It is shown that a similarity parameter has new mathematical structure. Heretofore, the correlation coefficient is used for quantifying the correlation relationship between two ensemble members. Koster et al1 introduced a statistical parameter, called Ω to quantify the similarity among several ensemble members with calculating the ensemble numbers and the two types of variances. However the mathematical structure of Ω had not been revealed in their studies. The present authors applied to derivate Ω for understanding the mathematical meaning of it. As results, we could have a knowledge that Ω consists of mainly two terms. One is the average value of cross correlation coefficients (ACCC) across all ensemble members. Another is the similarity of the mean value and the variance across all ensemble members. Therfore, the authorscan conclude that Ω shows the similarity of the ‘shape’ of all ensemble members and the mathematical characteristics is more capacious than the correlation coefficient. The paper ends with some remarks on the mathematical characteristics of ‘as a new evaluation methodology for the predictabity of numerical forecast.’ in monthly or seasonal time scale.
With the recent occurrence of frequent droughts or floods exceeding conventional predictions, an existing assumption of stationary rainfall is no longer applicable to frequency analyses in many cases. On such a background, this paper reports the results of frequency analyses of non-stationary monthly, seasonal and annual rainfalls with a non-exceedance probability of the design level, a key index in discussing drought problems.
This study is conducted as part of the study on estimation of the peak discharge of “the Yokota-Gire Flood” in July 1896, one of the historical floods in the Shinano River Basin. This study analyses relationship between rainfalls and discharges of relatively large floods since 1956 with well recorded rainfalls and flood discharges, and particularly compares the Yokota-Gire Flood with the flood in September 1982, in which the similar amount of rainfall was recorded despite the difference of the number of precipitation stations exists. Making clear the characteristics of hourly and spatial distribution of the rainfalls which caused large floods in the Shinano River Basin, not only helped to improve the accuracy of the peak discharge estimation of the Yokota-Gire Flood and to verify its validity, but also clarified the statistical positioning of the Yokota-Gire Flood through rainfall characteristics.
This paper first reports the current situation of the IDF (Intensity-Duration -Frequency) equations used in all the local prefectural governments in Japan, based on the authors' questionnaire responded from all the forty-seven prefectures in summer 2004. The responses to the questionnaire showed that the traditional methods for IDF equations are still widely used, regardless of the recent remarkable developments of data processing including computer technologies, frequency analysis methods and accumulation of systematic meteorological and hydrological data. Based on the results of the questionnaire and rainfall data provided by a number of prefectures, this paper also discusses how to select probability distribution functions and IDF equations from their candidate ones.
This paper describes relationships between Niño 3 sea water temperature (SST) and statistical characteristics of daily rainfalls in Tokyo. Daily rainfall series for 1961-2002 is separated the series in the years with positive SST-anomalies from the one in the years with negative SST-anomalies. The frequency analysis based on partial duration series is carried out individually. The quantiles in the years with negative SST-anomalies are 1.68-2.35 times as large as the ones in the years with the positive SST-anomalies, indicating that the heavy rainfall events occur frequently in the years with negative SST-anomalies. Despite the condition of no artificial factor in the climate changes such as a global warming, the numerical simulation results using the ocean-atmosphere coupled general circulation model also show that the daily rainfall series in Tokyo is unstationary.
In recent years, global warming is aggravated and is feared to cause climate change. The IPCC reportsthat the anthropogenic CO2 has affected the steep global warming. Therefore, many researches have been carried out about CO2. But there are many unclarified factors in CO2 behavior. So the fluctuation of atmospheric CO2 concentration is not grasped yet. In this paper, a time series analysis has been carriedout using observation data. And we have aimed at getting a hold of the fluctuation characteristics of atmospheric CO2 concentration and clarifying factors affecting it. In result, we confirmed quantitatively the cause of seasonal cycle in atmospheric CO2 concentration. Moreover, we discussed effect of SOI, solar activity and economic activity on unclarified long periodic behavior.
