Extraction of Rice Phenological Differences under Heavy Metal Stress Using EVI Time-Series from HJ-1A/B Data.

Shuyuan Liu,Meiling Liu,Ling Wu,Xiangnan Liu,Chao Ding,Zhi Huang

doi:10.3390/s17061243

Shuyuan Liu, Meiling Liu + Show 4 more

Open Access

https://doi.org/10.3390/s17061243

Copy DOI

Abstract

An effective method to monitor heavy metal stress in crops is of critical importance to assure agricultural production and food security. Phenology, as a sensitive indicator of environmental change, can respond to heavy metal stress in crops and remote sensing is an effective method to detect plant phenological changes. This study focused on identifying the rice phenological differences under varied heavy metal stress using EVI (enhanced vegetation index) time-series, which was obtained from HJ-1A/B CCD images and fitted with asymmetric Gaussian model functions. We extracted three phenological periods using first derivative analysis: the tillering period, heading period, and maturation period; and constructed two kinds of metrics with phenological characteristics: date-intervals and time-integrated EVI, to explore the rice phenological differences under mild and severe stress levels. Results indicated that under severe stress the values of the metrics for presenting rice phenological differences in the experimental areas of heavy metal stress were smaller than the ones under mild stress. This finding represents a new method for monitoring heavy metal contamination through rice phenology.

Highlights

With rapid development of the industry, rice paddies in some areas have been subjected to heavy metal pollution
Approximately 50 pixels from CCD images were extracted in each of the four experimental areas to fit the time-series of the enhanced vegetation index (EVI) and extract phenological periods by pixel, before experimental areas to fit the time-series of the EVI and extract phenological periods by pixel, before averaging the pixel information in the same experimental area
The methodology was identified through extracting rice phenological differences under mild and severe heavy metal stress levels by the EVI time-series, which was obtained from HJ-1A/B CCD images and fitted with asymmetric Gaussian model functions (AG)

Summary

Introduction

With rapid development of the industry, rice paddies in some areas have been subjected to heavy metal pollution. Most research on heavy metals in rice are carried out from the aspects of chlorophyll, leaf area index, cell structure (using a handheld radiometer for extracting spectral or crop parameter information in the field) [7,8,9], or by adopting empirical or semi-empirical models which were established based on the relationships between sensitive spectral characteristics and the heavy metal concentrations, or physiological parameters of crops for the dynamic monitoring of heavy metal stress [10,11,12,13] These studies are predominantly based on the growth parameters above ground and root weight under heavy metal stress, and these methods are not easy to directly implement.

Objectives

Methods

Results

Discussion

Conclusion