H31B-0352
Combined Radar and Radiometer Analysis of Precipitation Over Land
Passive microwave imagers have long been used to detect rainfall over the world's oceans. Over land, passive microwave imagers have historically relied on the use of high frequencies (i.e., 85 GHz) to detect ice scattering signals and relate this scattering signal to a surface rainfall. Errors with this retrieval are caused by inexact retrievals of the ice water paths, the relationship between ice water path and surface rainfall and by the parallax effect - a geometric displacement between the ice scattering and surface rainfall introduced by the viewing geometry of microwave sensor. This study analyzed precipitation over the southeast United States in the summer months of 1998 to 2000, using the TRMM Microwave Imager (TMI) and Precipitation (PR) on board the Tropical Rainfall Measuring Mission (TRMM). Correlations between 85-GHz brightness temperatures (Tb) and PR-derived surface rainfall, without accounting for the parallax effect, ranged from -0.16 to -0.47 for individual months. By accounting for the parallax effect and retrieving a surface rain rate directly below the ice layer that is being observed, the correlations with Tb 85 improved slightly to the -0.23 to -0.52 range. Further analysis of the ice scattering signal showed that stratiform precipitation with moderate ice scattering had improved correlations with rainfall rate when observed 37-GHz Tbs were used instead of 85 GHz. Correlations between Tb37 and surface rainfall ranged from -0.36 to -0.71, while Tb 85 versus surface rainfall had correlations that ranged from -0.22 to - 0.47 in this rainfall category. This was a result of 37-GHz Tbs being influenced more directly by scattering and emission from the liquid layer. Correlation between the IWP, as derived from the TRMM radar, and the 85-GHz Tb depressions ranged from -0.47 to -0.72. This indicates that significant improvements in rainfall retrievals can still be achieved from either an improved understanding of the ice physics (needed to improve the above correlations) as well as improved understanding of the relationship between ice water aloft and surface rainfall in different meteorological regimes. To further explore the relation between radar derived IWP and radiometer signals, ice water paths were examined with a combination of both radar and radiometer observations. Specifically, different ice density and ice particle number concentrations were examined on the radar inferred ice water path. Stratiform precipitation showed good agreement with both radar and radiometer observations when relatively low density ice (e.g.. snow) particles were assumed. For convective pixels with significant ice scattering, significantly denser ice particles were generally required. Moreover, liquid water above the freezing level had to be added in over 50% of the profiles before a solution that was consistent with both radar and radiometer observations could be constructed. These results appear consistent with current cloud physics understanding.
H31B-0353
An Explicit Microphysical Model for the Transient Evolution of the Vertical Structure of Warm Stratiform Rain: Application to Rainfall Radar Estimation
A rainshaft model of the stochastic advection equation with explicit representation of microphysics (coalescence, collisional breakup, condensation, and evaporation) is presented here as a dynamic simulator of raindrop size distributions in active storm systems (DSD) for use in physically-based retrieval of radar rainfall. The model was used to simulate the evolution of a stratiform rainfall observed event during TWP-ICE (Tropical Warm Pool – International Cloud Experiment). Transient DSD based on vertically Pointing Radars (VPR) retrieval, were imposed as realistic boundary conditions at the top of the column. The transient evolution of simulated vertical profiles of integral parameters such as drop number concentration, liquid water content (LWC), rainrate (RNT) and radar reflectivity factor (Z) were compared against Vertically Pointing Radars (VPR) estimates and ground based observations (Joss-Waldvogel Disdrometer). Results obtained suggest that the model is able to emulate with high accuracy the temporal and spatial evolution of the Reflectivity factor and with an acceptable level of confidence the evolution of the Rainrate and Liquid Water Content. More importantly the cross-comparison of modeling results with VPR and JWD observations showed that the model was able to significantly improve the description of the rain event for the lower level of the atmosphere where estimates provided VPR are not reliable. Recent progress including the extension of the one dimensional model to ice formation/melting mechanisms and aerosols processes will be also discussed.
H31B-0354
The drop-like nature of rain and its invariant statistical properties
We look for statistically invariant properties of the sequences of inter drop time intervals and drop diameters. We provide evidences that these invariant properties are 1) Large inter drop time intervals, ≥10s, separate drop of small diameter, ≤0.6mm. 2) The rainfall phenomenon has 2 phases; a quiescent phase whose contribution to the total cumulated flux is virtually null, and an active, non-quiescent, phase responsible for the bulk of the precipitated volume. 3) The probability density function of inter drop time intervals has a power law scaling regime in the range 1min-1h. 4) Once the moving average and moving standard deviation are removed from the sequence of drop diameters, an invariant shape emerges for the probability density function of drop diameters during active phases.