The present paper describes the heat, moisture and salt transfer in a soil column with a salt affected layer due to watering-evaporatiocny cles, e specially the leaching and the re-accumulation of soil salinity at the soil surface under different watering conditions. The experiment was carried out by changing the watering frequency and the amount of each watering under the same total amount of watering. A fter watering on the soil surfacec overed by a salt affectedl ayer, the time rate of the re-accumulationo f soil salinityi n the surfacel ayer is large but the durationo f the re-accumulationis short, when the amount of watering is small. On the other hand, after watering on the salt-free soil layer, on which lies an intermediate salt affected layer, the re-accumulation of soil salinity in the surface layer becomes more distinct as the amount of watering and the watering times increase. The present experimental resultsimply the importance of the water management to prevent the second salinization.
For an assessment of how hydrologic processes influence the location and timing of landslide or to assess how landslide risk might change in response to changed landuse pattern, some sort of mathematical model is needed, which can incorporate all triggering factors explicitly. Proposed landslide model explicitly analyzes the effect of high-intensity and short-duration rainfalls on development of pore water pressure. Richards equation is used to evaluate the pore waterpressure response to the changing rainfall pattern and attendant change in Safety Factor of the stability of a location. The numerical solution of the unsteady pore waterpressure is experimentally verified in lab. The model is able to predict the timing and location of the shallow landslides. It is used to simulate the landslides occurred in 1973/10/26 at the Takora basin located upstream of Kizu river basin. The result of model encourages for development of more generalized model in future incorporating more factors that influence landslide.
When heavy rains fall, the landslides frequently occur in valley heads. These reasons are that underground water concentrates easily and piping occurs because of groundwater discharge. This study is the basis to predict a landslide generation in valley heads. The prediction of a landslide in valley heads needs the 3D subsurface flow simulation. The purpose of this study is, first, to analyze subsurface flow in 3D considering land form, soil layers and pipe flow, and second, to evaluate the landslide potential in valley heads. We targeted the landslide that actually occurred in Fukushima in August, 1998. The result is that our model can almost accurately predict the time of landslide and piping generation, and there is a possibility that as the history of heavy rain continues, piping grows and landslide potential increases in valley heads.
The effects on slope failure due to seepage flow under the heavy rain are investigated numerically. The numerical modelling simulates the interacting overland flow and groundwater flow. The kinematic wave equation is used to describe the overland flow on the slope. The groundwater flow in sloped soil layer, as governed by the saturated-unsaturated equation, is considered in the model. As the results, in case of concave slope, by accumulating the rain water into lower side of slope, the region of saturated throughflow is increased, and return flow leads to fluidisation of surface layer of slope. These features of groundwater flow have implication for large-scale of slope failure.
It is important to know the characteristics of temporal change and spatial distribution of water content and infiltration rate of surface soil layer for remote sensing and unsaturated flow analysis. In this paper, we examined the water content change caused by rainfall and infiltration at 24 points (depth=0-10cm) surrounding a single tree by reflectometers. Results showed that the water content.Θ at the points near the stem had bigger average and smaller standard deviation than that at distant points. The relationship between Θ and its differential calculus immediately after rainfall differed in different water content. At lower content dΘ/dt was proportional to (Θ-Θr) n suggesting gradient of hydraulic head is 1. At higher content, the absolute value of dΘ/dt was bigger than that estimated by dΘ/dt=k (Θ-Θr) n. At the highest content, dΘ/dt suggested that the water was supplied from tree to surface soil after rainfall or that thegradient of hydraulic head is smaller than 1 because of high water content below 10 cm layer. The surface supply water estimated by using dΘ/dt=k (Θ-Θr) n corresponded with rainfall intensity but there is a time lag between them.
Change in the land use induces the reduction of rainwater infiltration and the increase of surface runoff. It is therefore important to minimize the influence of land use alteration on the infiltration and river discharge. In this research, the facilities of rainwater storage and infiltration for groundwater conservation was planned at the University of Kyushu, and the time series of infiltration of the facilities was calculated by the set of the water balance equations. From this study, the ability of the facilities for rainwater infiltration and flood control was proven. It is expected that the present approach could be useful for designing the facilities of rainwater and infiltration.
