To investigate the dependencies of the profiles on the atmospheric stability (z'/L), field measurementsw ere carried out for wind velocity, temperature and turbulent fluxes in and above a dense residential canopy. 1) The vertical profile of wind velocity varies with (z'/L). It deviates from the conventional similarity function in the unstable condition. 2) Momentum flux slowly decrease with height. This tendency is more significant in the unstable condition. 3) Skewness of u above the canopy is larger than the value reported in the vegetation studies. 4) Inversion layer is formed within the canopy around the sunrise due to the local heating of air near the roof top. 5) The gradient of potential temperature under unstable condition is dependent on the atmospheric stability but is relatively larger than the one of the conventional similarity function. 6) The range of roughness sub-layer for heat is smaller than the one for momentum.

View full abstract

This study aims to evaluate the mitigation effects on summer thermal stress by the current distribution of urban vegetation in Tokyo Metropolitan Area using an urban climate simulation. In previous urban climate researches, land use data were used generally to generate physical characteristics of ground surface. In contrast with them, this paper emphasizes on making use of vegetation coverage ratios calculated from satellite remote sensing data, which make it possible to assess the effects of small vegetation such as roadside trees that are unable to discriminate from land use data. The result shows that air temperature is decreased 1.5°C due to vegetation during daytime on a typical summer day in Tokyo. Vertical distribution and diurnal variation of cooling effects, and direction and velocity of wind are also analyzed.

View full abstract

Local climate change with urbanization in Kobe City area was investigated by using a turbulence closure model. Wind and temperature fields in air were computed with respect to the present ground conditions in the city and those of about fifty years ago when the city area was less extended. Parameters of ground surface such as albedo, geometrical roughness and water content of the old age were estimated from correlation between land use and categories of governmental districts. The analysis showed that urbanization in the last fifty years brings local warming such air temperature increases of about 0.5 degree in daytime and 0.3 degree in nighttime at the maximum. Intensification of sea-land breezes by the urbanization is also confirmed.

View full abstract

To investigate the diurnal variation of SPM (Suspended Particulate Matter) emitted from automobiles in metropolitan area, meso-scale numerical simulations were carried out using RAMS (Atmospheric Modeling System). The following results were obtained;
(1) The wind field dominately affect the diurnal variation of SPM. When simulated wind fields agree with those of the observation, diurnal variations of SPM also agree well with those of the observation.
(2) Sea breeze and boundary layer have strong influence on the SPM variation. In the morning, SPM is relatively high concentration, since a stable layer restrains vertical diffusion of SPM. In the afternoon, SPM concentrations suddenly decrease in accordance with the sea breeze and boundary layer development. Atmosphere in the BL is well mixed and SPM is diffused and advected.

View full abstract

Water and energy consumed by various urban activities are eventually released to the atmosphere and hydrosphere. Their impacts may not be negligible as the great amount of water and energy are dumped according to the growth of urban activities. To demonstrate the impact of urban growth on the water temperature released from wastewater treatment plants via a sewerage system, the decadal trend of the water temperature, as well as water amount and heat flux are revealed, and the factors affecting the trend are analyzed for the central Tokyo area. It is found that the effluent water temperature and the heat flux have been increasing over the past 36 years. Based on the model that quantifies the effluent water temperature, it is indicated that the increase of water temperature is attributed to the increase of supply-water temperature, the increase of the water used for bathing and the increase in the energy consumption for hot water supply due to the change of life style and the total energy demand increase.

View full abstract

In this paper the results of a number of idealized simulations of the airflow over a mountain ridge, using a fully compressible non-hydrostatic mesoscale atmospheric model is presented. First, the simulations were conducted with a steady background flow field as lateral boundary conditions. The results are interpreted with a special emphasis on formation and properties of lee waves and rotors. The effects of the changes in the atmosphere, topography and external wind field on lee wave and rotor properties are examined. It was found out that the flow fields generated by the present model are significantly different and much more complicated than those predicted by the linear theories. Most of the conditions produced single or multiple rotors that caused boundary layer separation on the lee side. Secondly, the effect of unsteadiness of the background flow on the lee wave dynamics was examined. It was found out that a gradual drop of wind speed could cause the lee wave train to move upstream and to cause a considerable rotor-like motion over the mountain ridge. An attempt is made to link this behavior of the model under the unsteady wind with the observed abundance of thunder-storm activities over mountain ridges compared with surrounding lowlands.

