Elizabeth C. Atwood
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Elizabeth C. Atwood has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Evangelos Spyrakos
Earth and Planetary Observation Sciences (EPOS), Biological and Environmental Sciences, Faculty of Natural Sciences, University of Stirling, Stirling, UK
3 shared publications - Giulia Sent
MARE—Marine and Environmental Science Centre, ARNET—Aquatic Research Network, Faculdade de Ciências, Universidade de Lisboa, 1749-016 Lisbon, Portugal
3 shared publications Thomas Jackson
3 shared publications- Ana C. Brito
MARE—Marine and Environmental Science Centre, ARNET—Aquatic Research Network, Faculdade de Ciências, Universidade de Lisboa, Lisboa, Portugal
2 shared publications - Bror Jönsson
University of New Hampshire, USA
2 shared publications Steve Groom
2 shared publications
Latest Publications
- Open Access
- 14 July 2026
- Published by Elsevier in SSRN Electronic Journal
How Validation Design Affects the Comparison of Remote Sensing Algorithms
Satellite remote sensors provide frequent, global views of ocean color, whereas in-situ observations of ocean variables are spatiotemporally sparse. Oceanographers therefore use remote sensing algorithms to infer biogeochemical variables from satellite data. Many algorithms exist, but choosing among them is challenging because performance varies with the deployment scenario---for example, whether applied globally or regionally. Standard practice in machine learning when comparing models is to train them on the same training data and evaluate them on held-out test data. In this paper, we discuss two effects that can arise without a careful validation strategy: (1) when the design of the data split does not mimic the intended deployment scenario, algorithm rankings can be misleading, and (2) when spatiotemporally proximate in-situ observations fall on opposite sides of the split, performance estimates can be inflated---and this can advantage some models more than others. We illustrate these points by comparing ocean-color chlorophyll-a algorithms.
- 14 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessHow Validation Design Affects the Comparison of Remote Sensing Algorithms
Satellite remote sensors provide frequent, global views of ocean color, whereas in-situ observations of ocean variables are spatiotemporally sparse. Oceanographers therefore use remote sensing algorithms to infer biogeochemical variables from satellite data. Many algorithms exist, but choosing among them is challenging because performance varies with the deployment scenario---for example, whether applied globally or regionally. Standard practice in machine learning when comparing models is to train them on the same training data and evaluate them on held-out test data. In this paper, we discuss two effects that can arise without a careful validation strategy: (1) when the design of the data split does not mimic the intended deployment scenario, algorithm rankings can be misleading, and (2) when spatiotemporally proximate in-situ observations fall on opposite sides of the split, performance estimates can be inflated---and this can advantage some models more than others. We illustrate these points by comparing ocean-color chlorophyll-a algorithms.
- Open Access
- 28 September 2025
- Published by Elsevier in International Journal of Applied Earth Observation and Geoinformation
- Vol. 144
Can optical water types be used as ecological indicators? Insights from a temperate estuary
- 28 September 2025
- Published by Elsevier in International Journal of Applied Earth Observation and Geoinformation
- Vol. 144
Open AccessCan optical water types be used as ecological indicators? Insights from a temperate estuary
- Open Access
- 10 December 2024
- Published by Elsevier in Remote Sensing Applications: Society and Environment
- Vol. 37
What time is the tide? The importance of tides for ocean colour applications to estuaries
- 10 December 2024
- Published by Elsevier in Remote Sensing Applications: Society and Environment
- Vol. 37
Open AccessWhat time is the tide? The importance of tides for ocean colour applications to estuaries
- Open Access
- 3 September 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (17) , 3267
Framework for Regional to Global Extension of Optical Water Types for Remote Sensing of Optically Complex Transitional Water Bodies
Water quality indicator algorithms often separate marine and freshwater systems, introducing artificial boundaries and artifacts in the freshwater to ocean continuum. Building upon the Ocean Colour- (OC) and Lakes Climate Change Initiative (CCI) projects, we propose an improved tool to assess the interactions across river–sea transition zones. Fuzzy clustering methods are used to generate optical water types (OWT) representing spectrally distinct water reflectance classes, occurring within a given region and period (here 2016–2021), which are then utilized to assign membership values to every OWT class for each pixel and seamlessly blend optimal in-water algorithms across the region. This allows a more flexible representation of water provinces across transition zones than classic hard clustering techniques. Improvements deal with expanded sensor spectral band-sets, such as Sentinel-3 OLCI, and increased spatial resolution with Sentinel-2 MSI high-resolution data. Regional clustering was found to be necessary to capture site-specific characteristics, and a method was developed to compare and merge regional cluster sets into a pan-regional representative OWT set. Fuzzy clustering OWT timeseries data allow unique insights into optical regime changes within a lagoon, estuary, or delta system, and can be used as a basis to improve WQ algorithm performance.
- 3 September 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (17) , 3267
Open AccessFramework for Regional to Global Extension of Optical Water Types for Remote Sensing of Optically Complex Transitional Water Bodies
Water quality indicator algorithms often separate marine and freshwater systems, introducing artificial boundaries and artifacts in the freshwater to ocean continuum. Building upon the Ocean Colour- (OC) and Lakes Climate Change Initiative (CCI) projects, we propose an improved tool to assess the interactions across river–sea transition zones. Fuzzy clustering methods are used to generate optical water types (OWT) representing spectrally distinct water reflectance classes, occurring within a given region and period (here 2016–2021), which are then utilized to assign membership values to every OWT class for each pixel and seamlessly blend optimal in-water algorithms across the region. This allows a more flexible representation of water provinces across transition zones than classic hard clustering techniques. Improvements deal with expanded sensor spectral band-sets, such as Sentinel-3 OLCI, and increased spatial resolution with Sentinel-2 MSI high-resolution data. Regional clustering was found to be necessary to capture site-specific characteristics, and a method was developed to compare and merge regional cluster sets into a pan-regional representative OWT set. Fuzzy clustering OWT timeseries data allow unique insights into optical regime changes within a lagoon, estuary, or delta system, and can be used as a basis to improve WQ algorithm performance.