Improving the accuracies of bathymetric models based on  multiple regression for calibration (case study: Sarca River, Italy)

Milad Niroumand-Jadidi; Alfonso Vitti

doi:10.1117/12.2242083

19 October 2016 Improving the accuracies of bathymetric models based on multiple regression for calibration (case study: Sarca River, Italy)

Milad Niroumand-Jadidi, Alfonso Vitti

Author Affiliations +

Proceedings Volume 9999, Remote Sensing of the Ocean, Sea Ice, Coastal Waters, and Large Water Regions 2016; 99990Q (2016) https://doi.org/10.1117/12.2242083
Event: SPIE Remote Sensing, 2016, Edinburgh, United Kingdom

Abstract

The optical imagery has the potential for extraction of spatially and temporally explicit bathymetric information in inland/coastal waters. Lyzenga’s model and optimal band ratio analysis (OBRA) are main bathymetric models which both provide linear relations with water depths. The former model is sensitive and the latter is quite robust to substrate variability. The simple regression is the widely used approach for calibration of bathymetric models either Lyzenga’s model or OBRA model. In this research, a multiple regression is examined for empirical calibration of the models in order to take the advantage of all spectral channels of the imagery. This method is applied on both Lyzenga’s model and OBRA model for the bathymetry of a shallow Alpine river in Italy, using WorldView-2 (WV-2) and GeoEye images. Insitu depths are recorded using RTK GPS in two reaches. One-half of the data is used for calibration of models and the remaining half as independent check-points for accuracy assessment. In addition, radiative transfer model is used to simulate a set of spectra in a range of depths, substrate types, and water column properties. The simulated spectra are convolved to the sensors’ spectral bands for further bathymetric analysis. Investigating the simulated spectra, it is concluded that the multiple regression improves the robustness of the Lyzenga’s model with respect to the substrate variability. The improvements of multiple regression approach are much more pronounced for the Lyzenga’s model rather than the OBRA model. This is in line with findings from real imagery; for instance, the multiple regression applied for calibration of Lyzenga’s and OBRA models demonstrated, respectively, 22% and 9% higher determination coefficients (R²) as well as 3 cm and 1 cm better RMSEs compared to the simple regression using the WV-2 image.

Conference Presentation

Citation Download Citation

Milad Niroumand-Jadidi and Alfonso Vitti "Improving the accuracies of bathymetric models based on multiple regression for calibration (case study: Sarca River, Italy)", Proc. SPIE 9999, Remote Sensing of the Ocean, Sea Ice, Coastal Waters, and Large Water Regions 2016, 99990Q (19 October 2016); https://doi.org/10.1117/12.2242083

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available