View full abstract

The so-called energy imbalance problem is investigated through numerical experiments using Large Eddy Simulation (LES). LES experiments are done for daytime atmospheric boundary layers over a flat surface with inhomogeneous heat flux. The heat flux is represented by a one-dimensional sinusoidal curve with a variable wavelengtha nd amplitude.
The following are the main results. 1) When the amplitude is enough, the bias and variance of imbalance is increase. A weak inhomogenity of surface heating, however, may relax the negative bias of imbalance. This is attributed to the heat transport efficiency of vertical mean wind. 2) Although longer wavelengths generally give a smallerb ias of imbalance, the 2km wavelength singularly increases the bias of imbalance. 3) The bias and variance of imbalance decrease in accordance with horizontal wind. Longer averaging time gives a smaller bias but a larger variance of imbalance. The tendency of these two results is almost the same as in the homogeneoush eatingc ase.

View full abstract

In this study, the dataset that is homogeneous in time and space was created by using meteorological and hydrological data obtained during HUBEX-IFO (1998/5/1-8/31). The domain of this dataset is from E110 to E122, and from N31 to N36 with 5min (about 10km) resolution. This dataset was produced to be used as forcing data for Land Data Assimilation by land surface scheme (SIBUC). Although the dataset created from the surface meteorological variables has much difficulty in the description of diurnal variation of radiation, since the original data are daily accumulated values. Then GMS data and the simple physical model are introduced to improve the diurnal variation of downward short-wave radiation. Furthermore, the mesh-data of hourly short-wave radiation is made using this method.

View full abstract

The purpose of this study is to investigate radiation transfer in vegetation layer and apply it to soil moisture estimation by passive microwave imager in global scale. A radiation transfer scheme with multiple parameters for leaf, stem, and branch is used. Sensitivity analysis shows that not only LAI (Leaf Area Index) but SAI (Stem Area Index) has strong influence on PD (Poralization Difference of brightness temperatures) and FD (Frequency Difference of brightness temperature). Among several different estimation algorithms, an algorithm which calculates soil moisture by PD with temporal variable LAI and spatial distributed SAI can give the most realistic seasonal variation of surface soil moisture. In this algorithm, spatial distributed SAI is derived from annual minimum FD.

View full abstract

Numerical simulations of canopy albedo for 3 dimensional model cities are performed using the Simple Urban canopy Model for Meso-scale simulation (SUMM). The algorism of the SUMM is described in this study. Model parameters and analytical conditions appropriate for validation studies are examined, and then the simulation results are compared with the experimental data. The model results agree fairly well with experimental results, even though slight disagreements exist especially for low solar angles. Both the simulations and experiments demonstrate that urban geometries significantly influence canopy albedos.

View full abstract

The interest of this study is twofold for a thunderstorm occurred in Tohoku district of Japan. The one is to investigate kinematic features of a multi-cellular line associated with heavy rainfall. Dual-Doppler radar data by 3-cm wavelength research radar are used. The 3-D wind fields derived from dual-Doppler analysis indicate that the pronounced low-level convergence coincident with the upward air motion is consistent with the formation of the line.
An extended version of the Volume Velocity Processing (EVVP) is proposed for single-Doppler retrieval of horizontal wind field. The results of application demonstrate that there is a discrepancy of scales between dual-Doppler derived wind field and EVVP-derived one. Overall, some improvements involved in the EVVP method could provide reliable estimation even when actual wind field is more complicated than the model of wind field.

View full abstract

About the heavy rain which brought great damage to Hiroshima prefecture in June 1999, various studies have been made. But, the cause of its heavy rain has not been examined from meteorological viewpoint. Since generation of the heavy rain is considered to originate in the overall weather condition, analysis of the weather data observed widely is important. Then, in this study characteristic of the front and the relevance with generation of the heavy rain have been examined in analyzing the GMS images at the time of the disaster. From characteristics of vector field of moving cloud by the mutual correlating method between the GMS images, we have grasped front position and the generating position of the rainy area moving with the front. Also, we examined relationships between the front positions by the GMS images and the GPV data.

View full abstract

Rainfall-topography relationship was analyzed by running numerical simulations using the MM5, which is the mesoscale model constructed in the Pennsylvania State University and National Center for Atmospheric Research (PSU-NCAR). For a quantitative estimation of topographic effect on rainfall distribution, the Dependence Line on Topographic Elevation (DLTE), one of the models which represent rainfall-topography relationship, were simulated and analyzed in four regions in Japan. The properties of DLTE and the requirements for establishing the relation of DLTE were determined in each region. Next, the influence of the scale of topography on rainfall-topography rolationship was investigated by running simulations under the conditions where the resolution of topography has various scales. It was found that the relation between the scale of topography and the degree of topographic effect differs from region to region, and that the degree depends largely on the scale of mountains in a region.

