Abstract

A soil–vegetation–atmospheric boundary layer model was developed to study the performance of two local-closure and two nonlocal-closure boundary layer mixing schemes for use in meteorological and air quality simulation models. Full interaction between the surface and atmosphere is achieved by representing surface characteristics and associated processes using a prognostic soil–vegetation scheme and atmospheric boundary layer schemes. There are 30 layers in the lowest 3 km of the model with a high resolution near the surface. The four boundary layer schemes are tested by simulating atmospheric boundary layer structures over densely and sparsely vegetated regions using the observational data from the First ISLSCP (International Satellite Land Surface Climatology Project) Field Experiment (FIFE) and from Wangara.

Simulation results indicate that the near-surface turbulent fluxes predicted by the four boundary layer schemes differ from each other, even though the formulation used to represent the surface-layer processes is the same. These differences arise from the differing ways of representing subgrid-scale vertical mixing processes. Results also indicate that the vertical profiles of predicted parameters (i.e., temperature, mixing ratio, and horizontal winds) from the four mixed-layer schemes differ from each other, particularly during the daytime growth of the mixed layer. During the evening hours, after the mixed layer has reached its maximum depth, the differences among these respective predicted variables are found to be insignificant.

There were some general features that were associated with each of the schemes in all of the simulations. Compared with observations, in all of the cases the simulated maximum depths of the boundary layer for each scheme were consistently either lower or higher, superadiabatic lapse rates were consistently either stronger or weaker, and the intensity of the vertical mixing was either stronger or weaker. Also, throughout the simulation period in all case studies, most of the differences in the predicted parameters are present in the surface layer and near the top of the mixed layer.

Abstract

Hurricane Opal (1995) experienced a rapid, unexpected intensification in the Gulf of Mexico that coincided with its encounter with a warm core ring (WCR). The relative positions of Opal and the WCR and the timing of the intensification indicate strong air–sea interactions between the tropical cyclone and the ocean. To study the mutual response of Opal and the Gulf of Mexico, a coupled model is used consisting of a nonhydrostatic atmospheric component of the Naval Research Laboratory’s Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS), and the hydrostatic Geophysical Fluid Dynamics Laboratory’s Modular Ocean Model version 2 (MOM 2).

The coupling between the ocean and the atmosphere components of the model are accomplished by conservation of heat, salt, momentum, as well as the sensible and latent heat fluxes at the air–sea interface. The atmospheric model has two nests with spatial resolutions of 0.6° and 0.2°. The ocean model has a uniform resolution of 0.2°. The oceanic model domain covers the Gulf of Mexico basin and coincides with a fine-mesh atmospheric domain of the COAMPS. The initial condition for the atmospheric component of COAMPS is the archived Navy Operational Global Atmospheric Prediction System operational global analysis, enhanced with observations. The initial ocean condition for the oceanic component is obtained from a 2-yr MOM 2 simulation with climatological forcing and fixed mass inflow into the Gulf. The initial state in the Gulf of Mexico consists of a realistic Loop Current and a shed WCR.

The 72-h simulation of the coupled system starting from 1200 UTC 2 October 1995 reproduces the observed storm intensity with a minimum sea level pressure (MSLP) of 918 hPa, occurring at 1800 UTC 4 October, a 6-h delay compared to the observation. The rapid intensification to the maximum intensity and the subsequent weakening are not as dramatic as the observed. The simulated track is located slightly to the east of the observed track, placing it directly over the simulated WCR, where the sea surface temperature (SST) cooling is approximately 0.5°C, consistent with buoy measurements acquired within the WCR. This cooling is significantly less over the WCR than over the common Gulf water due to the deeper and warmer layers in the WCR. Wind-induced currents of 150 cm s⁻¹ are similar to those in earlier idealized simulations, and the forced current field in Opal’s wake is characterized by near-inertial oscillations superimposed on the anticyclonic circulation around the WCR.

Several numerical experiments are conducted to isolate the effects of the WCR and the ocean–atmosphere coupling. The major findings of these numerical experiments are summarized as follows.

1. Opal intensifies an additional 17 hPa between the times when Opal’s center enters and exits the outer edge of the WCR. Without the WCR, Opal only intensifies another 7 hPa in the same period.

2. The maximum surface sensible and latent heat flux amounts to 2842 W m⁻². This occurs when Opal’s surface circulation brings northwesterly flow over the SST gradient in the northwestern quadrant of the WCR.

3. Opal extracts 40% of the available heat capacity (temperature greater than 26°C) from the WCR.

4. While the WCR enhances the tropical cyclone and ocean coupling as indicated by strong interfacial fluxes, it reduces the negative feedback. The negative feedback of the induced SST cooling to Hurricane Opal is 5 hPa. This small feedback is due to the relatively large heat content of the WCR, and the negative feedback is stronger in the absence of the WCR, producing a difference of 8 hPa in the MSLP of Opal.

Abstract

Data from the Special Sensor Microwave/Imager (SSM/I) on board a Defense Meteorological Satellite Program satellite are used to study the precipitation patterns and wind fields associated with Hurricane Florence (1988). SSM/I estimates indicate that the intensification of Florence was coincident with the increase in total latent beat release. Additionally, an increase in the concentration and areal coverage of heavier rain rates near the center is observed. SSM/I marine surface winds of Florence are examined and compared to in situ data, and to an enhanced objective isotach analysis over the Gulf of Mexico. Results indicate that the SSM/I winds are weaker than those depicted in the enhanced objective analysis and slightly stronger than in situ observations. Finally, center positions of Florence are estimated using the 85-GHz brightness temperature imagery. Much improved estimates are achieved using this imagery compared to using GOES infrared imagery. These results concur with previous studies that applications of SSM/I data could be valuable in augmenting current methods of tropical cyclone analysis.

