Probabilistic Forecast of the Extended Range Heatwave Over Eastern China

Several probabilistic forecast methods for heatwave (HW) in extended-range scales over China are constructed using four models (ECMWF, CMA, UKMO, and NCEP) from the Subseasonal-to-Seasonal (S2S) database. The methods include four single-model ensembles (SME; ECMWF, CMA, UKMO, and NCEP), multi-model ensemble (MME), and Bayesian model averaging (BMA). The construction and verification of reforecasts are implemented by a defined heat wave index (HWI) which is not only able to reflect the actual occurrence of heatwaves, but also to facilitate forecast and verification. The performance is measured by traditional verification method at each grid point of the 105°E to 132°E; 20°N to 45°N domain for the July, August, and September (JAS) of 1999–2010. For deterministic evaluations of HWI forecast, BMA shows a better pattern correlation coefficient than SME and MME and comparable equitable threat score (ETS) with ECMWF and MME. The good performance of ECMWF and MME take advantage of setting the percentile thresholds for forecasting HW. For the probabilistic forecast, the Brier score of BMA is comparable (superior) to that of MME and ECMWF at short (long) lead-time. BMA also demonstrates an improvement on the reliability of probabilistic forecast, indicating that BMA method is a useful tool for an extended-range forecast of HW. Meanwhile, in the real-time extended-range probabilistic forecast, the beginning date, end date, and probability of HW event can be predicted by the HWI probabilistic forecast of BMA.

A Study on Tensile Behavior According to the Design Method for the CFRP/GFRP Grid for Reinforced Concrete

In the present study, fiber-reinforced plastics (FRP) grid-reinforced concrete with very rapid hardening polymer (VRHP) mortar composites were fabricated using three types of design methods for the FRP grid (hand lay-up method, resin infusion method, and prepreg oven vacuum bagging method), along with two types of fibers (carbon fiber and glass fiber) and two types of sheets (fabric and prepreg). The FRP grid was prepared by cutting the FRP laminates into a 10 mm thick, 50 mm × 50 mm grid. The tensile behavior of the FRP grid embedded in composites was systematically analyzed in terms of the load extension, fracture mode, partial tensile strain, and load-bearing rate. The CFRP grid manufactured by the prepreg OVB method showed the best tensile behavior compared to the CFRP grid manufactured by the hand lay-up and resin infusion methods. The load-bearing of each grid point was proportional to the height from the load-bearing part when reaching the maximum tensile load. In addition, finite element analysis was conducted to compare the experimental and analysis results.

A diagnostic study of flood producing rainstorm of September 1988 over northwest India with the aid of a fine mesh numerical analysis system

A numerical analysis of the synoptic situation leading to devastating floods in Punjab and adjoining states during September 1988 has been carried out. The analysis is done by three dimensional multivariate optimum interpolation (OI) scheme cast on 1° x 1° Lat./Long. Grid. Software has been developed for computation of several derived parameters and linked with the basic flow variable analysis. A diagnostic study of day-to-day rainfall versus the objectively analysed grid point fields of integrated horizontal flux divergence of water vapour is carried out, The study brings out a close spatial correspondence between the area of net moisture flux convergence on the analysis day and the area of heavy rainfall on the following day. The study suggests that the numerical analysis products can be of a good predictive value to a synoptic forecaster In heavy rainfall predictions under difficult and uncertain synoptic situations.

On the use of outgoing longwave radiation data in incorporating divergent part of the wind in analysis

In the present study, kinematic divergence computed using ECMWF grid point data at 850 hPa is enhanced by using the relationship between OLR and divergence. This new enhanced divergence is used to compute the velocity potential and then, the divergence part of the wind is obtained from velocity potetial. To obtain the rotational part of wind, we computed the vorticity from wind data, and subsequently stream function and obtained and the rotational part of the wind from the stream function. The total wind is the combination of divergent part obtained from modified velocity potential (using OLR data) and rotational part from unmodified stream function. This total wind field is used as initial guess for univariate objective analysis by optimum interpolation scheme so that Initial Guess field contained the more realistic divergent part of the wind. Consequently, the analysed field also will contain the divergent part of the wind.

A Stochastic Rockfall Model Related to Random Ground Roughness Based on Three-Dimensional Discontinuous Deformation Analysis

