| Summary: | 碩士 === 國立成功大學 === 地球科學系碩博士班 === 96 === The general approach of monitoring the water quality in a reservoir relies on the point measurements made at ground stations, which is both time and labor consuming. Employing the technique of remote sensing, by contrast, would enable us to acquire a near-real-time and synoptic view of the entire reservoir. To relate the remote sensing data to the water quality, however, we need to establish the relationship between the water constituents and various optical properties. This can be investigated in a water tank with detailed measurements of optical properties for various water samples of controlled constituents. By employing the forward Monte Carlo ray-tracing (FMCR) technique to simulate the radiative transfer process in a water tank, we attempt to establish a quantitative relationship between the Inherent Optical Properties (IOP) and the Apparent Optical Properties (AOP).
As described in the earlier work by Liu and Woods (2004), the basic principle of our FMCR model is to simulate a beam of light by a very large number of photons. Following the path of each photon, we can use a series of random numbers to determine the photon’s life history according to different probabilities for different phenomena. The final light field is the cumulative contribution of total photons. The water tank is a cubic box with Lambertian surfaces. One unit light comes from the top of the box and the detectors can be deployed at any place in the box to collect photons. For each simulation, the IOPs are specified by various constituent-optical models.
We carefully validate our FMCR model against the commercial Hydrolight model for the case of infinitely deep bottom and the case of one-dimensional parallel bottom. The results verify the simulation of interaction between the photon and the Lambertian surface. After adding the walls of the box, the light near eight corners of the box is apparently brighter than the one simulated in the case without walls. This phenomenon is further enhanced as the size of box is reduced.
From simulations of model, there are some reasonable phenomenons we could discover, like high single scattering albedo or high reflectance will lead to high radiance. This FMCR model can be employed to systematically investigate the relationship between IOPs and AOPs by running a comprehensive FMCR simulation to generate a look-up-table (LUT). This LUT would assist us to derive the IOPs directly from the measurements of AOPs in the future.
|
|---|