H31B-0355
The Environmental Parameters That Influence Simulated Convective Storm Precipitation: Results From a Large Parameter Space Study
Relationships between environmental conditions and the precipitation characteristics of simulated storms in a large eight-dimensional parameter space study are investigated. Specifically, we explore which environmental parameters are associated with the greatest production of rain and hail, both aloft and near the surface. The rainwater and hail mixing ratios for each experiment are averaged during the second hour of 2 h simulations to assess the precipitation production of each storm. Multiple linear regressions of the mixing ratios to the environmental parameters are then performed, to identify the parameters most correlated to the production of rain and hail. Although our results do not characterize the amount of point rainfall that would be observed from a particular storm, relationships between storm rainfall and properties of the environmental profile are found. The environmental temperature, related to atmospheric precipitable water (PW), exerts considerable influence over the precipitation characteristics of the simulated storms. When atmospheric PW increases, storms generally produce more rain and hail aloft, but less hail is observed near the surface owing to the increased depth of the melting layer. The ambient shear profile also affects the amount of rainfall, since general storm morphology and evolution are affected by the environmental wind shear. The level of free convection (LFC) and vertical distribution of buoyancy have lesser influence. Rain and hail production exhibit reduced predictability (in a linear regression sense) compared to other storm properties, such as updraft intensity. These findings demonstrate the degree of influence that environmental conditions can have on convective storm organization and may offer insight into the difficulties associated with forecasting convective rainfall. http://space.hsv.usra.edu/COMPASS/
H31B-0356
Stochastic Simulation of Daily Rainfall for Flood Risk Assessment Using a Mixed Distribution
Stochastic weather generators are often used in the construction of long time series of rainfall that can be used in conjunction with rainfall-runoff models for risk assessment in the planning of water resources and flood mitigating facilities. The basic requirements of the weather generators used for such purposes are that they be able to reproduce the statistical properties of the historical rainfall series at each site and the spatial covariance structure between sites. Although a single type of distribution has frequently been implemented to model the amount of daily precipitation with seasonally varying parameters, it can sometimes be inadequate to capture some of the statistical properties of the daily rainfall that have relevance to the purpose to which the model is applied. In the present work, we demonstrate applicability of a stochastic model for the generation of daily time series of rainfall at multiple locations in which the amount of daily rainfall is modelled by a mixture of two different probability distribution functions. A two stage modelling procedure is implemented. In the first stage, a multivariate autoregressive model is used to model the local probability of occurrence of rainfall and the amount while keeping the inter-site covariance structure using a truncated and power transformed normal distribution. In the second stage, the amount simulated using the power transformed normal distribution is further transformed so that it can be regarded as coming from a mixture of Gamma and Gumbel distribution. The annual cycles of the amount as well as the temporal and spatial correlations are incorporated using a Fourier representation. Application was made on 122 stations within the Unstrut catchment in Eastern Germany. Results show that the model can fairly well reproduces the monthly mean rainfall and the corresponding variability as well as the extreme value distribution of the annual maximum daily rainfall.
H31B-0357
Probabilistic Prediction Of Heavy Rainfall Using Pattern Recognition Technique Based On Self-Organizing Map (SOM)
Most of the heavy rainfall systems are closely related to spatially-extended meteorological information, in other words, multi-dimensional information. Therefore, in this study, pattern recognition technique based on Self- Organizing Map (SOM) was applied to the prediction of heavy rainfall in the rainy season in Japan in combination with Back Propagation (BP: supervised ANN). The SOM is an unsupervised ANN-based pattern recognition technique, projecting high-dimensional input variables onto two-dimensional regularly-arranged units for visualization. Here, the patterns of meteorological field characterizing the rainy season (BAIU) in Japan were classified using the SOM, and related to rainfall data using the BP. From the results, the rainfall prediction technique succeeded in constructing meteorologically-significant relationships between complicated meteorological field patterns and rainfall by focusing on the inherent properties of the SOM. Particularly, it was clearly shown that high probability of heavy rainfall corresponds to a meteorological field pattern characterized by Low-Level Jet (LLJ) and ample water vapor, which was closely associated with disastrous rainfall events in Japan.
H31B-0358
Volume and Flood Frequency Characteristics of Ephemeral Streams in the Semi-arid U.S.
Large runoff events in the arid and semi-arid regions of the U.S. are generally a result of intense, short-duration (less than 24 hours) precipitation events and frequently result in considerable property damage and loss of life. Urban flood mitigation and channel restoration projects in these areas currently depend on short duration volume frequency relationships derived from precipitation characteristics in the course of performing rainfall-runoff analyses, rather than directly from observed streamflow data as is standard practice for longer durations (>1- day). Until recently, only average daily flows were reported for USGS stream gauges so hydrograph characteristics could not be used to assess rainfall-runoff models or evaluate whether mitigation structures were meeting design specifications. Archived high-temporal resolution streamflow records for 14 uncontrolled ephemeral streams in the Las Vegas region were digitized so that hydrograph characteristics (peak, shape, volume) could be measured. Where possible, flood frequency and volume-frequency-duration analyses were performed for the ephemeral streams. These analyses demonstrate the importance of using partial duration series rather than annual maxima for frequency analyses as well as the significance of using maximum daily flow rather than average daily flow when assessing the hazards posed by ephemeral streams.