View full abstract

Since 1997, rainfall has been measured by using tipping-bucket rain gauges in a mountainous area as a part of GAME-Tropics (the Global Energy and Water Cycle Experiment (GEWEX) Asian Monsoon Experiment (GAME)). The fact that a larger amount of rainfall was recorded at a high altitude was attributed to duration and frequency rather than intensity. After investigating orographic precipitation using a regional climate model, the duration and frequency of rainfall at high altitudes were found to be caused by two mechanisms: 1) convective clouds over mountains are activated in the evening, and 2) stratus formed by radiative cooling or heating at the tops of clouds and low-level orographically triggered convective clouds cause a seeder-feeder mechanism. The results of a numerical simulation were quite consistent with the measurements acquired from the tipping-bucket rainfall gauges.

View full abstract

In this study, regional land-atmosphere coupled model (JSM-SiBUC) is applied to the GAME-HUBEX region to simulate the activity of the Baiu front observed on 2nd July 1998. To perform the regional 4DDA (four dimensional data assimilation) acculately, not only the used model but also the external dataset is very important. The main target of this study is to investigate the accuracy of the external dataset to be used for the regional 4DDA in future. Three dataset-GANAL, GAME-Reanalysis Verl.1 and Ver 1.5, which were produced by JMA (Japan Meteorolocal Agency)-are used for initial and external boundary condition. Through the comparison with upper air sounding data, raingauge data, and cloud image from GMS, the following results were obtained. 1) lower level specific humidity of GAME-Reanalysis data is smaller than observed one. 2) lower level potential temperature of GANAL data is larger than observed one. 3) due to high temperature in the GANAL data, it is difficult to simulate the meso-vortex observed on 2nd July 1998. 4) due to low humidity in the GAME-Reanalysis data, it is more difficult to simulate the above case.

View full abstract

In recent years the economical conditions and the rising public criticism against large-scale publicworks represent the key issue toward the new directions of flood management in Japan. In consequencethe future flood actions is shifting from the well known “traditional hard protection measures stating thatno flood should occur” to the so called “soft countermeasures” stating that larger efforts should beconcen trated on new alternatives aiming to reduce the loss of life and damage of important propertieswhen a natural flood does occur. Therefore, under this new shift it becomes a primary issue to improvethe current soft abilities to forecast the size and impact of flood disaster before a flood does occur. Thepresent research is a growing assessment tool that aims to provide valuable decision support informationlikely to enhance the judgment abilities of river managers in providing disaster countermeasures actionssuch as refuge guidance. At earlier stage of this study the correlation nature between the actual AMeDASprecipitation data and the 6-hour forecast precipitation results of the Japan Meteorological Agencymesoscale numerical weather prediction model (Meso Spectral Model: MSM) was investigated. Theresults showed that rainfall forecast by MSM model would still need further research and improvement inorder to be used in practice. At a second stage, the MSM climatological data where used to derive newoutput variables (such as water divergence) by applying a straight for ward mathematical trans for mationmodel. The correlation results between the actual AMeDAS precipitation data and the transformed outputvariables were relatively high and promising.

View full abstract

To evaluate the effect of the global warming on precipitation is essential for future global water resourcemanagement. Previous studies have indicated the precipitation increase in Asian monsoon region. Fewprevious studies; however, have mentioned the interannual variability as the coefficient variation of precipitation (CVP) in sub continental scale. This paper has analyzed the dataset of the global warmingexperiments by general circulation models (GCMs) in JJA summer. The result has confirmed the increasein precipitation and the standard deviation of precipitation in East Asia region but revealed no increasein the interannual precipitation variability as CVP.

View full abstract

This research interprets simulated changes in characteristic of precipitation over Japan derived from an integrated hydrologic-atmospheric model for analysis of water uses on a river basin scale. First accuracy and reliability of simulated changes are evaluated in comparison with ground data, and changes averaged over each climatic region are evaluated in terms of mean and standard deviation that will be used to generate two sets of hydrologic data series under present and CO2 doubling scenarios for sensitivity analysis of a water system to climate change. It was found that simulated precipitation represented the actual regional precipitation distribution with better accuracy but was generally underestimated, that the MRI-GCM global warming scenario would induce monthly decrease in precipitation in almost all climatic regions, and that regions with smaller precipitation tend to have larger degree of precipitation fluctuation.