Abstract

A system for the frequent intermittent assimilation of surface observations into a mesoscale model is described. The assimilation begins by transforming the surface observations to model coordinates. Next, the lowest-level model fields of potential temperature, relative humidity, u and v component winds, and surface pressure are updated by an objective analysis using the successive correction approach. The deviations of the analysis from the first guess at the lowest model layer are then used to adjust the other model layers within the planetary boundary layer. The PBL adjustment is carried out by using the model's values of eddy diffusivity, which are nudged to reflect the updated conditions, to determine the influence of the lowest-layer deviations on the other model layers. Results from a case study indicate that the frequent intermittent assimilation of surface data can provide superior mososcale analyses and forecasts compared to assimilation of synoptic data only. The inclusion of the PBL adjustment procedure is an important part of generating the better forecasts. Extrapolation of the results here suggests that two-dimensional data can be successfully assimilated into a model provided there is a mechanism to smoothly blend the data into the third dimension.

Abstract

Fields of rainfall rates, integrated water vapor (IWV), and marine surface wind speeds retrieved by the Special Sensor Microwave/Imager (SSM/I) during the intensive observational period 4 on 4 January 1989 of the Experiment on Rapidly Intensifying Cyclones over the Atlantic (ERICA) were analyzed. Subjectively analyzed and model-simulated frontal structures were used to examine the spatial relationship of the SSM/I observed fields to the rapidly intensifying storm and the associated fronts. Qualitative and quantitative comparisons of SSM/I retrievals with GOES imagery, conventional observations, and results produced from the Naval Research Laboratory's (NRL) limited-area numerical model were also made.

SSM/I rainfall was found along the cold and warm fronts, with heavy precipitation within frontal bands. The spatial pattern and characteristics of SSM/I precipitation closely resembled those simulated by the model. Both the warm and the cold front were found to be located near the area of the strongest gradient in IWV. In the warm sector, areas of IWV greater than 40 mm were found, an amount supported by model simulations. Both SSM/I rain rate and IWV distribution were found to be useful in locating the cold and warm fronts. There was good agreement on the relationship of frontal locations to the precipitation patterns and IWV gradients. Most of the high-wind area near the storm center was obscured by clouds for marine surface wind retrieval. SSM/I-retrieved marine surface winds outside the cloud shield (flag 0) were compared to ship- and buoy-reported winds. It was found that the retrieved wind estimates were within 0–3 m s⁻¹ of in situ observation over areas of slow wind shifts. The errors became larger in regions of rapid wind shifts.

Abstract

A technique is described that adds diabatic forcing from stratiform precipitation to a vertical normal-mode initialization of a mesoscale model. The technique uses observed precipitation amounts and cloud-top height estimations with analyzed thermodynamic and kinematic fields to vertically distribute diabatic heating that arises from stratiform precipitation. Simulation experiments reveal the importance of incorporating this heating into the initialization. An adiabatic initialization recovered about 65%–75% of the maximum upward vertical motions, whereas a diabatic initialization, with respect to stratiform precipitation, recovered nearly all the original vertical motions. A real-data case study is presented using combined rain gauge-satellite precipitation analyses with cloud-top heights estimated from Geostationary Operational Environmental Satellite infrared brightness temperatures. The short-term precipitation forecasts from a diabatically initialized model, with respect to stratiform precipitation, demonstrate improvement over forecasts from an adiabatically initialized model.

Abstract

The Flux-Adjusting Surface Data Assimilation System (FASDAS) uses the surface observational analysis to directly assimilate surface layer temperature and water vapor mixing ratio and to indirectly assimilate soil moisture and soil temperature in numerical model predictions. Both soil moisture and soil temperature are important variables in the development of deep convection. In this study, FASDAS coupled within the fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) was used to study convective initiation over the International H₂O Project (IHOP_2002) region, utilizing the analyzed surface observations collected during IHOP_2002. Two 72-h numerical simulations were performed. A control simulation was run that assimilated all available IHOP_2002 measurements into the standard MM5 four-dimensional data assimilation. An experimental simulation was also performed that assimilated all available IHOP_2002 measurements into the FASDAS version of the MM5, where surface observations were used for the FASDAS coupling. Results from this case study suggest that the use of FASDAS in the experimental simulation led to the generation of greater amounts of precipitation over a more widespread area as compared to the standard MM5 FDDA used in the control simulation. This improved performance is attributed to better simulation of surface heat fluxes and their gradients.

The objective of this research is to determine whether poorly sited long-term surface temperature monitoring sites have been adjusted in order to provide spatially representative independent data for use in regional and global surface temperature analyses. We present detailed analyses that demonstrate the lack of independence of the poorly sited data when they are adjusted using the homogenization procedures employed in past studies, as well as discuss the uncertainties associated with undocumented station moves. We use simulation and mathematics to determine the effect of trend on station adjustments and the associated effect of trend in the reference series on the trend of the adjusted station. We also compare data before and after adjustment to the reanalysis data, and we discuss the effect of land use changes on the uncertainty of measurement.

A major conclusion of our analysis is that there are large uncertainties associated with the surface temperature trends from the poorly sited stations. Moreover, rather than providing additional independent information, the use of the data from poorly sited stations provides a false sense of confidence in the robustness of the surface temperature trend assessments.