The tendency in river management is to become more integrated in fluvial fan. Especially, the interaction of the surface and subsurface flow should be considered. The pool-riffle-pool sequences in a river with alternate sandbars govern the surface water profile and also subsurface flow in the river. Surface water flow usually changes near the alternate sandbar with high permeability due to the interaction of surface and subsurface flow. The study on the relationship of the upstream total discharge, the required minimum discharge for certain purpose (such as fishery, navigation), and the amount of leakage from surface to subsurface becomes very significant in the river management. As the initial framework, this problem is discussed in a river reach scale in this study. The required environmental discharge is assumed to be the minimum surface flow, and the relationship between leaking flux and the required upstream total discharge is discussed based on the result of flume experiment.
In Sarobetsu Mire, the groundwater level has been dropping for the past few decades. There is great concern regarding the increase in sasa bamboo distribution and the displacement of the original vegetation. In the region of the mire, snow plays an important role in stabilizing the ground water level, especially in the bog area, because it recharges abundant amount of water to the groundwater. Observation data showed that the ground water level in the bog area is kept high in winter by meltwater recharge from the bottom of the accumulated snow. An attempt was made to analyze the effect of snowmelt on groundwater level variation using a two-dimensional unsteady model. The simulation results indicated that lowering of groundwater level in the bog area could strongly relate to a decrease in snowmelt water. This means that mire conservation must take into account not only artificial influences like reclamation but also climate influences.
Continuous measurement at four stations in Takamatsu plain demonstrates that the confined groundwater levels in the central part of the plain has been recovering over the last decade. The present study aims to clarify its mechanism to provide fundamental information for the water resources. development Two-dimensional numerical simulation is carried out for the confined groundwater using the estimated groundwater withdrawals, the measured unconfined groundwater levels and the interpolated hydrogeologic parameters. The numerical results are found to well reproduce the trends in water level at four stations as well as the water levels at other points that were measured this year. The increase in groundwater level can be explained by the reduction in pumpage by some factories located in the central part of the plain.
The floodplain in Phichit province, Thailand always has flood in rainy season and water lacking in dry season. This area is a major recharge area of Thailand and farmers set many wells for drawing groundwater, because of insufficient surface water during two times of paddy crop and unavailable irrigation systems. This causes severe decreasing of groundwater level with no sign of recovery. In inundation area, the average flood depth is about 2m and has duration 1-2 month. The authors set up the observation systems in this study area to monitor their daily change in groundwater level, river water level, and rainfall. Flood water in this area recharges shallow aquifer and continuously flows down to deep aquifers. This paper presents the simulation of groundwater budget in this area based on the Tank model concept using data in 2002-2004. The results show that recharge through ground surface by flood is about 55% of annual recharge and 73% of annual recharge flow down to deeper aquifer.
Effect of sea level fluctuations on the movement of the freshwater saltwater interface was analyzed by means of a sharp interface model. The simulation was carried out over 100 years period for two case studies based on available data; western coast of America and the Bay of Bengal and adjacent continental shelf. The position of freshwater-saltwater interface was estimated and it reflects the effect of long term average sea level rise on the position of interface. Using the location of the interface, the related loss of freshwater resources was estimated in both areas and results show that volumetric freshwater losses due to sea level rise is 1%- 5% of the aquifer volume in western American coast and 1%-3% in Bay of Bengal. The effects and influences of the loss of fresh groundwater resources were discussed, considering the groundwater use and groundwater recharge and adaptation proposals were introduced for both areas.
The seawater intrusion in coastal aquifers has long been attracting attention of researchers for the management of coastal water resource and environmental protection. As the saltwater intrudes into the freshwater aquifers, the quality of salt groundwater will depend on the length of its residence time in aquifers and the geo-chemical composition of the aquifer material. This paper presents the estimation of residence time of freshwater and saltwater in the coastal aquifers and their characteristics. This study utilized the data from the borehole taken from Saitozaki, Fukuoka City. The method of stream function was applied in evaluating the residence time both in the saltwater and freshwater. It was found out that the residence time at saltwater region is longer than that in the freshwater region.