View full abstract

The water stable isotopes (D and ¹⁸O) in precipitation, which are widely used as tracers in the global hydrological cycle, have large spatial and temporal fluctuations. However, a primary factor of the fluctuations has not been quantitatively described by previous studies. Thus, this study has developed a new isotope circulation model (ICM), which has higher spatial and temporal resolution such as 1.25°1.25° and daily time-scale than former isotope model studies. Notice that ICM has only one vertical layer, which means the isotopic behavior are vertically integrated in each grid. ICM has reproduced the daily variation of precipitation δ¹⁸O in Chiangmai with good correlation (0.76), as well as the monthly δ¹⁸O in the whole globe. The results have (1) revealed the primary factor of the isotope fluctuations and (2) supported the validity of the assumption in ICM.

View full abstract

In this study, we investigate the effect of size distribution of rain drop on the wet deposition of nitrate. The numerical acid rain model which uses the bin method for calculating the cloud microphysical processes is used for the investigation. The results are as follows, i) rain which mainly consists of large size droplets makes much amount of wet deposition of nitrate, ii) accumulated rainfall amount is important to estimate the amount of wet deposition of nitrate, iii) spatial variation of wet deposition of nitrate is large, and iv) wet deposition of nitrate can be estimated by space borne radar with considering the graupel fall.

View full abstract

Taking account of the variation of the water vapor content and the temperature in the air parcel, a new model for the estimation of the cloud droplet size, the temperature and the saturation ratio has been developed by modifying traditional Kohler model for the equilibrium state. The modification of Kohler model is based on the mass and the heat conservation laws for the water (vapor and liquid) in the air parcel. Variations of size, temperature and saturation ratio with number density of CCN are simulated numerically. The results of simulations show that the temperature variation of the cloud droplet and air parcel is so small to change the values of physical properties but large enough to make the air parcel unstable (or convective) in the atmosphere.

View full abstract

As a rainfall field generation model, a random cascade is applied at the Miyama Radar observation area (192x192km) and the model parameters are identified by using radar rainfall data. To include the heterogeneity of rainfall fields into a random cascade, the dependence line on topographic elevations (DLTE) developed by Nakakita et al.(2001) is introduced, and the topographic effects of spatial rainfall fields are well simulated by using the random cascade generator.

View full abstract

We investigated temporal and spatial precipitation data from the view point of scale invariance or scaling, and examined applicability of random cascades whose generators are defined by β model, since random cascades generate data which exhibit fractal behavior. Although β model is too simple to simulate daily precipitation time series, 151 of observation points are classified into two groups depending on ρ value of β model; one is a set of observation points along the Japan Sea, and the other is along the Pacific. Regardless of the groups, daily precipitation at 151 observation points exhibit properties of multifractal. Regarding spatial data, we investigated scaling properties using relatively long term's hourly precipitation data. Although each scan exhibits well scaling, the relationship between ρ value and spatial mean precipitation is not one to one function, if all data are used. If relatively short term's data, namely, data of each precipitation event is used, the relationship is relatively clear and parameters in the relationship depend on events.

View full abstract

The purpose of this study is to elucidate temporal characteristics of rainfall intensity averaged at different time intervals. Using AMeDAS 10min rainfall product, the features of 10min and 1h rainfalls on Kanto area were examined. It was found out that, on average, the maximum rainfall intensities of 10min series are between 1.5 to 2.0 times of the corresponding 1h rainfall amount. The frequency distributions of 10min rainfall intensity in each 1h intensity range could be modeled as a gamma distribution. The properties of thunderstorm and non-thunderstorm rainfall were compared in plain and mountainous areas. For thunderstorm-rainfalls, the higher 1h rainfalls in plain areas are characterized by larger 10min intensities compared with those of mountainous areas. The non-thunderstorm rainfall does not show such a difference.

View full abstract

Hourly rainfall is one of the most important hydrological variables. In hydrological engineering, we are often more interested in maximum 60 minutes rainfall rather than hourly rainfall recorded at an interval of one hour or 10 minutes. This paper describes the method to estimate maximum 60 minutes rainfall from hourly rainfall. Maximum 60 minutes rainfall is estimated by multiplying the bigger value of two consecutive hourly rainfalls recorded at an interval of one hour by 1.092, or the average of them by 1.587. Maximum 60 minutes rainfall is estimated by multiplying the bigger of two consecutive hourly rainfalls recorded at an interval of 10 minutes by 1.025, or the average ofthem by 1.102.

