scholarly journals Hybrid Chlorophyll-a Algorithm for Assessing Trophic States of a Tropical Brazilian Reservoir Based on MSI/Sentinel-2 Data

2019 ◽  
Vol 12 (1) ◽  
pp. 40 ◽  
Author(s):  
Carolline Cairo ◽  
Claudio Barbosa ◽  
Felipe Lobo ◽  
Evlyn Novo ◽  
Felipe Carlos ◽  
...  
Keyword(s):  
Decision Tree ◽  
Chlorophyll A ◽  
Spatiotemporal Variability ◽  
Inland Waters ◽  
Decision Tree Classifier ◽  
A Algorithm ◽  
Spectral Angle Mapper ◽  
Chl A ◽  
Tree Classifier ◽  
Sentinel 2

Using remote sensing for monitoring trophic states of inland waters relies on the calibration of chlorophyll-a (chl-a) bio-optical algorithms. One of the main limiting factors of calibrating those algorithms is that they cannot accurately cope with the wide chl-a concentration ranges in optically complex waters subject to different trophic states. Thus, this study proposes an optical hybrid chl-a algorithm (OHA), which is a combined framework of algorithms for specific chl-a concentration ranges. The study area is Ibitinga Reservoir characterized by high spatiotemporal variability of chl-a concentrations (3–1000 mg/m3). We took the following steps to address this issue: (1) we defined optical classes of specific chl-a concentration ranges using Spectral Angle Mapper (SAM); (2) we calibrated/validated chl-a bio-optical algorithms for each trophic class using simulated Sentinel-2 MSI (Multispectral Instrument) bands; (3) and we applied a decision tree classifier in MSI/Sentinel-2 image to detect the optical classes and to switch to the suitable algorithm for the given class. The results showed that three optical classes represent different ranges of chl-a concentration: class 1 varies 2.89–22.83 mg/m3, class 2 varies 19.51–87.63 mg/m3, and class 3 varies 75.89–938.97 mg/m3. The best algorithms for trophic classes 1, 2, and 3 are the 3-band (R2 = 0.78; MAPE - Mean Absolute Percentage Error = 34.36%), slope (R2 = 0.93; MAPE = 23.35%), and 2-band (R2 = 0.98; MAPE = 20.12%), respectively. The decision tree classifier showed an accuracy of 95% for detecting SAM’s optical trophic classes. The overall performance of OHA was satisfactory (R2 = 0.98; MAPE = 26.33%) using in situ data but reduced in the Sentinel-2 image (R2 = 0.42; MAPE = 28.32%) due to the temporal gap between matchups and the variability in reservoir hydrodynamics. In summary, OHA proved to be a viable method for estimating chl-a concentration in Ibitinga Reservoir and the extension of this framework allowed a more precise chl-a estimate in eutrophic inland waters.

scholarly journals Regional Models for High-Resolution Retrieval of Chlorophyll a and TSM Concentrations in the Gorky Reservoir by Sentinel-2 Imagery

2019 ◽  
Vol 11 (10) ◽  
pp. 1215 ◽  
Author(s):  
Alexander A. Molkov ◽  
Sergei V. Fedorov ◽  
Vadim V. Pelevin ◽  
Elena N. Korchemkina
Keyword(s):  
Optical Properties ◽  
Chlorophyll A ◽  
Russian Federation ◽  
Freshwater Ecosystems ◽  
In Situ Measurements ◽  
Spatiotemporal Variability ◽  
Chl A ◽  
The Russian Federation ◽  
Sentinel 2

The possibilities of chlorophyll a (Chl a) and total suspended matter (TSM) retrieval using Sentinel-2/MSI imagery and in situ measurements in the Gorky Reservoir are investigated. This water body is an inland freshwater ecosystem within the territory of the Russian Federation. During the algal bloom period, the optical properties of water are extremely heterogeneous and vary on scales of tens of meters. Additionally, they vary in time under the influence of currents and wind forcing. In this case, the usage of the traditional station-based sampling to describe the state of the reservoir may be uninformative and not rational. Therefore, we proposed an original approach based on simultaneous in situ measurements of the remote sensing reflectance by a single radiometer and the concentration of water constituents by an ultraviolet fluorescence LiDAR from a high-speed gliding motorboat. This approach provided fast data collection including 4087 synchronized LiDAR and radiometric measurements with high spatial resolutions of 8 m for two hours. A part of the dataset was coincided with Sentinel-2 overpass and used for the development of regional algorithms for the retrieval of Chl a and TSM concentrations. For inland waters of the Russian Federation, such research was performed for the first time. The proposed algorithms can be used for regular environmental monitoring of the Gorky Reservoir using ship measurements or Sentinel-2 images. Additionally, they can be adapted for neighboring reservoirs, for example, for other seven reservoirs on the Volga River. Moreover, the proposed ship measurement approach can be useful in the practice of limnological monitoring of inland freshwater ecosystems with high spatiotemporal variability of the optical properties.