This study shows that the distributed model of permeabilities in geostatistics is proposed and uniquely solved in the numerical test. A hypothetical aquifer divided into twenty piecewise zones is designed. Twenty constant-permeabilities are individually identified and some of them result in inaccurate solution. The difficulties of an inverse problem with ill-posedness prevent us from getting the stable, unique and accurate solutions of identified permeabilities. A distribution model of permeabilities mapped through geostatistics is herein proposed. A sensitivity of observation updating to the parameters to be identified is then enhanced over the whole area. The mathematical foundation for the advances of the sensitivity is also investigated and discussed. It can be proven that the proposed approach is greatly helpful in identifying a spatial distribution of piecewise permeabilities.
Three models were compared on simulating NO3--N stream water solute concentration and estimating annual NO3--N load from a forested watershed. These models are (1) tank-model C-Q method, (2) tank-model concentration coefficient method, and (3) solute runoff tank-model. The models based on the same rainfall runoff tank-model developed by authors, have different methods for calculating stream-water NO3--N concentration. As a result of applying each model to the data observed in the experimental forested watershed, solute runoff tank-model was most excellent in the reproducibility of the seasonal variation of stream water solute concentration. However, the remarkable difference in calculation result of annual runoff load in each model did not appear. It was cleared that the accuracy of rainfall runoff model is the most important in calculation of runoff load from a forested watershed.
An applicability of quinone biomarker as tracer to the analysis of hillslope runoff was investigated. At first, vertical quinone profile in the under and central parts of a hillslope was examined. Quinone content, which corresponds to the amount of bacteria, decreased with depth in both parts. The predominant ubiquinone (Q) and menaquinone (MK) species were Q-10 and MK-10 (H4), respectively. Then, the dynamic changes in quinone species included in three different types of runoff under different rainfall events and ground water levels were examined. The three different runoff pathways brought distinct quinone profile. Consequently, the behavior of specific quinone species in the runoff was evaluated with aid of the vertical quinone profile of the hillslope. Q-10, which existed aspredominant quinone species in the hillslope, probably becomes an index for runoff through a surface soil layer in the hillslope. Furthermore, Q-8 and Q-11 are expected to represent base flow and runoff through a deep soil layer, respectively.
Sayanokami spring water exists in the new campus of Kyushu University, which is located on the western part of Fukuoka City. This spring water is one of the important water resources for agriculture around the new campus area. The construction of the new campus started last June, 2000. In this study, to understand the hydrological properties of the Sayanokami circumference, 222Rn and tritium were used as tracers, respectively. Consequently, it was found out that the residence time of Sayanokami spring water is 10-20 years. Moreover, the residence time and the catchment area of Sayanokami spring water were estimated using groundwater flow model. As a result, the residence time of Sayanokami spring water was evaluated to be about 25 years at most and the catchment area was specified. It is concluded that the accuracy of estimation of residence time and catchment area of the spring water can be improved by analysis of radioactive isotopes and groundwater flow model.
It is known that observed macroscopic dispresivity depends on observation scale, so that a single value cannot represent the field property for dispersion. In this study, numerical simulations of the advection dispersion phenomena in various random-permeability fields generated by using stochastic fractal model were carried out to investigate the relationship between spatial distributions of hydraulic conductivity and dispersivity. The results show that various concentration distributions occur even when the hydraulic conductivities of flow fields are statistically the same. It is also shown that a dispesivity distribution model expressed by function of scale can well reproduce the ensemble mean of concentration distributions.
Groundwater quality is affected by the various reactions such as precipitation of dissolved ions, solid-liquid phase interaction, and the bacteria mediated reduction-oxidation process. The geochemical properties in coastal aquifers are unique and important for the management of groundwater environment. Although the studies analyzing the hydrological behaviors have been intensively done up to now, a study of coupling both groundwater flow and geochemical analysis is limited. In this study, the one-dimensional multicomponent solute transport model which accounts for the cation exchange reaction and the reduction reaction induced by anaerobic degradation is studied. It is found out through both the experiment and numerical simulation that the cation exchange process is significant at the frontal part of the seawater infiltration, while reducing process by the bacteria mediation takes places at the entire soil column except for the top ofthe soil where dissolved oxygen infiltrates.
This paper evaluated the results of a field test for grou ndwater remediation by in-situ bioremediation at a site contaminated by tetrachloroethylene. The technique carried out in this test uses reductive dechlorination of contaminant in a subsurface environment. In the observation well during the field test, the percentages of daughter products like dichloroethylenes were observed to increase with the rise in reductive condition of the aquifer. Moreover, it was shown that the first order rate constants of all substances became greater than those of natural groundwater by more than a few orders. This meant that the reductive dechlorinataions could be accelerated during the field test.