View full abstract

This paper expanded 1 variate maximum entropy distribution into 1 variate nonstationary maximum entropy distribution (1VNME distribution). The given information to the 1VNME distribution is the statistical moments. By using the 1VNME distribution, nonstationary frequency analysis was applied to monthly, seasonal and annual precipitation. The result obtained was as follows. Although fitting rate of the 1VNME distributions were different a little in monthly, seasonal and annual precipitation, the 1VNME distributions were almost fitted to histograms. Therefore this distribution could confirm that it was efficiency. Besides, because the shape of the distribution changed in time series, it was clear that hydrological value had the nonstationarity. Finally, Although the probabilistic value of hydrological variable had the difference in monthly, seasonal and annual precipitation, it showed the secular change. As a whole, the probabilistic value of hydrological variable decreased in the month of winter and increased in the month of summer.

View full abstract

In order to conduct adaptive water management for an entire watershed, long-term hydrologic conditions must be clarified. In particular, the hydrologic process related to snow plays an important role in cold, snowy regions such as Hokkaido. Therefore, reliability of estimations of snow accumulation and snowmelt are needed to perform effective water management, while evapotranspiration relating to the conditions of soil and vegetation is also needed to clarify the water budget in an entire watershed.
This paper focuses on the hydrologic process reproduced by the Two-layer model that deals with the heat balance in the atmosphere, the vegetation layer and the ground surface, as well as the long-term runoff characteristics calculated by the Tank model. In the model, the hydrologic process was evaluated in relation to the conditions of snow pack and snowmelt. Specifically, the parameters of the Tank model were effectively fine-tuned by applying the Filter-separation AR method and the Newton-Raphson method. Thus, the proposed model gives a promise to reasonably estimate the hydrologic process in cold, snowy region.

View full abstract

The influences of snow condition on the seasonal variation of vegetation are investigated in little snow year (1998), normal snow year (1999) and heavy snow year (2000) in the Taki Dam basin of the Agano River. At first, the relationships among NDVI, WI, evapotranspiration and snow covered area are evaluated. The area of snow cover is proportional to mean basin NDVI in snowmelt season. Evapotranspiration in April is 36mm/month in little snow year and 15mm/month in heavy snow year. Secondly, the distribution of vegetation growth ratio to temperature is calculated with a proposed new index. As the results, vegetation growth ratio is faster in the low elevation area (0-1300m), slower in the middle elevation area (1300-1700m) and faster in high elevation area (over 1700m), as temperature increase, respectively. This is related to snow condition and vegetation kinds.

View full abstract

Sensitivity of Land Surface Model (LSM), which is designed to simulate water and energy on large scale, to the temporal precipitation resolution is examined. The results of two sets of off-line experiments by a LSM are investigated. The first set of the experiments is global 1°× 1° simulation in 1987, which is driven by the same atmospheric forcings but different precipitation input, varying in temporal scale frnm high (6-hourly) to low (deca-day or monthly). Low temporal precipitation is obtained by averaging original 6-hourly precipitation. Original 6-hourly and obtained deca-day averaged or monthly averaged precipitation forcing are provided to the model and the results are compared at each grid pixel. The results show that increases in surface runoff are obtained at most of the grids with the low temporal precipitation, due to large increase in canopy interception. At the place where monthly LAI value is large and monthly precipitation is around 100 mm/mon, the effects of the temporal scale of precipitation tend to be the largest. Another set of experiments is also 1°× 1° off-line simulation at one grid pixel using the accumulation of the observed precipitation. Three downscaling schemes for retrieving daily precipitation from the monthly value are compared in terms of water balance calculated in the LSM. Both the number of days and intensity are shown to be important for the hydrological estimation in LSM. The water balance calculated in the experiment with γ distribution method shows fairly well accorded with the experiment using observed daily precipitation, and shows the applicability of the scheme for downscaling of daily precipitation from monthly data.

View full abstract

In recent years, the Sarobetsu Mire is suffering from aridification allowing for the invasion of Sasa bamboo grass. In order to clarify the cause of aridification, the hydrologic process in the entire watershed must be properly assessed. Specifically, reliability of assessments of the snow accumulation and snowmelt are needed to evaluate the water balance in cold, snowy regions.
This paper focuses on the hydrologic process estimated by the two-layer model that deals with the heat balance in the atmosphere, vegetation layer and ground surface as well as the simulations of groundwater. Accordingly, the hydrologic processes including the obscure factors such as snow accumulation, snowmelt and evapotranspiration were properly assessed. Based on such assessment, the characteristics of groundwater are analyzed using FEM. The Proposed method is considered to be proper is assessing water balance in cold, snowy regions and its effect on the groundwater.

