As one of the most important indicators for studying marine primary productivity, chlorophyll concentration in seawater can be quickly measured by laser-induced fluorescence (LIF) technology. In the traditional theory for obtaining chlorophyll concentration by LIF, the chlorophyll concentration nchl=CIＦ／Ｒ, where ＩＦ and Ｒ are fluorescence intensity of chlorophyll a (Chl-a) at 685 nm and Raman scattering intensity of water respectively, and Ｃ is a system constant. Withoutconsidering induced fluorescence saturation, this theory is based on an assumption that both the fluorescence intensity at 685 nm and water Raman intensity are linear with the intensity of incident laser. However, experimentsconfirmed the existence of non-linear relationships between the induced fluorescence energy at 685 nm and laser energy. While the linear relationships between water Raman intensity and pulse intensity have always existed without saturation excitation. In order to explore the effect of non-linear fluorescence change under saturation excitation, two series of measurement were done in the experiments. Fluorescence of the solution with constant Chl-a concentration was measured by varied laser powers, and a constant laser power was used to obtain the solution fluorescence of varied Chl-a concentrations. The third harmonic of Nd∶YAG laser at 355 nm was the excitation source. Thus, Raman scattering at 404 nm and fluorescence at 685 nm of Chl-a solutions were the key part of emission spectra. The experiment results were discussed in section 3. In the first part, the emission spectra of Chl-a solutions were measured by LIF with excitation light intensity variation. It shows a linear relationship between Raman scattering and excitation intensity, while fluorescence intensity at 685 nm appeared nonlinear change under saturated excitation. Moreover, fluorescence intensity of Chl-a solution with higher concentration increased to plateaus earlier, and the ratio of Raman scattering intensity to excitation intensity in the linear relationship decreased with Chl-a concentration. The data analysis shows that a polynomial of degree 4 fitting the changes of fluorescence intensity and the value of the Raman scattering efficiency can qualitatively characterize the saturation of fluorescence at 685 nm. Secondly, for the purpose of analyzing the effect of fluorescence nonlinearity on the applicability of traditional theory in chlorophyll concentration inversion, with considering the phenomenon of fluorescence saturation existing in the application of ocean Lidar for detecting chlorophyll concentration, the emission spectra of samples with different Chl-a concentrations were measured with a constant excitation intensity. The relationship between IＦ/R and Chl-a concentration was obtained under the excitation power at 52.00, 80.70, 132.10 and 197.30 mW·cm-2. Experiments show that IＦ／Ｒ is still in linear relationship with Chl-a concentration under the condition that the exciting radiation is not changed. But, the concentration from the traditional inversion theory by LIF is less than the real Chl-a concentration measured by a high excitation intensity which leads to fluorescence saturation effect. Therefore, the inversion module is necessary to be corrected with ＣＦ which is related to fluorescence nonlinearity under saturation excitation. A more accurate inversion is based on IＦ／Ｒ＝ｎchl/C+CＦ. And, it is worth mentioning that the system constant C in this correction module increases with the exciting intensity. Consequently, saturation excitation causesfluorescence nonlinearity and affects the measurement of Chl-a concentration by LIF technology. It is regrettable that the polynomial obtained by the fluorescence data fitting cannot quantify the impact of fluorescence saturation effect, due to the complexity of the nonlinear factors. However, when the excitation power is constant, a corrected inversion can be experimentally obtained and used to measure Chl-a concentration by LIF in field surveys.