Arsenic contamination had been found all over Bangladesh in 1993, however, the groundwater flow and the mechanism of arsenic contamination have not been made clear yet. Because the arsenic contamination spreads through the groundwater, it is important to understand the groundwater flow. We performed the field measurement from December 23rd to 25th, 2003 and from May 19th to 21st, 2004. Then we performed the numerical analysis of groundwater flow using MODFLOW to grasp the field groundwater condition. As a result of the field measurement, the underground condition of our study area is reductive, and we consider the arsenic elutes into groundwater through a reductive mechanism in a dry season. Besides, using MODFLOW, we could reproduce the groundwater flow properly.
The purpose of the present study is to clarify formation process of the water quality hydrograph. We obtained two ordinary equations to approximately and conveniently express the hydrographs of discharge and water quality. The characteristics of water quality hydrograph observed in mountainous basins can be explained rationally by the method. The method to estimate generation function in formation process of water quality hydrograph using inverse estimation from observed water quality concentration, discharge and rainfall data is newly proposed in this paper. A method is applied to several water quality hydrograph at three different mountainous basins. The result of estimation show eduction and re-suspension is generated from rainfall and influenced by initial condition of water quality concentration.
The Kalman filter theory is coupled with a distributed hydrological model to update spatially distributed state variables by using several techniques proposed here. To acquire the total water storages of a basin from discharge observations at the outlet, a Q-S curve is used as an observation equation. After updating total water storage with the Kalman filter, the ratio method is introduced to reset the distributed storage amount of a basin, maintaining the spatially distributed pattern. A Monte Carlo simulation is adopted to predict state variables and error variance propagations. A distributed model coupled with the Kalman filter theory gives updated simulation results with improved forecasting accuracy.
The purpose of present study is to clarify runoff mechanism and to understand hydraulic processes in mountainous basins. Considering multilayer flow in a slope, generation of overland flow for large scale flood is important. The method of multilayer runoff analysis which can calculate overland flow, subsurface flow and vertical infiltration flow based on soil property and rainfall form is newly proposed in this paper. Characteristics of overland flow in mountainous basins can be expressed rationally using our method. A method is applied to the Kusaki dam basin (254km2). The results of runoff analysis match well with the observed data in Kusaki dam basin and overland flow is expressed for peak time in a large scale flood.
Rainfall-runoff phenomena generally belong to a random process. Thus, a rainfall runoff system is described by random differential equations. However, the stochastic response characteristics of rainfall runoff systems for a larger basin are not clear. In this paper, the Kinematic Wave model is used as a rainfall runoff system, derivation of theoretical differential equations to calculate the first-and second-order moments of discharge is shown. The validity of derived equations is shown by the results of simulation. By using a similar way, estimation of the higher-order moments of discharge output and its probability density function that change with time is possible. The results of this study can be applicable to the evaluation method of models based on their stochastic response characteristics.
There are several commercial river water simulation models, with the progress of mathematical techniques for hydraulic calculation in open channel and pipeline. As a result, it is getting easier for engineers to model a river basin as a distributed model. But the problems are still left in application, such as, which is suitable for rainfall-runoff model for sub-basins, how to determine parameters with insufficient data, the lack of detailed cross sections of river beds and so on. With a view to promote the applicability of the distributed model, the authors applied it to the Japanese mountainous river basin for continuous simulation including dry weather condition. The countermeasure for above problems are the application of SCS method for surface runoff model, the adoption of new interflow model, the conversion of hourly rainfall data into 5 minutes intervals, and the presumed triangular river cross sections and falling works. Based on these assumptions, the simulation results fairly agreed with the observed data. It means basic concept of this study is in the right way and applicable to any other basins.