View full abstract

In order to simulate global river discharge, global river routing network models (GRRNMs) have rapidly developed over the past decade. One practical problem of current GRRNM, is that they do not deal with reservoir operation. In this paper, simple reservoir operation model (ROM) for Total Runoff Integrating Pathways (TRIP), which is one of the GRRNMs, is introduced. The model has simple structure and its global parameters such as reservoir information, distribution of irrigated area or climatic rainfall are derived from published global data sets, aims its global application. ROM is applied to Bhumibol and Sirikit dam in the Chao Phraya river in Thailand as a case study. The calculated release and storage of them compared well with the observation, for 17 years simulation. Consequently the discharge at the lower reach calculated by TRIP-with ROM is dramatically improved compared with original TRIP calculation. It supports the usefulness of our model and possibility of global application.

View full abstract

Distributed TOP model was applied to the middle Mekong river basin divided into 5 sub-basins. The model parameters were estimated from land use data. The TOP model simulation remained regular deviation for observed data. The deviation was considered as storage effects such as reservoirs and irrigation. The regular variation was extracted as a storage model. The amplitude of the regular deviation was decided by catchment area. The storage model was substituted to TOP model and solved the regular deviation.
Water balance model with distributed T
OP model evaluated the distribution of potential water resources in the middle Mekong basin. As the results, east and west region of the middle Mekong have much water in wet season and poor water during one year, respectively. This potential water resource is used as one of indices to evaluate actual water resources.

View full abstract

Topography has dominant effects in hydrological modeling. The purpose of this study is to explore a proper subdivision level for distributed hydrological model BTOPMC (Block-wise use of TOPMODEL with Muskingum-Cunge method), so that simulation uncertainty due to basin subdivision could be reduced to quite a low level. The Pfafstetter subdivision algorithm was successfully integrated into the model and related problems and solutions were discussed. Using this BTOPMC, the effects of sub-basin scale on runoff simulations were investigated for the Fuji-kawa and the Nakagawa basin, respectively. It was found that as the average area of sub-basins becomes smaller, the simulated total runoff usually tends to be increased in flood/wet period and decreased in dry period, this is considered as the effect of averaging scale of the topographic index in the model. However, the results indicated that model performance appears stabilized when average sub-basin scale is about 1/150 of the entire basin.

View full abstract

A grid-based hydrological cycle analysis model, developed and verified in an urbanized small river watershed, is applied to Tama river basin. This model is a physically based distributed one considering not only natural hydrological cycle but also artificial water use. In this paper, a modeling technique is described for the Tama river basin that has 1, 240km2 and a population of about 3.4 million. Then the model calibration and verification are carried out with several river and ground water observation records. The simulated results show relatively good agreement with the observation. Interflow conductivity is found to be an important parameter for river discharge during the flood and its recession period.

View full abstract

Local scientists and engineers in Thailand have recognized and pointed out a decreasing trend in discharge in the Chao Phraya river in recent years. However, no solid data or analysis to show a decline has been reported. This study is to find trends in river discharge data and to analyze significant causes of the trend. The following results were obtained:(1) As a result of examining discharge at 10 gauging stations in the Chao Phraya river basin, the significant decline of the annual discharge is found at 8 stations out of the ten;(2) 17 out of 18 sub-basins of the Chao Phraya river basin have no tendency in annual rainfall change, but the only one sub-basin in the upper Yom river basin has tendency of declining rainfall according to the Kendall rank test at 95% significance level;(3) and it is considered that major cause of the discharge decline is a change of the land use to cropland such as the orchard from paddy field rather than rainfall.

View full abstract

A study of flood flow pattern in floodplain of a large river in the natural groundwater recharge area during flood period from 2001 to 2002 is presented. The floodplain of the Yom River in Phichit Province, Thailand, was carried out during flood period 2001 and 2002. Some phenomenon in hydrological processes especially runoff, infiltration and flood flow patterns has been studied. Field experiments for infiltration testing by using ring-infiltrometers were carried out corresponding recorded time and water depth. Laboratory testing was done for studying some properties of soil samples especially for infiltration speed through soil column, which were collected from the field. Due to the site of study is located in ungauged catchments with lack of for hydrological data but some had been generated using available data from other catchments nearby the site. The simple water balance model was applied to this study. Mapping represents flood flow patterns and distribution of infiltration rates. The results showed the potential for further study of flood reduction via deep percolation from floodplain.

View full abstract

It is necessary to understand the hydrological and pollutant cycle system of the watershed where a nature system element is intertwined to an artificial system element to promote the lake watershed management in the cooperation of labor. Additionally, we should have an objective tool which is appreciable of the effect of an each measure in the entire system.
We constructed a distributed pollutant flow model continuously to a distributed hydrological cycle model. And, we applied these models to the Lake Kasumigaura watershed which was a large area and general watershed. As a result, the pollutant flow process of the watershed was able to be understood for the unit of a standard grid cell. In addition, it was shown that the simulation tool which these constructed models cooperated was effective for the lake watershed management.