scholarly journals Spatiotemporal Variability of Surface Phytoplankton Carbon and Carbon-to-Chlorophyll a Ratio in the South China Sea Based on Satellite Data

2020 ◽  
Vol 13 (1) ◽  
pp. 30
Author(s):  
Wenlong Xu ◽  
Guifen Wang ◽  
Long Jiang ◽  
Xuhua Cheng ◽  
Wen Zhou ◽  
...  
Keyword(s):  
South China Sea ◽  
Chlorophyll A ◽  
South China ◽  
Satellite Data ◽  
Phytoplankton Biomass ◽  
The South China Sea ◽  
Spatiotemporal Variability ◽  
The South ◽  
Chl A ◽  
China Sea

The spatiotemporal variability of phytoplankton biomass has been widely studied because of its importance in biogeochemical cycles. Chlorophyll a (Chl-a)—an essential pigment present in photoautotrophic organisms—is widely used as an indicator for oceanic phytoplankton biomass because it could be easily measured with calibrated optical sensors. However, the intracellular Chl-a content varies with light, nutrient levels, and temperature and could misrepresent phytoplankton biomass. In this study, we estimated the concentration of phytoplankton carbon—a more suitable indicator for phytoplankton biomass—using a regionally adjusted bio-optical algorithm with satellite data in the South China Sea (SCS). Phytoplankton carbon and the carbon-to-Chl-a ratio (θ) exhibited considerable variability spatially and seasonally. Generally, phytoplankton carbon in the northern SCS was higher than that in the western and central parts. The regional monthly mean phytoplankton carbon in the northern SCS showed a prominent peak during December and January. A similar pattern was shown in the central part of SCS, but its peak was weaker. Besides the winter peak, the western part of SCS had a secondary maximum of phytoplankton carbon during summer. θ exhibited significant seasonal variability in the northern SCS, but a relatively weak seasonal change in the western and central parts. θ had a peak in September and a trough in January in the northern and central parts of SCS, whereas in the western SCS the minimum and maximum θ was found in August and during October–April of the following year, respectively. Overall, θ ranged from 26.06 to 123.99 in the SCS, which implies that the carbon content could vary up to four times given a specific Chl-a value. The variations in θ were found to be related to changing phytoplankton community composition, as well as dynamic phytoplankton physiological activities in response to environmental influences; which also exhibit much spatial differences in the SCS. Our results imply that the spatiotemporal variability of θ should be considered, rather than simply used a single value when converting Chl-a to phytoplankton carbon biomass in the SCS, especially, when verifying the simulation results of biogeochemical models.

scholarly journals PERFORMANCE ANALYSIS OF BREAST CANCER CLASSIFICATION USING DECISION TREE CLASSIFIERS

2017 ◽  
Vol 9 (2) ◽  
pp. 19 ◽  
Author(s):  
P. Hamsagayathri ◽  
P. Sampath
Keyword(s):  
Breast Cancer ◽  
Decision Tree ◽  
Ductal Carcinoma ◽  
Research Work ◽  
The United States ◽  
Breast Cancer Dataset ◽  
Decision Tree Classifier ◽  
Cancer Dataset ◽  
Term Survival ◽  
Tree Classifier

Breast cancer is one of the dangerous cancers among world’s women above 35 y. The breast is made up of lobules that secrete milk and thin milk ducts to carry milk from lobules to the nipple. Breast cancer mostly occurs either in lobules or in milk ducts. The most common type of breast cancer is ductal carcinoma where it starts from ducts and spreads across the lobules and surrounding tissues. According to the medical survey, each year there are about 125.0 per 100,000 new cases of breast cancer are diagnosed and 21.5 per 100,000 women due to this disease in the United States. Also, 246,660 new cases of women with cancer are estimated for the year 2016. Early diagnosis of breast cancer is a key factor for long-term survival of cancer patients. Classification plays an important role in breast cancer detection and used by researchers to analyse and classify the medical data. In this research work, priority-based decision tree classifier algorithm has been implemented for Wisconsin Breast cancer dataset. This paper analyzes the different decision tree classifier algorithms for Wisconsin original, diagnostic and prognostic dataset using WEKA software. The performance of the classifiers are evaluated against the parameters like accuracy, Kappa statistic, Entropy, RMSE, TP Rate, FP Rate, Precision, Recall, F-Measure, ROC, Specificity, Sensitivity.

Sign in / Sign up

Export Citation Format

Share Document