of salt marsh vegetation over larger spatiotemporal scales. This study presents a critical comparison of vegetation index (VI) and curve fitting methods—two key factors for time series construction that potentially influence vegetation classification performance. To accomplish this objective, the stability of five different VI time series, namely Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), Enhanced Vegetation Index (EVI), Green Normalized Difference Vegetation Index (GNDVI), and Water-Adjusted Vegetation Index (WAVI), was compared empirically; the suitability between three curve fitting methods, namely Asymmetric Gaussian (AG), Double Logistic (DL), and Two-term Fourier (TF), and VI time series was measured using the coefficient of determination, and the salt marsh vegetation separability among different combinations of VI time series and curve fitting methods (i.e., VI time series-based curve fitting model) was quantified using overall the Jeffries–Matusita distance. Six common types of salt marsh vegetation from three typical coastal sites in China were used to validate these findings, which demonstrate: (1) the SAVI performed best in terms of time series stability, while the EVI exhibited relatively poor time series stability with conspicuous outliers induced by the sensitivity to omitted clouds and shadows; (2) the DL method commonly resulted in the most accurate classification of different salt marsh vegetation types, especially when combined with the EVI time series, followed by the TF method; and (3) the SAVI/NDVI-based DL/TF model demonstrated comparable efficiency for classifying salt marsh vegetation. Notably, the SAVI/NDVI-based DL model performed most strongly for high latitude regions with a continental climate, whilst the SAVI/NDVI-based TF model appears to be better suited to mid- to low latitude regions dominated by a monsoonal climate.

[1]  M. Kirwan,et al.  Tidal wetland stability in the face of human impacts and sea-level rise , 2013, Nature.

[2]  M. Mahdianpari,et al.  Random forest wetland classification using ALOS-2 L-band, RADARSAT-2 C-band, and TerraSAR-X imagery , 2017 .

[3]  Matthias Drusch,et al.  Sentinel-2: ESA's Optical High-Resolution Mission for GMES Operational Services , 2012 .

[4]  Cécile Gomez,et al.  Use of Sentinel-2 Time-Series Images for Classification and Uncertainty Analysis of Inherent Biophysical Property: Case of Soil Texture Mapping , 2019, Remote. Sens..

[5]  C. Woodcock,et al.  Continuous monitoring of forest disturbance using all available Landsat imagery , 2012 .

[6]  P. Villa,et al.  Assessing macrophyte seasonal dynamics using dense time series of medium resolution satellite data , 2018, bioRxiv.

[7]  Raymond F. Kokaly,et al.  Mapping changing distributions of dominant species in oil-contaminated salt marshes of Louisiana using imaging spectroscopy , 2016 .

[8]  A. Skidmore,et al.  Vegetation phenology from Sentinel-2 and field cameras for a Dutch barrier island , 2018, Remote Sensing of Environment.

[9]  Deren Li,et al.  A review of vegetation phenological metrics extraction using time-series, multispectral satellite data , 2020 .

[10]  J. Zedler,et al.  Wetland resources : Status, trends, ecosystem services, and restorability , 2005 .

[11]  Jennifer N. Hird,et al.  Noise reduction of NDVI time series: An empirical comparison of selected techniques , 2009 .

[12]  A. Gitelson,et al.  Quantitative estimation of chlorophyll-a using reflectance spectra : experiments with autumn chestnut and maple leaves , 1994 .

[13]  J. A. Schell,et al.  Monitoring vegetation systems in the great plains with ERTS , 1973 .

[14]  Venkat Lakshmi,et al.  Comparison of Normalized Difference Vegetation Index Derived from Landsat, MODIS, and AVHRR for the Mesopotamian Marshes Between 2002 and 2018 , 2019, Remote. Sens..

[15]  Bin Zhao,et al.  Spectral Discrimination of the Invasive Plant Spartina alterniflora at Multiple Phenological Stages in a Saltmarsh Wetland , 2013, PloS one.

[16]  Frank E. Müller-Karger,et al.  mproved coastal wetland mapping using very-high 2-meter spatial esolution imagery , 2015 .

[17]  Mariano Bresciani,et al.  Comparative Assessment of Broadband Vegetation Indices Over Aquatic Vegetation , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[18]  Kaishan Song,et al.  Mapping the distribution of mangrove species in the Core Zone of Mai Po Marshes Nature Reserve, Hong Kong, using hyperspectral data and high-resolution data , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[19]  Edoardo Pasolli,et al.  Harmonized Landsat 8 and Sentinel-2 Time Series Data to Detect Irrigated Areas: An Application in Southern Italy , 2020, Remote. Sens..

[20]  Alexis J. Comber,et al.  Random forest classification of salt marsh vegetation habitats using quad-polarimetric airborne SAR, elevation and optical RS data , 2014 .