View full abstract

The potential for flood and sediment material is strongly affected by river meandering and changes in land use. Therefore the modelling pre-channelization and land use change are important with respect to prediction of flood, sediment yield and its on-site consequence. A combination of land use transformation maps and soil properties shows certain decision rules for the conservation of forest into arable or vice versa. The proposed model obtained from this study was used to simulate possible past and/or future pre-channelization and land use patterns. Consequently, the outcome of this simulation was employed to assess flood, sediment material and eroded soil risk under different scenarios. Pre-channelization and land use change in the Ishikari river basin were analyzed by comparing three scenarios. The results indicate that pre-channelization has a significant impact on flood peak discharge, but no significant effect on sediment yield. In contrast, land use change has no significant effect on flood and sediment yield for Ishikari river basin.

View full abstract

Generally, it is well known that rainfall-runoff phenomena belong to a random process. So, the runoff system is described by random differential equations. In this paper, I adopted Kinematic Wave model for tow slopes and a river channel and derived differential equations whose solutions provided the average and variance of discharge output from the runoff system under the condition of mutually independent rainfall input. The validity of proposed theoretical equations was cross-checked by a simulation. By using similar way, it is possible to estimate the higher order moments of discharge output and its probability density function which change with time.
The results of this study can be applied to the evaluation method of models based on its stochastic response characteristics.

View full abstract

The equivalent frequency transfer function (EFTF) has originally been used by control engineers to describe dynamics non-linear elements such as threshold saturation. Previous study by authors had derived EFTF between rainfall input and discharge using distributed parameter runoff models. Moreover, authors had pointed out that the obtained EFTF is approximated by the delay system. It is considered that the general description of channel flow is the dynamic wave equation. This paper makes clear that the dynamic wave equation is characterized by two non-dimensional parameters, K₁, K₂. In the case of K₁→∞, the dynamic wave equation approaches the diffusion wave equation. Moreover, the diffusion wave equation approaches the kinematic wave equation when K₂→∞. In this paper, authors focus on the diffusion and kinmatic wave equation. This paper finds the critical value of K₂ which divides the both equations based on parameters include in the delay system for the both equations.

View full abstract

Digital elevation models (DEM) are extensively used in hydrological analysis to obtain distributed hydrological attributes and the flow direction. Demerits are consistently hindering hydrologists despite the ease of using DEM with several existing methods. A new method of assigning flow directions based on flow tendency is proposed in this paper. Iterative search of continuous pits and flats are embedded inside. This method is named as Ranked Flow Tendency (RAFT) method. A storage function method is used to simulate the water movement taking the DEM of Kamishiiba Reservoir Site (210 km²) as the case. The results are compared against the conventional D8 method and modified D8 method for flow networks. The proposed RAFT method yields better flow path and is able to trace the reservoir sites. This method shows its better ability to involve more cells at off-peak wet region in flow routing.

View full abstract

We applied “water census database”, which we had estimated by a distributed model in standard grid cell, to the watershed management of Kasumigaura where watershed management has been proposed for its water quality improvement. We took up “sewage connection rate in local sewage area”as an item of watershed management, suggesting“E_{CR - 100/100P}” which means the effect of river water quality improvement by growing sewage connection rate as an incentive index to local people. Growing sewage connection rate from the cell where E_{CR -100/10OP} is larger in a same municipality, contributes to more effective improvement of river water quality in other cells. So that way is thought to be useful method of watershed management for getting local people being cooperative.

View full abstract

A lumped runoff model is developed. This model consists of two runoff components, i.e. river flow and subsurface flow in saturated zone. These water movements are represented by the basic differential equations derived from the hydraulic principles, such as the equations of unsteady flow and the Darcys law based on Dupuit-Forchheimer assumption. The identification procedure of the model parameters is presented and the estimation method of runoff hydrograph from the calibrated model is given. The model is applied in the experimental catchment (373.6 km²) of Kanna river basin, Kanto area, Japan. The results show the good agreement between estimated and observed flood flow and low flow hydrographs.

View full abstract

The purpose of this study is to clarify the nonlinearity of runoff phenomena and to understand hydraulic processes in mountainous basins. Universal lumped analysis method for runoff in mountainous basins is proposed in this study. This model can express the characteristics of nonlinearity of runoff and base-flow recession on small mountainous basins. Runoff in mountainous basins are uniquely determined by model parameters concerning slope gradient, slope length, the thickness of surface soil layer, unsaturated hydraulic conductivity and effective porosity.

