| The temporal and spatial distribution of chlorophyll a concentration in Taihu Lake was analyzed using the data measured from January to October in 2005. Otherwise, chlorophyll a retrieval model was built using the synchronous spectrum data obtained in different season to evaluate the effect of chlorophyll a concentration temporal and spatial variation on the retrieval model. Firstly, temporal distribution of chlorophyll a concentration was analyzed. Then, chlorophyll a concentration was induced by Inverse Distance Weight method. Lastly, the chlorophyll a concentration estimation models based on spring, summer, autumn and different trophic states were built. An obvious spatial-temporal distribution was found for chlorophyll a concentration in Taihu. Chlorophyll a concentration had the maximal value in summer and the minimal value in winter,and the average chlorophyll a concentration was 56.29 μg/L, 13.61 μg/L respectively. Due to the effect of autumn, the average chlorophyll a concentration in autumn was higher than that in spring, and the average chlorophyll a concentration was 26.43 μg/L, 34.78 μg/L respectively. Marked spatial difference for chlorophyll a concentration was found in summer, while in winter the spatial variation was ambiguous. The spatial variation in autumn was greater than that in spring as influenced by summer. Spatially, the variation in north area was greater than those in the other areas and the south area had less variation during the year. Because of the different chlorophyll a concentration, different season had different retrieval model. Band-ratio had better retrieval result in spring and autumn as there were lower concentration and less variation correspondingly. While in summer, because of the high chlorophyll a concentration and great spatial variation, among the arithmetic built in summer, the differential arithmetic was the optimal model, and the difference of Chl-a concentration retrieval model of different trophic state was minor. |