The purpose of this study is to understand effect of different litter layers on water storage. An experimental system with rainfall variation expresses the storage process through measurements of discharge and soil moisture. Also 2D Richards model with hysteresis and the experimental results show the permeability of different litters, broadleaf, acicular, and laurel. The litter layers increases the storage effect as slope angle increases. Broadleaf and laurel have larger storage capacity than acicular. Application of the model with hysteresis is approximated to survey data. In Richards equation with hysteresis, The vertical water permeability of a litter layer is expressed in the half permeable of the soil.
Stormwater detention facilities have been used to reduce flood runoff from urbanized areas. They are recognized as means to reduce peak flows of downstream rivers, and it is needed that the planning of them are done on a watershed-wide basis. In this paper a method to estimate the flood control effects of detention facilities on a watershed-wide basis, and the results of its application to Tsurumi river basin are presented. We could estimate the flood control effects of detention facilities by comparing the hydrographs calculated by the present method for Tsurumi river basin with and without detention basins respectively. It is found that there was little effect on flood control of the small scale detention facilities.
Recently, Geographic information data such as Digital Elevation Model is available in runoff analysis because of improvement of GIS (Geographic information system). In particular, various hydrological models have made a proposal based on DEM. However, there has been least study on land use model that is the origin of flow generation. Subcathments in slope element runoff model or grids in distributed hydrological model are subdivided into impervious area and pervious area from land-use data. Effective Precipitation is calculated based on them as the specific value in the region. Using the land-use data, there is a problem that the position of runoff control equipment doesn't reflect the runoff model in detail. It is unable to set Equipment characteristic value (discharge point, equipment scale), because land-use model and equipment dose not have one to one correspondence. In this study, hydrological model using polygon format data to express land-use is proposed, and applied to Kotta River basin. The simulated hydrograph show relatively good agreement with the observation.
The objective of this study is to calculate the maximum discharge distribution corresponding to the distribution of annual maximum daily rainfall by using random cascade model.This model is used to probabilistically disaggregate daily rainfall generated according to its probabilistic distribution, then maximum discharge is calculated from disaggregated hyetographs by using a runoff model. This technique is applied to the Doki river basin, a small basin, where the influence of human activities is very limited. The observed discharge of annual maximum from 1976 to 1998 was used to verify the calculated distribution of maximum discharge. It is shown that calculated distribution of maximum discharge reproduced well. Furthermore, the cover ratio of the conventional method to compute design flood is evaluated.
We propose a methodology to identify prediction uncertainty through recognizing and quantifying the different uncertainty sources in a hydrologic model. Statistical second moment is used as a measure of uncertainty; also an index which originated from Nash coefficient of efficiency named Model Structure Indicating Index (MSII) is proposed to quantify model structure uncertainty. The results show that MSII can well reflect the goodness of model structure, while a larger value of MSII indicating a poorer structure of hydrologic model. The index can be used as a tool for implementing model quantitative comparison (selection).
Development of a transferable rainfall runoff model and transferable model parameter values seeks connections among physical processes at disparate scales and possible linkage of hydrological similarities between catchments, which makes scale transformation and scale invariance as the fundamental requirement. We successfully transferred 50m DEM resolution TOPMODEL parameters identified at Kamishiiba catchment (210 km2, Japan) to Kuwanouchi catchment (187 km2, Japan) and Balaphi catchment (650 km2, Nepal) by downscaling the topographic index distribution for the later catchments from coarse resolution DEM with 1 km grid size to target resolution DEM with 50m grid scale. This leads to the conclusion that the method of transferring scale independent hydrological relationships can serve as a potential tool for regionalization of parameters and for predicting ungauged basins.
The present study investigates effects of grid size of DEMs on hydrograph applying distributed runoff model to five catchments with different scales. This work is carried out in order to give some information to manipulate spatial data and to acquire the reliable result in applications of grid-based model. The variations of goodness-fit-index and runoff ratio are presented for geomorphologic resolution and catchment scale. It is showed that the fine resolution is appropriate for application to small catchment and the application with different grid sizes mostly meets with satisfactory results.