H31B-0359
Modeling Water Shortage Management Using an Object-Oriented Approach
As a result of the increasing global population and the resulting urbanization, water shortage issues have received increased attention throughout the world . Water supply has not been able to keep up with increased demand for water, especially during times of drought. The use of an object-oriented (OO) approach coupled with efficient mathematical models is an effective tool in addressing discrepancies between water supply and demand. Object-oriented modeling has been proven powerful and efficient in simulating natural behavior. This research presents a way to model water shortage management using the OO approach. Three groups of conceptual components using the OO approach are designed for the management model. The first group encompasses evaluation of natural behaviors and possible related management options. This evaluation includes assessing any discrepancy that might exist between water demand and supply. The second group is for decision making which includes the determination of water use cutback amount and duration using established criteria. The third group is for implementation of the management options which are restrictions of water usage at a local or regional scale. The loop is closed through a feedback mechanism where continuity in the time domain is established. Like many other regions, drought management is very important in south Florida. The Regional Simulation Model (RSM) is a finite volume, fully integrated hydrologic model used by the South Florida Water Management District to evaluate regional response to various planning alternatives including drought management. A trigger module was developed for RSM that encapsulates the OO approach to water shortage management. Rigorous testing of the module was performed using historical south Florida conditions. Keywords: Object-oriented, modeling, water shortage management, trigger module, Regional Simulation Model
H31B-0360
Measured and Modeled Water Balances For Three Snow Dominated Forested Catchments With Different Canopy Cover In The Inland Pacific Northwest
There is a need to better understand the dominant components of the catchment water balance in complex vegetated terrain to advance our understanding of basic hydrological processes and develop effective land management practices. A lack of paired or other detailed watershed studies in the inland Pacific Northwest has limited our understanding of this hydroclimatically, biophysically, and topographically complex region. Empirical analyses of long term data sets, and results from detailed investigations were used to assess impacts of contemporary timber harvest practices on the water balance components of the Mica Creek Experimental Watershed (MCEW). Measurements at the MCEW include precipitation, rainfall interception, snow water equivalent, sap flux, soil moisture, and streamflow. Results were applied to annual averages for the 2002 through 2005 water years directly following canopy removal. Of total precipitation (1401 mm/wy), 755 mm, 628 mm, and 475 mm/yr resulted in streamflow from clearcut, partial cut, and fully forested catchments respectively. Study results showed that canopy interception of rain was 17.2 % of rainfall for a full canopy, and 13.3 % for a partial cut (i.e. 50 % harvest) canopy. Canopy interception of snow was calculated as 43 % and 60 % for partial cut and full forest respectively. Based on sap flow measurements, transpiration was calculated to be 1.5 mm/day for approximately 200 days per year, or 300 mm/year. Based on these findings, estimates of evaporation (including sublimation) were 161 mm, 361 mm, and 470 mm/yr, and estimates of transpiration were 148 mm, 221 mm, and 296 mm/yr, for clearcut, partial cut, and fully forested catchments respectively. This suggests that water yield increased 30 % following clearcut harvest, and 20 % following partial cut harvest, and evaporation dropped to nearly 30 % of pre-harvest evaporation following clearcut, and nearly 60 % of pre-harvest evaporation following partial cut harvest. Soil evaporation was considered to be negligible, but nonetheless could be a source of error, especially for the clearcut catchments. Transpiration estimates for partial cut and fully forested catchments may also be somewhat lower than current estimates. The difference in Error ranged from 156 mm/wy between paired fully forested control watersheds, to 338 mm/year in the clearcut catchments. Errors may result from multiple processes including soil evaporation, spatial distribution of snow, storage changes, and transpiration estimate errors. Results of the empirical water balance components will be compared to modeled values using the Distributed Hydrology Soil Vegetation Model (DHSVM) and will result in improved understanding and modeling predictive power in this topographically and hydroclimatically complex region. http://www.cnr.uidaho.edu/micacreek/
H31B-0361
Improvements on the Regional Climate Predictions Using Initialized Soil Moisture and Snow Fields
The Experimental Climate Prediction Center's (ECPC's) Regional Spectral Model (RSM) was used for three experiments simulating the North American climate from 1979 to 2004. A 26-year control experiment used the National Centers for Environmental Prediction-Department of Energy Reanalysis II (R-2) as initial and boundary conditions. A climatology forcing experiment used daily climatological values of soil moisture and snow obtained from the control simulation. In a 3rd simulation, differences between the ECPC RSM predicted precipitation and observed precipitation were used to correct the simulated soil moisture. These three simulations then provided the initial soil moisture and snow fields for monthly predictions. Preliminary evaluations indicate that knowledge of the surface initial state overall improves the skill of monthly forecasts, particularly in spring and summer.