Hyperspectral remote sensing offers the potential to detect water quality variables such as Chl-a by using narrow spectral channels of less than 10 nm,which could otherwise be masked by broadband satellites such as Landsat TM.Fluorescence peak of the red region is very important for the remote sensing of inland and coastal waters,which is unique to phytoplankton Chl-a that takes place in this region.Based on in situ water sampling and field spectral measurement from 2004 to 2006 in Nanhu Lake,the features of the spectral reflectance were analyzed in detail with peak position shift.The results showed:An exponential fitting model,peak position=a(Chl-a)b,was developed between chlorophyll-a concentration and fluorescence peak shift,where a varies between 686.11 and 686.29,while b between 0.006 2 and 0.006 5.It was found that the better the spectral resolution,the higher the precision of the model.Except that,the average of peak shift showed a high correlation with the average of different Chl-a grades,and the determination coefficient (R2) was higher than 0.81.It contributed significantly to the increase in the accuracy of the derivation of chlorophyll values from remote sensing data in Nanhu Lake.There is satisfactory correspondence between hyperspectral models and chl-a concentration,therefore,it is possible to monitor the water quality of Nanhu lake throngh the hyperspetral remote sensing data.