For comprehensive water management and water environmental planning in large river catchment areas, the longterm hydrologic cycle need to be understood. In cold, snowy regions, the snow process (falling, accumulation, and melting) plays an important role on the hydrologic cycle. An understanding of snow cover volume and snowmelt volume is essential. Analysis of the water balance of the catchment area also requires accurate estimation of evapotranspiration, which is affected by soil and vegetation. This report applies a two-layer model toward estimating the evapotranspiration, snow cover volume, and snowmelt volume of 1-km meshes. The model incorporates heat balance between the air, the vegetation layer, and the ground surface. Observed values for evapotranspiration, snow cover volume, and snowmelt volume were compared with estimates generated by the model. These estimates were used to reproduce long-term runoff, in order to examine the validity of the estimates. The estimates were found to accurately reproduce the measurements. This has enabled the quantification of long-term hydrologic factors for cold, snowy regions.
Supplementary water should be estimated even in areas where data is scarce because of the importance of proper water management. The Mekong River, particularly the lower part (Lower Mekong River Basin), is one area where supplementary water has not been well researched despite the fact that this international river runs through four countries: Laos, Thailand, Cambodia and Vietnam. Studies to grasp the amount of supplementary water are restricted due to a lack of hydrological data, thus methods for obtaining these amounts are needed. One solution is model simulations. In this study, supplementary water was estimated by calculating back from rice production targeted at a macro scale. As a result, the supplementary water in Cambodia, Laos and Thailand was estimated from 1991 to 1995 through model calibration with rice production.
In this study, complicated atmospheric patterns in a rainy season (BAIU) in the south-western Japan was classified into eight groups, using the Self-Organizing Map (SOM) algorithm, which converts complex nonlinear features into simple two-dimensional relationships. The groups can be basically represented by five meteorological fields;(1) dry air masses, (2) anti-cyclonic flow due to the Pacific high pressure, (3) the existence of the BAIU front between dry and wet regions, (4) the intrusion of a large amount of water vapor, (5) passages of typhoon and low pressure system. One of the groups has notable feature represented by high precipitable water accompanied by strong wind components (Low Level Jet), which is a typical meteorological field that causes disastrous heavy rainfall events in the northern Kyushu. Therefore, it may be expected that the classification of meteorological fields and associated extraction of heavy rainfall phases contribute to determining whether heavy rainfall occurs or not in a target area as well as the enhancement of the accuracy for the rainfall prediction.
Taking account of variation of water vapor content and temperature in an air parcel, a new model for the size estimation of the cloud droplets condensed on composite CCN, whose constituents are waterinsoluble dust of kernel and water-soluble substance of coat, has been developed. New model is constructed by coupling chemical potential equilibrium with mass and heat conservation equations. Variations of the droplet size and the individual and populational amounts of droplet with the dust size are simulated numerically. The numerical results show that the larger dust size is, the more the droplet size and the condensation amounts increase.
In order to downscale spatial rainfall field from a coarse scale into a finer one, the non-homogenous multiplicative random cascade method is often employed. Currently, this kind of downscaling method is less reliable even though it correctly preserves the statistical description and a long term average spatial pattern. It fails reproducing the spatial patterns in repeated trials; and there is a higher chance of having random outputs. These drawbacks are needed to overcome for applying the downscaled rainfall products in further analysis of hydrological and meteorological studies. In this study, a new method, named as Multiplicative Random Cascade Hierarchical and Statistical Adjustment (HSA) method, is introduced and tested to downscale 1.25 degree GAME Re-analysis data into 10-minute spatial resolution. The obtained results are highly improved, quite robust and reliable than the previous method.
In this study, the effect of land cover and thermally induced local circulation on convective precipitation in mountainous region around the Lake Biwa is investigated by numerical simulations. Firstly, numerical simulations, which have realistic land cover, are carried out on 15th Aug. 2001 (precipitation was observed around the Lake Biwa) and 14th Aug. 2001 (precipitation was not observed around the Lake Biwa) and it is found that both local circulation and atmospheric environment are important for the precipitation. Secondly, some numerical simulations which has imaginary land cover (forest is changed into paddy field) are carried out on 15th Aug. 2001 and it is found that transportation of water vapor by local circulation induced by heating from the land surface is more important for the precipitation than evaporation from the land surface.
