Remote sensing estimation of the biomass of floating Ulva prolifera and analysis of the main factors driving the interannual variability of the biomass in the Yellow Sea

Yanfang Xiao; Jie Zhang; Tingwei Cui; Jialong Gong; Rongjie Liu; Xiaoying Chen; Xijian Liang

doi:10.1016/j.marpolbul.2019.01.037

Remote sensing estimation of the biomass of floating Ulva prolifera and analysis of the main factors driving the interannual variability of the biomass in the Yellow Sea

Mar Pollut Bull. 2019 Mar:140:330-340. doi: 10.1016/j.marpolbul.2019.01.037. Epub 2019 Feb 2.

Authors

Yanfang Xiao¹, Jie Zhang², Tingwei Cui², Jialong Gong³, Rongjie Liu², Xiaoying Chen², Xijian Liang³

Affiliations

¹ First Institute of Oceangraphy, Ministry of Natural Resources, Qingdao, Shandong 266061, China. Electronic address: xiaoyanfang@fio.org.cn.
² First Institute of Oceangraphy, Ministry of Natural Resources, Qingdao, Shandong 266061, China.
³ College of Information Science & Engineering, Ocean University of China, Qingdao, Shandong 266100, China.

PMID: 30803652
DOI: 10.1016/j.marpolbul.2019.01.037

Abstract

Since 2007, green tide blooms with Ulva prolifera as the dominant species have occurred every summer in the Yellow Sea. Biomass is a critical parameter used to describe the severity of green tide blooms. In this study, we analyzed the relationships between several indices (normalized difference vegetation index (NDVI), floating algae index (FAI), ratio vegetation index (RVI), enhanced vegetation index (EVI), ocean surface algal bloom index (OSABI), Korea Ocean Satellite Center (KOSC) approach) and the biomass per unit area of Ulva prolifera by using the in situ measurements from a water tank experiment. EVI, NDVI, and FAI showed strong exponential relationships with Ulva prolifera biomass per unit area. In order to apply the relationships to satellite remote sensing data, the impacts of the atmosphere (different aerosol optical depth at 550 nm) and mixed pixels to the relationships were analyzed. The results show that atmosphere has little effect on the relationship between EVI and Ulva prolifera biomass per unit area with R² = 0.94 and APD (the average percentage deviation) = 19.55% when EVI is calculated from R_rc (Rayleigh-corrected reflectance), and R² = 0.95 and APD = 17.53% when EVI is calculated from R_toa (top-of-atmosphere reflectance). Due to the low sensitivity to the atmosphere, the EVI relationship can be directly utilized in the top-of-atmosphere (TOA) reflectance without atmospheric correction. In addition, the EVI was slightly affected by mixed pixels with the APD only increased by ~10%. The EVI relationship was then applied to a long MODIS image time series to obtain the maximal total biomass of floating Ulva prolifera in the Yellow Sea from 2007 to 2016. The results showed that the maximum and minimum total biomass occurred in 2016 (~1.17 million tons) and 2012 (~0.074 million tons), respectively. The main factors that caused the inter-annual biomass variability were analyzed. The total amount of nutrients from Sheyang River which was the largest river on the northern coast of Jiangsu Province, and Porphyra cultivation in the Radial Sand Ridges of Jiangsu Province had both strong correlation with Ulva prolifera total biomass.

Keywords: Atmosphere effect; Biomass; EVI; Ocean color; Remote sensing; Ulva prolifera.

MeSH terms

Biomass
Environmental Monitoring / methods*
Eutrophication*
Oceans and Seas
Remote Sensing Technology*
Republic of Korea
Seasons
Ulva / growth & development*