The use of feasible 3-D numerical methods has become essential for addressing problems related to rockfall hazard. Although several models with various degrees of complexity are available, certain trajectories and impact dynamics related to some model inputs could fall in the rockfall observations area but are rarely calibrated against reflecting its range, especially the lateral deviations. A major difficulty exists in the lack of simulating the apparent randomness during the impact-rebound process leading to both ground roughness and block irregularities. The model presented here is based on three-dimensional discontinuous deformation analysis (3-D DDA). Despite similarities to previous simulations using 3-D DDA, the model presented here incorporates several novel concepts: (1) ground roughness is represented as a random change of slope angle by height perturbation at a grid point in DEM terrain; (2) block irregularities are modelled directly using polyhedron data; (3) a scaled velocity restitution relationship is introduced to consider incident velocity and its angle. Lateral deviations of rebound velocity, both direction and value, at impact are similarly accounted for by perturbing the ground orientation laterally, thus inducing scatter of run-out directions. With these features, the model is capable to describe the stochastic rockfall dynamics. In this study, 3-D DDA was then conducted to investigate the dynamic behavior of the rockfall and examine the role of sphericity of the rock block travelling on bench slopes with different ground roughness levels. Parametric analyses were carried out in terms of cumulative distribution function (CDF) to investigate for spatial distribution (both runout distance and lateral displacement), velocity and jumping height. The effects of block shape and ground roughness revealed by these factors were discussed. It suggests that ground roughness amplifies the randomness and plays important roles on the dynamic behavior of the system; irregularity from block sphericity will further amplify the randomness especially when the size of the rock is relatively small compared to the roughness level. Both irregularities should be taken into consideration in simulating rockfall problems. Further calibration of the new model against a range of field datasets is essential.

Reducing transport errors in WRF modeling of greenhouse gas distributions through a combination of grid-nudging and regular restarts

<p>Atmospheric transport models are often used to simulate the distribution of Greenhouse Gases (GHGs) for atmospheric inverse modeling. However, errors in simulated transport are often neglected in the context of inverse flux estimation. We coupled the commonly used Weather Research and Forecasting (WRF) model with the greenhouse gas module (WRF-GHG), to enable passive tracer transport simulation of CO<sub>2</sub> and CH<sub>4</sub>. As a mesoscale numerical weather prediction model, WRF’s transport is only constrained by global meteorological fields via initialization and at the lateral boundaries; over time the winds in the center of the domain can deviate considerably from these (re-)analysis fields that are constrained by observations. The aim of this study is to have the WRF-simulated transport represent reality as closely as possible, which in this case means staying consistent with the (ERA5) reanalysis fields used as boundary conditions.</p> <p>Therefore, two ways of blending ERA5 with WRF-GHG were tested: (a) regularly restarting the model with fresh initial conditions from ERA5, and (b) nudging the atmospheric winds, temperatures, and moisture to those from ERA5 continuously, using the built-in FDDA option (four-dimensional data assimilation). FDDA constantly forces the model towards the physical reference state (ERA5) by adding an additional tendency term at each grid point and time step.</p> <p>Meteorological variables, as well as the concentrations of CO<sub>2</sub> and CH<sub>4</sub>, were analyzed by comparing with observations. We also compared mixed layer heights (PBLH) with radiosonde-derived observation. We found that performance in horizontal winds and PBLH are slightly better in the nudged simulation (NS) compared to the simulation incorporating frequent restarts (RS). The advantage of grid-nudging is notable when comparing CH<sub>4</sub> with aircraft measurements from the CoMet campaign. However, differences in soil moisture increase over time, as soil moisture is not used for nudging. The consequence is a change in the Bowen ratio and thus in vertical mixing, impacting the distribution of GHG tracers in general.</p> <p>To preserve the benefits of nudging and avoid the divergence of soil moisture, we recommend a hybrid approach, combining nudging with daily re-initializations. This technique will be used in an ensemble-based regional inversion system currently under development to make use of satellite-based measurements of GHGs.</p>

MODEL- AND FULL-SCALE RANS SIMULATIONS FOR A DRIFT TANKER

The flow around a full-scale (FS) ship can be simulated by means of Reynolds-Averaged Naiver-Stokes (RANS) method, which provides a way to obtain more knowledge about scale effects on ship hydrodynamics. In this work, the viscous flow around a static drift tanker in full scale is simulated by using the RANS solver based on the open source platform OpenFOAM. The k - w SST model is employed to approximate the eddy viscosity. To reduce computational time, wall function approach is applied for the FS simulation. The flow around the ship in model scale is simulated as well, but without using any wall function, i.e., using Low-Reynolds number mode. In order to verify the computations, de- tailed studies on the computational grid including investigation of the sensitivity of computed forces to y+ (dimension- less distance of first grid point to wall) and grid dependency study are carried out. The computed forces are compared with available measured data. The scale effects are analysed and discussed by comparisons.

Spatial rainfall patterns associated with Indian northeast monsoon derived from high resolution rainfall estimates of Chennai DWR