[21]  P. Atkinson,et al.  Inter-comparison of four models for smoothing satellite sensor time-series data to estimate vegetation phenology , 2012 .

[22]  Chris J. Kennedy,et al.  The value of estuarine and coastal ecosystem services , 2011 .

[23]  A. Strahler,et al.  Monitoring vegetation phenology using MODIS , 2003 .

[25]  S. Sanjeevi,et al.  Jeffries Matusita based mixed-measure for improved spectral matching in hyperspectral image analysis , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[26]  Zhiqiang Yang,et al.  Continuous monitoring of land disturbance based on Landsat time series , 2020, Remote Sensing of Environment.

[27]  Yongxue Liu,et al.  Classification mapping and species identification of salt marshes based on a short-time interval NDVI time-series from HJ-1 optical imagery , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[28]  R. Nicholls,et al.  Future response of global coastal wetlands to sea-level rise , 2018, Nature.

[29]  J. Tukey.  Comparing individual means in the analysis of variance. , 1949, Biometrics.

[30]  Yongxue Liu,et al.  Classification mapping of salt marsh vegetation by flexible monthly NDVI time-series using Landsat imagery , 2018, Estuarine, Coastal and Shelf Science.

[31]  Claudia Giardino,et al.  A rule-based approach for mapping macrophyte communities using multi-temporal aquatic vegetation indices , 2015 .

[32]  A. Huete.  A soil-adjusted vegetation index (SAVI) , 1988 .

[33]  Marco Heurich,et al.  Spatially detailed retrievals of spring phenology from single-season high-resolution image time series , 2017, Int. J. Appl. Earth Obs. Geoinformation.

[34]  M. Kearney,et al.  The Effects of Tidal Inundation on the Reflectance Characteristics of Coastal Marsh Vegetation , 2009 .

[35]  A. Tamhane.  Multiple comparisons in model i one-way anova with unequal variances , 1977 .

[36]  Lei Wang,et al.  Mapping freshwater marsh species distributions using WorldView-2 high-resolution multispectral satellite imagery , 2014 .

[37]  B. Brisco,et al.  Spectral analysis of wetlands using multi-source optical satellite imagery , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[38]  Jennifer Pontius,et al.  Remote sensing of spring phenology in northeastern forests: A comparison of methods, field metrics and sources of uncertainty , 2014 .

[39]  Xiaoling Chen,et al.  Wetland changes of China's largest freshwater lake and their linkage with the Three Gorges Dam , 2018 .

[40]  Federico Filipponi,et al.  Exploitation of Sentinel-2 Time Series to Map Burned Areas at the National Level: A Case Study on the 2017 Italy Wildfires , 2019, Remote. Sens..

[41]  O. Iribarne,et al.  Distribution of saltmarsh plant communities associated with environmental factors along a latitudinal gradient on the south‐west Atlantic coast , 2006 .

[42]  José M. C. Pereira,et al.  Spectral discrimination of giant reed (Arundo donax L.): A seasonal study in riparian areas , 2013 .

[43]  C. Woodcock,et al.  Improvement and expansion of the Fmask algorithm: cloud, cloud shadow, and snow detection for Landsats 4–7, 8, and Sentinel 2 images , 2015 .

[44]  Chung-Hsin Chung,et al.  Forty years of ecological engineering with Spartina plantations in China , 2006 .

[45]  Per Jönsson,et al.  Seasonality extraction by function fitting to time-series of satellite sensor data , 2002, IEEE Trans. Geosci. Remote. Sens..

[46]  Gemma Bell,et al.  Using Continuous Change Detection and Classification of Landsat Data to Investigate Long-Term Mangrove Dynamics in the Sundarbans Region , 2019, Remote. Sens..

[47]  Sonia Silvestri,et al.  Tidal regime, salinity and salt marsh plant zonation , 2005 .

[48]  Mingguo Ma,et al.  Comparison of Eight Techniques for Reconstructing Multi-Satellite Sensor Time-Series NDVI Data Sets in the Heihe River Basin, China , 2014, Remote. Sens..

[49]  C. Hladik,et al.  Salt Marsh Elevation and Habitat Mapping Using Hyperspectral and LIDAR Data , 2013 .

[50]  P. Beck,et al.  Improved monitoring of vegetation dynamics at very high latitudes: A new method using MODIS NDVI , 2006 .

[51]  R. Pontius,et al.  Death to Kappa: birth of quantity disagreement and allocation disagreement for accuracy assessment , 2011 .

[52]  Li Wang,et al.  The Tidal Marsh Inundation Index (TMII): An inundation filter to flag flooded pixels and improve MODIS tidal marsh vegetation time-series analysis , 2017 .