Researching

Journals
  • Advanced Photonics
  • Photonics Insights
  • Advanced Photonics Nexus
  • Photonics Research
  • Advanced Imaging
  • View All Journals
  • Chinese Optics Letters
  • High Power Laser Science and Engineering
    • Articles
  • Optics
  • Physics
  • Geography
  • View All Subjects
    • Conferences
  • CIOP
  • HPLSE
  • AP
  • View All Events
    • News
    About CLP
    Search by keywords or author
    Journals >Acta Optica Sinica >Volume 42 >Issue 18 >Page 1830002 > Article
    • Acta Optica Sinica
    • Vol. 42, Issue 18, 1830002 (2022)
    XGBoost-Based Inversion of Phytoplankton Pigment Concentrations from Field Measured Fluorescence Excitation Spectra
    Linqi Wang1, Shengqiang Wang1,2,*, Deyong Sun1, Junsheng Li2..., Yuanli Zhu3, Yongjiu Xu4 and Hailong Zhang1|Show fewer author(s)
    Author Affiliations
  • 1School of Marine Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, Jiangsu, China
  • 2State Key Laboratory of Remote Sensing Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China
  • 3Second Institute of Oceanography, MNR, Hangzhou 310012, Zhejiang, China
  • 4School of Fishery, Zhejiang Ocean University, Zhoushan 316022, Zhejiang, China
    • show less
    DOI: 10.3788/AOS202242.1830002 Cite this Article Set citation alerts
    Linqi Wang, Shengqiang Wang, Deyong Sun, Junsheng Li, Yuanli Zhu, Yongjiu Xu, Hailong Zhang. XGBoost-Based Inversion of Phytoplankton Pigment Concentrations from Field Measured Fluorescence Excitation Spectra[J]. Acta Optica Sinica, 2022, 42(18): 1830002 Copy Citation Text show less
    Distribution histograms of eight pigment concentrations. (a) Perid; (b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax;(g) Chlb; (h) Tchla
    Fig. 1. Distribution histograms of eight pigment concentrations. (a) Perid; (b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax;(g) Chlb; (h) Tchla
    Download full size
    Excitation fluorescence spectrum curves measured in field
    Fig. 2. Excitation fluorescence spectrum curves measured in field
    Download full size
    Training performances of pigment concentration inversion models based on XGBoost machine learning algorithm. (a) Perid;(b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Fig. 3. Training performances of pigment concentration inversion models based on XGBoost machine learning algorithm. (a) Perid;(b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Download full size
    Validation performances of pigment concentration inversion models based on XGBoost machine learning algorithm. (a) Perid; (b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h)Tchla
    Fig. 4. Validation performances of pigment concentration inversion models based on XGBoost machine learning algorithm. (a) Perid; (b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h)Tchla
    Download full size
    Profile distributions of eight pigment concentrations in 32.8°N section estimated from fluorescence excitation spectra. (a) Perid;(b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Fig. 5. Profile distributions of eight pigment concentrations in 32.8°N section estimated from fluorescence excitation spectra. (a) Perid;(b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Download full size
    Training performances of pigment concentration inversion models based on least square regression method. (a) Perid;(b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Fig. 6. Training performances of pigment concentration inversion models based on least square regression method. (a) Perid;(b) 19Butfu; (c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Download full size
    Validation performances of concentration inversion models based on least square regression method. (a) Perid; (b) 19Butfu;(c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Fig. 7. Validation performances of concentration inversion models based on least square regression method. (a) Perid; (b) 19Butfu;(c) Fucox; (d) 19Hexfu; (e) Allox; (f) Zeax; (g) Chlb; (h) Tchla
    Download full size
    Comparison of accuracies of pigment concentration inversion models based on XGBoost machine learning algorithm and least square regression method. (a) Model training; (b) model validation
    Fig. 8. Comparison of accuracies of pigment concentration inversion models based on XGBoost machine learning algorithm and least square regression method. (a) Model training; (b) model validation
    Download full size
    SymbolPigment
    TchlaTotal chlorophyll a
    ChlbChlorophyll b
    FucoxFucoxanthin
    PeridPeridinin
    19Hexfu19′-Hexanoyloxyfucoxanthin
    19Butfu19′-Butanoyloxyfucoxanthin
    AlloxAlloxanthin
    ZeaxZeaxanthin
    Table 1. English names (symbols and full names) of phytoplankton pigments involved in this paper
    Spectral indicatorExponential form
    X1lgF(λ1)+lgF(λ2)lgF(λ1)/lgF(λ2)
    X2lgF(λ1)-lgF(λ2)lgF(λ1)/lgF(λ2)
    X3lgF(λ1)-lgF(λ2)lgF(λ1)+lgF(λ2)
    X4lgF(λ)
    X5lgF(λ1)lgF(λ2)
    X6lgF(λ1)F(λ2)
    X7lgF(λ1)+F(λ2)
    Table 2. Exponential forms of seven excitation fluorescence spectral indicators
    PigmentMinimum valueMaximum valueAverage value
    Tchla0.0786.7301.354
    Fucox02.0000.221
    Perid00.6160.048
    19Hexfu01.0380.125
    19Butfu00.3630.040
    Allox00.5250.032
    Chlb01.3250.163
    Zeax0.0030.8130.120
    Table 3. Statistics of pigment concentration measured by HPLC
    PigmentOptimal indictor of fluorescence excitation spectrumTraining datasetValidation dataset
    R2

