Earth observation data validation: implementation and performance evaluation of methods for chlorophylla retrieval

Khachatryan, Lilit

http://hdl.handle.net/10400.1/19849

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
KHACHATRYAN.pdf		2.25 MB	Adobe PDF	Download

Send Feedback

Authors

Khachatryan, Lilit

Advisor(s)

Goela, Priscila

Cristina, Sónia

Abstract(s)

High-Performance Liquid Chromatography (HPLC) allows determining the concentrations of phytoplankton pigments and is the most used technique to validate Earth Observation (EO) data of Chlorophylla (Chla). The challenges of current HPLC methods for pigment analysis are the use of toxic solvents and the coelution of important pigment peaks (e.g. mono and divinyl forms of chlorophylls). Despite that, these methods are established in reference laboratories, and its performance metrics were already assessed in previous work conducted by National Aeronautics and Space Administration (NASA) intercalibration exercises. A more recent method Sanz et al. (2015) is referred as having several chromatographic and instrumental advantages as: a) the use of a simpler binary elution gradient; b) the complete resolution of mono and divinyl forms of chlorophylls; and c) a mobile phase with low toxicity solvents (methanol and ethanol). In the first part of the thesis, the laboratorial performance of Sanz et al. (2015) methodology for the analysis of phytoplankton pigments was assessed, according to performance metrics defined in NASA HPLC Round-Robin Experiments, such as resolution, injection precision and retention time precision. Considerations about main coelutions and the response to different injection volumes are also discussed. In the second part of the thesis, phytoplankton pigments were quantified in 12 coastal water samples. The followed methodology implemented in CIMA-UAlg laboratorial conditions showed to be easy to implement for a routine level of work and the resolution between mono and divinyl forms of Chlorophylla was achieved. Implications of the results for the pigment quantification were also discussed and are presented. This study contributes to the overall effort of providing accurate in-situ data to validate satellite EO Chla data, and to accomplish the goal of improving the estimations of phytoplankton distribution.