View full abstract

Hysteresis effect is one of the soil characteristics, and it is important for groundwater analysis involving dynamic water movement. In this study, a 2D model based on Richards equation expressed on unsaturated infiltration theory with considered hysteresis effect, was developed. To discuss the model, numerical analysis and experiments were carried out on three conditions in the view of discharge and water balance.
As the analysis using the soil parameters measured by experiments, this model is more effective to simulate real water movement than the model without hysteresis effect. As these results, this model is useful to analyze 2D soil water movement and apply to 3D model.

View full abstract

This study is the foundation to predict a slope failure generation. The prediction of a slope failure needs the subsurface flow simulation. The purpose of this study is to propose a simplified two-dimensional infiltration model of landslide area, considering the unsaturated hydraulic conductivity and the slope of the impermeable layer, by which we can analyze the subsurface flow in the direction of the slope and the normal direction to the slope at the same time. To verify the proposed model, it is applied to the slope failure that actually occurred in Fukushima in August, 1998. The result is that the proposed model can almost accurately predict the time of slope failure generation, and this implies that the model has prospects of the prediction of a slope failure.

View full abstract

When discussing on the water budget in a basin scale with various land surface conditions, evapotranspiration is a key factor for the practitioners. It is known in forest hydrology that canopy interception, which is equal to the direct evaporation, accounts for around 20% of rainfall, while the transpiration through vegetation is affected by soil water content. However, rational method to interpret actual evapotranspiration is not sufficiently known to those who use the rainwater rechargeevapotranspiration discharge term in the quasi-three dimensional groundwater flow model. It is found that the proposed rainwater recharge-evapotranspiration discharge term functions properly, and the calculated water budget seems reasonable compared to the observation. On the other hand, however, it is also found that more examination is necessary for the initial loss of rainwater R₀ and the parameter H_g^* in the rainwater recharge model, which affect both groundwater recharge and actual evapotranspiration estimation.

View full abstract

Extensive field measurement of groundwater level and quality was carried out three times in 2002 at sixty five stations in Takamatsu area to clarify the effects of irrigation on the unconfined groundwater. Continuous measurement of groundwater level has also been made at seven stations in addition to the continuous measurement by the public sectors. It is found that the water level gradually increases in the irrigation period with the contours of water level running nearly parallel to those of topography. The rainfall and the pumping up have influence of the change in water level. The pH of groundwater is found to slightly increase in the irrigation season while Stiff diagrams and the cluster analysis demonstrate that the water quality can be classified into four groups and that the groundwater quality mostly remains unchanged. Nitrate nitrogen of unconfined water at a few locations has exceeded the environmental standard, which may have come from the overuse of fertilizer. Some wells in the coastal region have shown the effects of seawater intrusion.

View full abstract

When modelling quasi-3D groundwater flow in large area, the spatial distributions of unconfined and confined groundwater heads at each node are necessary. However, the heads of unconfined and confined aquifers are sampled at a limited number of sites whereas values at many un-sampled sites usually are necessary for modeling. On the other hand, the observed heads of the confined aquifer in Kofu basin are less than those of the unconfined aquifer. There is need to incorporate the unconfined head data as auxiliary information into the estimation of the confined head in space. The purpose of this study is to emphasize the supporting role of geostatistics-Residual Ordinary Kriging (ROK) and Universal Cokriging (UCK) in estimating the unconfined and confined heads in space from limited measurements in Kofu basin. Monthly field measurements of the unconfined and confined head in February of 1985 were collected and analyzed using ROK, UCK. The results indicate that accurate results can be obtained by using ROK and UCK.

View full abstract

The purpose of the present paper is to propose a new approach to the problem of identifying a model structure of groundwater flow with modeling errors to be complemented on the basis of groundwater hydraulics and geostatistical theory. The optimal parameters are identified to be in best agreement between the calculated and observed hydraulic heads. The head residual by all means causes an modeling error in computed parameters because the conventional groundwater model in use is of the physically based, deterministic type. An universal and new strategy to complement all errors at the observed points is now needed. The strategy also should take the spatio-temporal estimates of the errors at the rest of the given points. This paper theoretically describes a new approach and demonstrates its effectiveness in dealing with numerical studies in a hypothetical aquifer. Finally, a mathematical procedure to select the best model subsequent to improvement of fitting imperfection is used, and its validation is then discussed.

View full abstract