Performance Evaluation of Newly Proposed Seaweed Enhancing Index (SEI)

Muhammad Danish Siddiqui,Muhammad Abdullah,Arjumand Z Zaidi

doi:10.3390/rs11121434

Abstract

Seaweed is a valuable coastal resource for its use in food, cosmetics, and other items. This study proposed new remote sensing based seaweed enhancing index (SEI) using spectral bands of near-infrared (NIR) and shortwave-infrared (SWIR) of Landsat 8 satellite data. Nine Landsat 8 satellite images of years 2014, 2016, and 2018 for the January, February, and March months were utilized to test the performance of SEI. The seaweed patches in the coastal waters of Karachi, Pakistan were mapped using the SEI, normalized difference vegetation index (NDVI), and floating algae index (FAI). Seaweed locations recorded during a field survey on February 26, 2014, were used to determine threshold values for all three indices. The accuracy of SEI was compared with NDVI while placing FAI as the reference index. The accuracy of NDVI and SEI were assessed by matching their spatial extent of seaweed cover with FAI enhanced seaweed area. SEI images of January 2016, February 2018, and March 2018 enhanced less than 50 percent of the corresponding FAI total seaweed areas. However, on these dates the NDVI performed very well, matching more than 95 percent of FAI seaweed coverage. Except for these three times, the performance of SEI in the remaining six images was either similar to NDVI or even better than NDVI. SEI enhanced 99 percent of FAI seaweed cover on January 2018 image. Overall, seaweed area not covered by FAI was greater in SEI than NDVI in almost all images, which needs to be further explored in future studies by collecting extensive field information to validate SEI mapped additional area beyond the extent of FAI seaweed cover. Based on these results, in the majority of the satellite temporal images selected for this study, the performance of the newly proposed index—SEI, was found either better than or similar to NDVI.

Highlights

Seaweed is the name given to the numerous marine plants and algae that animate in seas, oceans, rivers, lakes, and other water forms
The results showed that normalized difference vegetation index (NDVI) extracted fewer seaweed areas as compared to floating algae index (FAI) and seaweed enhancing index (SEI)
Three indices—floating algal index (FAI), normalized difference vegetative index (NDVI), and seaweed enhancing index (SEI)—were applied on Landsat 8 temporal images to extract seaweed patches along the Karachi Coast

Summary

Introduction

Seaweed is the name given to the numerous marine plants and algae that animate in seas, oceans, rivers, lakes, and other water forms. Seaweeds can be of three types based on the pigments they contain [1] Their light-absorbing pigments can be either red, green, or brown. Depending upon these pigments, seaweeds perform their process of photosynthesis. Seaweed stock is an important component of the coastal ecosystem that provides living space for mangroves and coral reefs and breeding grounds and food for several types of nearshore fish, shrimp, marine reptiles, shellfish, and mammals [2,3]. The beneficial chemical properties and nutritional value of seaweed have made it a commercially important coastal product. It is consumed in many countries of the world as human food, livestock fodder, and agricultural fertilizer [5]. The world’s seaweed production in 2012 was estimated at around US$6 billion, and 95 percent of this production was from Asian countries’ aquaculture [6]

Methods

Findings

Discussion

Conclusion