Unlocking the potential of remote sensing for kelp biomass estimation in South African Kelp Concession Areas

Searle, Lauren Jane

Unlocking the potential of remote sensing for kelp biomass estimation in South African Kelp Concession Areas

Thesis / Dissertation

2024

Publisher

University of Cape Town

Department

Department of Biological Sciences

Faculty

Faculty of Science

Abstract

The use of high-resolution imagery (HRI) has the potential to improve the accuracy of kelp biomass estimates, ensuring the implementation of sustainable harvesting. However, definitive research on the potential of HRI in this application is lacking in the current literature. An accurate estimation of kelp biomass is crucial to calculate maximum sustainable yield (MSY) in South African kelp Concessions. This study seeks to fill the knowledge gap by exploring the effectiveness of HRI for estimating the biomass of kelp along a specific stretch of coastline. The study aim is achieved by analysing HRI of Concession Area 6 taken from an aircraft. Maps quantifying kelp extent are derived from image classification methods applied to the HRI. A total biomass figure is then determined using the product of the calculated kelp extent and an average biomass figure of 14.5 kg/m-2 , taken from the literature. A total biomass of 40527.9 tonnes wet weight was calculated for Concession Area 6. The classification of HRI provided an overall accuracy of 95%, which is relatively high when compared to Sentinel-2 satellite imagery which resulted in an overall accuracy of 75%. When compared to the kelp extent measured in previous studies, HRI-derived maps had consistently less kelp coverage than maps from other imagery, suggesting that other imagery overestimates kelp extent (likely due to resolution). However, this was confounded given different imagery used at different times and so it was not possible to rule out change in kelp coverage over time. The results demonstrate the value of HRI in the mapping of kelp extent, which can ultimately be used to produce more accurate MSY assessments and support sustainable harvesting practices. However, before HRI can be integrated into MSY assessments, it is imperative to calculate more accurate biomass figures that are specific to the Concession Area, rather than relying on region wide estimates. Additionally, it's important to acknowledge that while HRI excels in precision, other imagery may be more suitable for large-scale estimates where accuracy is not a primary concern and due to its cost-effectiveness.