Remote Sensing - Validation, spatial and temporal patterns in sea surface temperature and chlorophyll-a

Chapter of report - Southern Bluefin Tuna Aquaculture Subprogram: Risk and Response - Understanding the Tuna Farming Environment, 2009 / J. E. Tanner and J. Volkman (eds.), Ch. 6 pp. 127-150

This chapter provides the results of an investigation into the seasonal variability of sea surface temperature and chlorophyll-a within and adjacent to Spencer Gulf using satellite-based remote sensing imagery. Firstly, MODIS estimated chlorophyll-a and sea surface temperature are compared to field-based measurements made in the tuna farming zone (TFZ) to ensure that MODIS methods are valid in the region, and the results from MODIS are representative of the true conditions as measured using traditional techniques. This validation exercise showed that MODIS can accurately determine sea surface temperature measurements within Spencer Gulf, with MODIS estimates explaining 94% of the variation in field temperature measurements. The results for chlorophyll-a are less conclusive, with MODIS measurements explaining just 46% of the variation in the field data. Water depth may be a contributing factor to the accuracy of MODIS estimated chlorophyll-a, with measurements taken from water depths of more than 20 m displaying a better relationship than those taken from depths of less than or equal to 20 m (77% of variation explained vs. 47%). For depths > 20 m, the root mean square (rms) error of 27% falls well within the target error of 35% for MODIS chlorophyll estimations. Validation in areas of Spencer Gulf outside the TDZ is yet to be done. In the second part of this study, MODIS monthly composite imagery was used to investigate spatial and seasonal variation in both chlorophyll-a and sea surface temperature over a period of five years. Both the magnitude of chlorophyll-a concentration, and the timing of the seasonal patterns of chlorophyll-a, change with distance along a transect running down Spencer Gulf, with highest concentration in the shallower waters of the north and decreasing concentrations towards the open ocean. These was also an increase in chlorophyll-a concentration in close proximity to the TFZ compared to nearby waters of southern Spencer Gulf (by up to 100% in February-July). While bottom reflectance may be causing some interference in shallow northern areas, and thus quantitative measures of chlorophyll-a may be incorrect in these areas, the broader spatial and temporal patterns are likely to be correct. Further validation in other areas of the gulf will be undertaken to confirm this. Sea surface temperature measurements showed a clear contrast between gulf and non-gulf waters. Waters within Spencer Gulf showed both the highest summer temperatures and also the lowest winter temperatures, thus undergoing a greater seasonal transition than the waters outside the gulf.

Paul Bierman, Megan Lewis, Jason Tanner and Bertram Ostendorf