研究目的
To assess methods to remove the glint effect in inland water systems with widely differing optical properties and investigate the reliability of glint removal methods for quality control of remote sensing data.
研究成果
The spectral glint correction method (M3) provides the most accurate Rrs datasets for most optical compositions, reducing errors by almost 30%, but in CDOM-dominated environments with low Chl-a, the variable ρ method (M2) performs better. No single ρ value is universally applicable; spectral methods are recommended for unstable platforms and complex waters.
研究不足
The study is limited to specific inland water reservoirs with varying optical properties; methods may not generalize to all aquatic systems. HydroLight simulations assume certain conditions (e.g., clear sky, infinite depth) that may not fully capture field variability. Spectral glint correction (M3) is sensitive to outliers in Chl-a concentrations and CDOM-rich environments.
1：Experimental Design and Method Selection:
The study evaluates four glint removal methods (M1: constant ρ=
2：028, M
variable ρ from LUT based on viewing geometry and wind speed, M3: spectral ρ using hyperspectral optimization, M4: ρ as function of wind speed) for correcting Rrs spectra. HydroLight simulations are used to generate reference Rrs data for comparison.
3：Sample Selection and Data Sources:
Field campaigns conducted in four reservoirs (Barra Bonita, Bariri, Ibitinga, Nova Avanhandava) of the Tietê River Cascade System in Brazil, with measurements of radiometric quantities (Lt, Lsky, Es, Ed) and water quality parameters (SPM, Chl-a, etc.).
4：List of Experimental Equipment and Materials:
TriOS Hyperspectral radiometers (RAMSES TriOS?), UV-2600 UV-VIS spectrophotometer (SHIMADZU), Whatman GF/F filters, portable anemometer, Garmin GPS.
5：Experimental Procedures and Operational Workflow:
In situ measurements taken with sensors at specific viewing geometry (θv=40°, φv=90°), water samples collected and analyzed for optical properties, HydroLight simulations performed with inputs of IOPs and environmental conditions.
6：Data Analysis Methods:
Statistical errors (RMSD, nRMSD, MAPE) computed to compare corrected Rrs with HydroLight reference; spectral convolution applied for OLI band comparisons.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容-
UV-VIS Spectrophotometer
UV-2600
SHIMADZU
Used for absorbance readings to derive absorption coefficients.
-
RAMSES TriOS
ARC
TriOS
Measures Lt and Lsky radiance with 7° field of view and 3.3 nm sampling interval.
-
RAMSES TriOS
ACC
TriOS
Measures downwelling irradiance (Es and Ed) with cosine collector.
-
HydroLight
5.2
Sequoia Scientific, Inc.
Radiative transfer numerical model for simulating Rrs and assessing glint correction methods.
-
Garmin GPS
Garmin
Used for geographic location data collection.
-
Portable Anemometer
Measures wind speed during field campaigns.
-
Whatman GF/F Filter
47 mm diameter, 0.7 μm pore size
Whatman
Used for filtering water samples to measure SPM, PIM, POM, and Chl-a concentrations.
-
Nylon Membrane Filter
47 mm diameter, 0.22 μm pore size
Used for filtering water samples to measure CDOM absorption.
-
Quartz Cuvette
10 cm optical path
Used for spectral absorbance readings of filtrate to derive CDOM absorption.
-
登录查看剩余7件设备及参数对照表
查看全部