The purpose of this study is to clarify precipitation-topography relationships (PTRs) in mountainous regions on various spatiotemporal scales and to determine the physical mechanism of them. By running numerical simulations under realistic conditions using mesoscale model MM5, the PTRs in the Kinki region of Japan were investigated. PTRs on a spatial scale of a mountain slope were found to be described with a Gaussian-type function, which is referred to as GRIM (Gaussian-functional Relationship on an Isolated Mountain). And one of the conditions required for the formation of GRIM was determined: GRIM can be formed when regional average rainfall is more than 10 to 20 mm on a mountain slope. The spatiotemporal fluctuations of GRIM were, then, investigated to clarify how GRIM is connected with large-scale PTRs. As a result, large-scale PTRs were found to be formed by the summation of GRIM's. And it was clarified how large-scale PTRs vary according to the fluctuation properties of GRIM.
Monthly rainfall estimates by two spaceborne microwave sensors TRMM/TMI and TRMM/PR are compared. The originality of this study is that the difference between them is decomposed into retrieval error and rain/no-rain classification (RNC) error. Overestimation by the standard algorithm for TRMM/TMI against the standard algorithm for TRMM/PR has been known, but the difference decreased after the version-up of both algorithms. Still, it should be noted that retrieval error shows larger overestimation than total difference. Our developed RNC methods can replace the original RNC method of the standard algorithm for TMI. It is discussed how to set the global parameter k0 in our methods.
GPM project has a future plan of rainfall measurement with the core satellite loading dual frequency precipitation radars, accompanied by about eight companion satellites that also mount the microwave radiometers. Microwave radiometer data are expected to be the central data for producing the global maps of rainfall rates for the reason that they are used for the observation from satellite more frequently and have the wider observation swath width. In order to improve the rain retrieval algorithm for space-borne microwave radiometers, it is imperative to improve the microphysics model and the rain distribution models that are used in the retrieval algorithms. One of the important observations was simultaneous measurements of precipitation with COBRA, 400MHz WPR and ground-based measurements of rain drop size distribution. Therefore, we analyze the vertical characteristics of rainfall related to rainfall type observed by the COBRA and the 2D-Video-Distrometer. As a result, in convective rainfall, vertical variations of the polarimetric parameter (ZDR and Do) were very large. Height dependency of the vertical profile of radar reflectivity was become clear.
The objective of this study is to develop a new downscaling approach, which can take into account atmospheric and land surface heterogeneities for a better precipitation prediction. We address the effect of land surface heterogeneity on the land-atmosphere interactions through coupling a land data assimilation system with a land-atmosphere coupled model. This system relies on a mesoscale model as atmospheric part, a Land Surface Scheme as a model operator, a Radiative Transfer Model as an observation operator, satellite data and the Simulated Annealing method for minimization. To assess the effectiveness of the new system, a 2-dimensional numerical experiment was carried out in a mesoscale area of the Tibetan Plateau. The results showed significant differences compared with standard regional atmospheric model outputs and were more consistent with satellite microwave brightness temperature observations that improved the spatial distribution of soil moisture, which strongly affected the convection systems.
In this paper, an 1-D Cloud Microphysics Data Assimilation System (CMDAS) is developed for retrieving reasonable cloud distribution. The general framework of CMDAS includes the Kesslerwarm-rain cloud microphysics scheme, a microwave radiative transfer model in the atmosphere and a global minimization method of Shuffled Complex Evolution (SCE). The paper investigates potentials of the CMDAS to modify the cloud properties by considering integrated cloud liquid water content as a assimilation parameter and to introduce the heterogeneity into the initial state of the atmosphere, by applying the CMDAS to the satellite microwave radiometer data set obtained by the international cooperative observation experiment, “Wakasa Bay Experiment 2003”, in Japan. The simulated microwave brightness temperature by CMDAS is in good qualitative and quantitative agreement with the observed one.
Effect of forest on snow is evaluated with a proposed snow model. Wind speed and precipitation intensity and the forest condition (density of the forest and leaf) are acquired with the input data of the snow model. The snow model consists of the snowfall model and snowmelt model. The snowfall process and the amount of the snowcap are evaluated by the snowfall model. The amount of the snowmelt is estimated by the heat balance method in the snowmelt model. These models applied to various conditions (various wind speed, various LAI (Leaf Area Index)), and we can see that low wind speed deposits snow at the front of forest before higher one deposits it at the back of forest.