Shiomi Shikasho

Numerical quantification of the significance of aquatic vegetation affecting spatial distribution of Japanese medaka (Oryzias latipes) in an agricultural canal

Aquatic vegetation plays a very important role in providing food, shelter, and nursery habitat and is also regarded as hydraulic resistance in the stream environment. To achieve better ecological restoration, this trade-off should be solved both hydraulically and ecologically. This study quantifies the effect of aquatic vegetation on the spatial distribution of Japanese medaka (Oryzias latipes) to evaluate its importance to fish habitat preference. The preference for aquatic vegetation index is calculated using a fuzzy preference intensity model (FPIM) with interactions among water depth, current velocity and cover ratio in an agricultural canal. In this model, simplified fuzzy reasoning is introduced to explicitly take the essential vagueness of fish behavior into consideration, and a simple genetic algorithm is used to search for an optimum model representation. Uncertainties in measurement errors and dispersions of the physical environment are positively taken into the model using symmetric triangular fuzzy numbers. To overcome the difficulty in model construction with insufficient data observed in an agricultural canal, this model was conjugated with a model developed in a laboratory experiment. The model obtained was then assessed using the AIC (Akaike’s Information Criterion) to evaluate the significance of vegetation index with a statistical approach. The results suggest the significance of vegetation index to habitat selection by Japanese medaka and that utilization of the AIC enables us to grasp the validity of an additional factor contributing to habitat prediction with a view to a definite scale

Numerical modeling of environmental behavior and fate of tributyltin in a semi-closed bay

The environmental behavior and fate of tributyltin (TBT) in the northern Ariake Sea, resulting from the use of TBT-containing antifouling paints on hulls of ships, pleasure crafts and docking facilities, was assessed by numerical simulations. First, a mathematical model was devised on the basis of a non-steady state equilibrium, one box multi-compartment model consisting of the surface micro-layer, the water column, the mud-layer, and the bottom sediment compartments. The movement of TBT among the four compartments was modeled by resuspension of bottom mud, deposition of suspended sediment, film penetration and water advection in each compartment. Furthermore, a one-dimensional diffusion equation was introduced into the bottom sediment compartment to calculate profile distribution of TBT. The reactivity of TBT considered within the compartments included biological degradation, adsorption to particulate matter and diffusion. Next, the optimal amount of past TBT loads, reflecting the recent observations of TBT concentration in the waters and sediments of the northern Ariake Sea, was searched by a simple genetic algorithm. The relative sensitivity of various model parameters were also determined to identify the more important parameters for estimating the environmental behavior and fate of TBT. Finally, the future status of TBT contamination of the northern Ariake Sea was predicted assuming the discontinued use of TBT-containing antifouling paints. Despite its simple model structure relative to hydraulics, it was concluded that this multi-compartment model adequately estimated the environmental behavior and fate of TBT.

Vagueness ratio of storage function model parameters and its relation to the accuracy of hydrograph prediction

Fuzzy linear regression (FLR) has been applied in many areas, including several areas of engineering. For instance, it has been successfully applied in conceptual rainfall runoff modeling to identify the parameters of the storage function (SF) model. In the FLR, each model parameter is not a crisp value as found in traditional procedures but the model parameters have a fuzziness value characterized by the center (α) and the width (c) values. The width value could be further analyzed in terms of its contribution in the prediction accuracy. In this case, the vagueness ratio, defined as a ratio between the width value and its center value for the model parameters, is presented here as a surrogate for analyzing any possible relation to the prediction accuracy. Application of the vagueness ratio to a number of flood hydrographs at Ochohzu basin in Fukuoka, Japan, shows that the ratio is useful for assessing the accuracy of the hydrograph prediction; i.e., the higher the value of the ratio indicates the less accuracy in the prediction.

Mathematical Modeling of Preference Intensity of Japanese Medaka Fish for Instream Water Environment using Fuzzy Reasoning

Fish behavior is essentially vague, and the behavioral responses to an environmental change vary widely both among individuals and depending on circumstances. This suggests it is necessary to positively consider the uncertainty when constructing preference intensity models to environmental factors. In this paper, the essential vagueness of the behavioral responses of Japanese Medaka Fish (Oryzias latipes) was positively taken into consideration in studying the preference intensities of the fish to three environmental factors: water depth, current velocity, and cover. The factors were quantified by laboratory open-channel experiments and a fuzzy reasoning technique. The Japanese Medaka Fish, a river fish, was selected to quantify preference intensities because it has become symbolic of the restoration of the countryside ecosystem in Japan. A simple genetic algorithm was introduced to search for an optimal structure of a fuzzy preference intensity model. The fuzzy preference intensity model was then verified by on-the-spot examinations. The results indicated that the two environmental factors of current velocity and cover obviously affect the environmental preference of Japanese Medaka Fish, and the fuzzy preference intensity model was a reasonable model to account for the actual behavioral responses of the fish.

Experimental Study on the Fish Preference for Hydraulic Environments

The preference intensities of river fish to environmental factors were discussed in this paper. The three environmental factors of water depth, current velocity and cover were selected as the principal factors affecting the availability and quality of the fish habitat in natural rivers. Japanese Medaka Fish (Japanese killifish, Oryzias latipes) was selected as a river fish subject to quantify preference intensities because Japanese Medaka is now regarded as the symbolic fish for the restoration of countryside ecosystems in Japan. Mathematical models for preference intensities of Japanese Medaka to the three environmental factors were constructed in laboratory open-channel experiments. A simple genetic algorithm was newly introduced to search for the optimal functional representation of preference intensity and to determine parameter values of the preference intensity models. The results indicated that the maximum preference level of Japanese Medaka was determined at a water depth of 9.4 cm and at a current velocity of 2.8 cm/s, and that the All-cover condition had a markedly high level compared with other cover conditions. The relative weights necessary for calculating the combined preference level of the three environmental factors were evaluated as 0.32 for depth, 1.0 for velocity and 0.66 for cover, when the maximum weight was normalized to be unity. The weight values suggested that the current velocity has the strongest preference intensity among the three environmental factors and that the environmental preferences of Japanese Medaka were not greatly affected by water depth.