The Indian northeast monsoon (NEM) season benefits the southeastern parts of peninsular India during the period October-November-December (OND). In this study, which is a first of this type for the Indian region, certain new and salient features of the NEM rainfall (RF) have been derived utilising the very high resolution (333 m × 333 m) radar estimated rainfall (RERF) data generated by the Doppler Weather Radar (DWR) at Chennai for the 12 year period (2002-13), over a circular area of 100 km radius spreading over both land and ocean. More than 2.8 lakhs of grid point data per day have been processed. Rain gauge measured rainfall (RGRF) data of 34 inland stations has also been used. Monthwise spatial distributions of RERF for October, November and December and for the entire season OND have been generated. It is shown through rigorous analysis that RERF is heavier closer to the coast and for a given longitude over land, southern latitudes receive 10-15% more RF than the northern latitudes. Decrease of RF eastwards into Bay of Bengal (BoB) is gradual whereas westwards over inland it is sharp and almost linear. By and large, the climatological features of NEM derived from historical analysis of RGRF data are well-captured by the analysis based on RERF data. A few new features of monthly and seasonal RF have also been identified. For the 34 stations, 12 year data set for OND, the mean RERF and RGRF values are 629.8 mm and 627.4 mm respectively yielding a difference of just 2.4 mm but with a substantial mean absolute deviation of 69.2 mm. RERF during pre-NEM days of Oct contributed to 10% of the seasonal OND total. RERF in the area of study, during days of cyclonic disturbances (CD days) is nearly twice over outer oceanic areas of BoB than over land. It has been shown that during the onset to withdrawal period of NEM, RERF is heavier over areas close to the coast (75 cm) than oceanic areas (68 cm) within the 100 km radius of the DWR. High RF zones approximately extending 25-30 km westwards into land and around 30-40 km eastwards over the ocean have been delineated. Spatial distributions of RERF during the various phases of NEM, viz., dry, weak, normal, active and vigorous as identified from the RGRF data have been generated, critically analysed and results drawn. In the case of vigorous, active and vigorous (AV) NEM days excluding CD days, a relatively high daily RERF patch of 5-6 cm located approximately 5-10 km west of the coast inland and in the SW sector of Chennai DWR has been identified. During post-NEM withdrawal days of December, oceanic areas of eastern sector are shown to receive highest RF compared to land areas, a feature consistent with the withdrawal pattern of NEM. The instrumental limitations and artifacts of radars contributing to errors in RERF have been discussed.

Extreme metrics from large ensembles: investigating the effects of ensemble size on their estimates

We consider the problem of estimating the ensemble sizes required to characterize the forced component and the internal variability of a number of extreme metrics. While we exploit existing large ensembles, our perspective is that of a modeling center wanting to estimate a priori such sizes on the basis of an existing small ensemble (we assume the availability of only five members here). We therefore ask if such a small-size ensemble is sufficient to estimate accurately the population variance (i.e., the ensemble internal variability) and then apply a well-established formula that quantifies the expected error in the estimation of the population mean (i.e., the forced component) as a function of the sample size n, here taken to mean the ensemble size. We find that indeed we can anticipate errors in the estimation of the forced component for temperature and precipitation extremes as a function of n by plugging into the formula an estimate of the population variance derived on the basis of five members. For a range of spatial and temporal scales, forcing levels (we use simulations under Representative Concentration Pathway 8.5) and two models considered here as our proof of concept, it appears that an ensemble size of 20 or 25 members can provide estimates of the forced component for the extreme metrics considered that remain within small absolute and percentage errors. Additional members beyond 20 or 25 add only marginal precision to the estimate, and this remains true when statistical inference through extreme value analysis is used. We then ask about the ensemble size required to estimate the ensemble variance (a measure of internal variability) along the length of the simulation and – importantly – about the ensemble size required to detect significant changes in such variance along the simulation with increased external forcings. Using the F test, we find that estimates on the basis of only 5 or 10 ensemble members accurately represent the full ensemble variance even when the analysis is conducted at the grid-point scale. The detection of changes in the variance when comparing different times along the simulation, especially for the precipitation-based metrics, requires larger sizes but not larger than 15 or 20 members. While we recognize that there will always exist applications and metric definitions requiring larger statistical power and therefore ensemble sizes, our results suggest that for a wide range of analysis targets and scales an effective estimate of both forced component and internal variability can be achieved with sizes below 30 members. This invites consideration of the possibility of exploring additional sources of uncertainty, such as physics parameter settings, when designing ensemble simulations.

Automated temporal tracking of coherently evolving density fronts in numerical models

Oceanic density fronts can evolve, be advected, or propagate as gravity currents. Frontal evolution studies require methods to temporally track evolving density fronts. We present an automated method to temporally track these fronts from numerical model solutions. First, at all time steps contiguous density fronts are detected using an edge detection algorithm. A front event, defined as a set of sequential-in-time fronts representing a single time-evolving front, is then identified. At time step i, a front is compared to each front at time step i + 1 to determine if the two fronts are matched. An i front grid point is trackable if the minimum distance to the i + 1 front falls within a range. The i front is forward-matched to the i + 1 front when a sufficient number of grid points are trackable and the front moves onshore. A front event is obtained via forward tracking a front for multiple time steps. Within an event, the times that a grid point can be tracked is its connectivity and a pruning algorithm using a connectivity cutoff is applied to extract only the coherently evolving components. This tracking method is applied to a realistic 3-month San Diego Bight model solution yielding 81 front events with duration ≥ 7 hours, allowing analyses of front event properties including occurrence frequency and propagation velocity. Sensitivity tests for the method’s parameters support that this method can be straightforwardly adapted to track evolving fronts of many types in other regions from both models and observations.