Analysis of vegetation dynamics and application in agriculture using MERIS Terrestrial Chlorophyll Index

Abstract

Vegetation phenological stages are important indicators for monitoring vegetation growth, evaluating climate change impacts on vegetation, control atmospheric general circulations and carbon sequestration. Traditional phenology observations rely on fixed-point visual inspection. However, this method is labour-intensive and subjective, and often limited to few species. Remote sensing technology using vegetation indices provides a more objective, long-term, continuous and efficient way to monitor land surface phenology from regional to world wide scale. The European Space Agency (ESA)’s Medium Resolution Imaging Spectrometer (MERIS) data in red/NIR (near infrared) were used to produce the level-2 product of MERIS Terrestrial Chlorophyll Index (MTCI). The MTCI, is strongly linked with the red edge position (REP) in vegetation spectra, and in turn the foliar chlorophyll content, making the MTCI a useful product of vegetation phenology indicator.

In this thesis the MTCI data with different resolutions were applied to monitor vegetation phenological variables over mainland China, namely onset of greenness (OG) and end of senescence (ES). Then they were correlated with climatic factors of temperature and precipitation, demonstrated the main drives for major vegetation types in climate zones. Both MTCI and NDVI time-series captured the growth patterns for major vegetation types, the OG estimates were more consistent than the ES, and overall the NDVI gave later ES estimates than the MTCI. 9-year phenology maps showed that the OG was advanced and the ES was delayed in general. The OG was more related with latitude than the ES especially in the north China, while it was the opposite for the ES. And it was found in north China, the temperature was the main driver for the earlier OG, while in the south precipitation played a prominent role in advancing the OG. For the ES, both precipitation and temperature influenced partially. In Qinghai-Tibet Plateau, the precipitation was the main driver for both shifting OG and ES of grass, while less influenced by temperature. Among the vegetations that were examined, the broadleaf forest had the strongest correlation with climatic factors; the needle leaf forest was also greatly influenced by climate in cold temperate zone; the grass was highly affected by climate, while the mixed forest and crops were at moderate level.

In the light of the abilities of MTCI in monitoring vegetation phenology, MTCI was applied into specific situations to test the performance on phenology-based applications, including mapping paddy rice in northeast China and predict rice yield. The results were well consistent with the statistical data on the prefectural level and county level in spatial distribution and quantity from 2007 to 2011. The crop yield regression models indicated that the maximum value of MTCI time-series has a better correlation with rice yield.

In summary, MTCI has its own advantages than popular index such as normalised difference vegetation index (NDVI). It is more sensitive to high values of chlorophyll content and less sensitive to spatial resolution and atmospheric effects. Although the MERIS was not in operation anymore in May 2012, the successor, the Sentinel satellites were launched, with a wider range of wavelength from blue to shortwave infrared, including red edge bands. Therefore, index for estimating foliar chlorophyll can be produced to combine with other Sentinel products in agricultural, biological, and ecological studies.

More information

Identifiers

ORCID for Jadunandan Dash: orcid.org/0000-0002-5444-2109

ORCID for Peter Atkinson: orcid.org/0000-0002-5489-6880

Catalogue record

Export record