ANN BASED MODELLING FOR ESTIMATION OF DAILY SEA SURFACE TEMPERATURE OVER ARABIAN SEA USING MODIS DATA

Hits: 21
Research areas:
Year:
2017
Type of Publication:
Article
Keywords:
MODIS, SST,ANN
Authors:
Swathy Sunder, Balaji Ramakrishnan Raaj Ramsankaran
Abstract:
In this study, an Artificial Neural Network (ANN) based modelling has been doneto estimate daily Sea Surface Temperature (SST) usingModerate Resolution Imaging Spectroradiometer (MODIS) Aqua datasets. A feed-forward Back-Propagation Artificial Neural Network (ANN-BP) model was implemented in this work. This ANN based SST model was then trained and tested over the Arabian Sea for 10 years using MODIS data and in-situ reference data collected by CERSAT from Coriolis data centre. The application of the tested model requires only the reflectance of bands 31 and 32 at 11 and 12 μm as inputs. The obtained results were compared and analysed with the standard MODIS SST product as well as the reference data. Preliminary analysis of the obtained results show that the proposed ANN based technique is in good correlation with the above two datasets. The obtained results suggests that the machine learning techniques such as ANN has good potential and should be explored in detail for estimation of SST at both near shore and offshore waters.
Full text: 188.pdf [Bibtex]
You are here: Home ACRS ACRS Overview Proceedings ann-based-modelling-for-estimation-of-daily-sea-surface-temperature-over-arabian-sea-using-modis-data