    RMSE /

    （mg·m-3）

    		MAPE /%R2RMSE /（mg·m-3）MAPE /%
    PeridX30.840.03639.20.770.11049.9
    19ButfuX60.940.02524.10.670.02450.6
    FucoxX60.960.08325.70.870.38246.9
    19HexfuX60.780.10339.20.680.12535.8
    AlloxX50.960.03026.50.860.03738.2
    ZeaxX60.850.06434.50.860.13547.2
    ChlbX50.800.17141.10.590.24164.2
    TchlaX60.980.2107.50.871.16828.1
    Table 4. Optimal indictor forms of fluorescence excitation spectra and performances inverted by eight pigment concentrations
    PigmentOptimal indictor of fluorescence excitation spectrumBest band combination /nmTraining datasetValidation dataset
    R2RMSE /（mg·m-3）MAPE /%R2RMSE /（mg·m-3）MAPE /%
    PeridX1λ1=570, λ2=5050.540.06372.10.540.15158.7
    19ButfuX6λ1=505, λ2=5900.550.05395.40.560.02169.7
    FucoxX1λ1=375, λ2=4000.740.16294.50.870.77068.6
    19HexfuX6λ1=375, λ2=4350.430.14262.60.070.17257.3
    AlloxX6λ1=435, λ2=5050.510.091126.90.360.057120.2
    ZeaxX6λ1=435, λ2=5050.420.11980.00.460.186124.5
    ChlbX6λ1=505, λ2=5900.580.21171.00.380.26892.4
    TchlaX7λ1=420, λ2=5050.760.78643.40.750.89345.1
    Table 5. Optimal indictor forms of fluorescence excitation spectra, best band combinations and performances inverted by eight pigment concentration based on least square regression method
    Abstract
    Get PDF(in Chinese)
    Figures&Tables (13)
    Equations (0)
    References (28)
    Cited By (2)
    Get Citation
    Copy Citation Text
    Linqi Wang, Shengqiang Wang, Deyong Sun, Junsheng Li, Yuanli Zhu, Yongjiu Xu, Hailong Zhang. XGBoost-Based Inversion of Phytoplankton Pigment Concentrations from Field Measured Fluorescence Excitation Spectra[J]. Acta Optica Sinica, 2022, 42(18): 1830002
    Download Citation
    Set citation alerts for the article
    Tools
    Share
    Set citation alerts for the article
    Save the article for my favorites
    Paper Information
    Recommended Topics
    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology