SOARS Manuscripts

Papers and posters completed by protégés in the Significant Opportunities in Atmospheric Research (SOARS) program, dating back to the program's start in 1996. SOARS provides funded research opportunities to undergraduate and graduate students from backgrounds that are underrepresented in the atmospheric sciences.


Pages

A Comparison of ADCIRC Storm Surge Predictions Given	Different Forcing Inputs
A Comparison of ADCIRC Storm Surge Predictions Given Different Forcing Inputs
In September of 2008, Hurricane Ike swept across the Gulf of Mexico and hit the coast of Galveston, Texas causing high storm surges that led to billions of dollars in damage and killed hundreds of people. Predicting storm surges is difficult, due to limits in computational ability and forecasting errors, which hinders emergency evacuations and preparation. Simulations of Hurricane Ike were used to better understand the usefulness of different methods of constructing and imposing the storm in the surge model environment in an attempt to explore the resulting surge and to increase predictive abilities. The ADvanced CIRCulation (ADCIRC) model was forced with two different representations of the storm, both originating from forecast data generated with the Weather Research and Forecasting model (WRF). The comparison was made between the storm surge simulations forced with the full WRF wind and pressure fields, and a simplified storm approximation constructed from the WRF data, partly developed in this work. The simplified approach used for forcing, developed here, was obtained by measuring the nautical distance from the center of the hurricane, at 6 hour intervals, for the maximum extent of the 34-, 50-, 64-knot, and strongest winds. Differences were found among the resulting storm surges, most noticeably in the points of inundation near where the storm made landfall. Given further investigation, the simplified storm model may produce results more consistent with WRF, but currently the complicated WRF model is a better tool for determining where evacuation and emergency preparation efforts should be focused.
A FORTRAN program to calculate descriptive statistics of satellite-derived tropical prescription.
A FORTRAN program to calculate descriptive statistics of satellite-derived tropical prescription.
Precipitation can be calculated using satellite observations. Satellites can estimate cloud-tops by measuring the outgoing longwave radiation (OLR). After getting the daily OLR data we change it to daily precipitation estimates for the tropics. The focus of this paper is to examine and calculate spatial and temporal statistics for this daily precipitation. The calculations ate done using a FORTRAN program. The results are important because they determine the precipitation climatology and inter annual variability for the tropics.
A GEMPAK (General Meteorological Package) tutorial available via NCSA Mosiac, written in HTML (Hypertext Markup Language).
A GEMPAK (General Meteorological Package) tutorial available via NCSA Mosiac, written in HTML (Hypertext Markup Language).
My project entails designing a tutorial for the GEMPAK (GEneral Meteorological PAcKage) software distributed by Unidata, a division of UCAR (University Corporation for Atmospheric Research). GEMPAK analyzes, diagnoses, and displays geo-referenced data, usually meteorological data. The tutorial is designed using HTML (Hypertext Markup Language). HTML is a language that assigns structural elements to text, such as: emphasis, address, citation, code, etc. [CTAN, 1993] HTML documents are read and displayed using a client application called a browser. I worked primarily with Mosaic, one of the most common browser applications. It also can imbed images, sounds, video, and "links" to other documents. The browser, therefore, can be programmed or configured to display these structural elements as the programmer or user chooses. With Mosaic, links appear as text or bordered images highlighted in blue or purple. The user can select a highlighted object (with the mouse if using Mosaic) to view another document. The GEMPAK tutorial uses "next" and "prev" (previous) icons to link the sections so the user can progress sequentially, and it has a detailed table of contents so the user can skip around the document. At the end of each section is a set of exercises that test users' understanding and the tutorial can display the correct answer so that users can check their results. This interactive quality of HTML documents makes it an excellent medium for tutorials.
A citizen science campaign encouraging urban forest professionals to engage the public in the collection of tree phenological data
A citizen science campaign encouraging urban forest professionals to engage the public in the collection of tree phenological data
There are growing concerns among leading national and local organizations about American scientific literacy, fundamental understanding of science, and the value of scientific research. These organizations, including the University Corporation for Atmospheric Research, have been at the forefront to change this trend. In an effort to improve scientific literacy, research conducted by Sam Droege, amongst others, suggested using citizen science and public participation as one instrumental method to engage the public. Urban Tree Phenology (UTP), a project of Project BudBurst and the USDA Forest Service, is one such citizen science program that sought to educe the public, including the professionals and the amateurs among them, in collecting urban tree phenophase data. UTP participants monitored and reported the stages of phenological events, such as First Leaf and Leaf Fall, of twenty-four native and cultivated urban tree species. Data collected will support the long-term research of plant ecology, climate change, public health, urban heat islands on tree physiology, and urban tree management. UTP, using the architectures of online learning, has developed two instructional tutorials to assist data collection (Phase 1). The instructional tutorials were published online, in print and PowerPoint formats, at www.UrbanTreePhenology.com. By completing these tutorials, participants will gain the skills necessary to provide urban tree phenological data to national research databases via the Internet. Phase 2 will test and review the instructional materials developed, and in Phase 3, the administrators of UTP will distribute promotional materials, to national research organizations, and to participants of the Project BudBurst national citizen science campaign.
A comparison of large-scale influences on tropical cyclogenesis in the Eastern Pacific
A comparison of large-scale influences on tropical cyclogenesis in the Eastern Pacific
In a given hurricane season, several tropical disturbances propagate across environments favorable for development; however, only a few disturbances actually strengthen into tropical cyclones. The lack of a consolidated theory on tropical cyclogenesis makes it difficult for forecasters to predict a storm’s development. Previous studies have approached this problem by comparing large-scale influences on storms that developed into tropical cyclones and on those that did not. This study used a similar approach to characterize the environmental influences on cyclogenesis in the 2005 Eastern Pacific Hurricane season. Data for each storm were taken from the NCEP/NCAR Final Analysis model and analyzed over a 48-hour period during the development stage. The non-developing storms were selected based on certain atmospheric parameters to resemble the developing storms prior to cyclogenesis. Composites and spatial averaging were used to compare 12 developing storms and 11 non-developing storms during this season. The results showed that the environments of the developing storms had large regions of increased moisture above the boundary layer and greater temperatures in the upper troposphere. Regions of increased potential vorticity penetrated deeper into the troposphere for the developing storms. Lastly, the storms that developed were in environments with relatively strong wind shear to the south of the vortex. The results suggest that the moisture, temperature, and wind shear fields preceded development, while the vorticity fields were more of an indicator of development. Identifying these large-scale characteristics as possible determining influences can lead to a better understanding of tropical cyclogenesis.
A comparison of vapor pressure estimates from MTCLIMv3 and CLIMSIM
A comparison of vapor pressure estimates from MTCLIMv3 and CLIMSIM
Plant productivity is strongly dependent upon moisture. Ecosystem models estimate the moisture content in the air from measurements or vapor pressure. Realistic vapor pressure values allow these models to more accurately estimate plant productivity. Observational measurements of vapor pressure are relatively scarce, and a reliable means of estimating vapor pressure is needed. CLIMSIM and its later version, MTCLIMv3, are climate simulators that estimate daily vapor pressure. CLIMSIM calculates vapor pressure by assuming that the dewpoint is reached every night and that it is equal to the night-time minimum temperature, a more widely available parameter. Unfortunately, this method overestimates the vapor pressure in arid and semi-arid regions. MTCLIMv3 estimates vapor pressure by scaling down minimum-temperature derived calculations by a ratio of daily potential evapotranspiration to annual precipitation. This study compared the vapor pressure estimates from CLIMSIM and an improved version. MTCLIMv3. The input data needed for the climate simulators, daily maximum and minimum temperature and daily precipitation, were obtained from a WGEN weather generator. CLIMSIM and MTCLIMv3 simulated vapor pressure for six locations in the United States: five in arid and semi-arid regions in the West and one in a relatively wetter region in the East. The success of the most recent version , MTCLIMv3, was determined by its ability to match the average monthly vapor pressures in the Danny Marks data set. which were derived from a climatology and assumed to be accurate. MTCLIMv3 out-performed CLIMSIM in all five of the arid and semi-arid regions. However, for certain months, MTCLIMv3 did no better than CLIMSIM at two dry sites. Why MTCLIMv3 fails to improve vapor pressure estimates all year for some dry climates and only some months for other dry climates needs to be investigated further.
A flash flood risk assessment of the Colorado Front Range using GIS
A flash flood risk assessment of the Colorado Front Range using GIS
Although significant research has been performed on impacts and mitigation of flash flood events, the methodology for assessing social vulnerability and regions at risk has not been fully developed. This project explored the environmental-social links of flood hazards and developed a GIS-based methodology for flood risk assessment. The assessment was based on a model that risk was a product of exposure to a hazard and societal vulnerability. Vulnerability was represented by population characteristics and distribution of critical facilities. Exposure was estimated by combining the Areal Mean Basin Average Rainfall (AMBER) method combined with GIS techniques. This method involved relating precipitation accumulation, averaged over a stream basin, to National Weather Service flash flood guidance values to identify basins with flooding potential. The vulnerability and the exposure were integrated in a GIS to estimate the total risk. The 1997 extreme precipitation event in Fort Collins, Colorado was used as a model to assess potential flood risk in two metropolitan areas: Fort Collins and Denver. Results yielded a GIS-based model that combines hydrometeorological information with social data, and allowed for radar-derived precipitation data to be integrated into the GIS to map key areas at risk in Fort Collins and Denver. Early identification of risk areas can assist emergency and flood-plain managers in developing response and mitigation measures. These results can provide a framework to expand this study of flood risk by introducing near-real time precipitation data, hydrological models and, detailed socio-economic geographic data.
A fuzzy logic system for predicting hurricane intensity in the Eastern North Pacific
A fuzzy logic system for predicting hurricane intensity in the Eastern North Pacific
The primary goal of this research is to examine the efficacy of using a fuzzy logic/adaptive weighting (FLAW) technique to predict hurricane intensity change in the eastern North Pacific (ENP) basin. The intensity change forecasts, for 12-hour intervals ranging from 12 to 72 hours, produced by a model using FLAW are compared to a model developed using the standard statistical technique of multiple linear regression (MLR). To this end, climatology and persistence variables, as well as observed intensity changes, are computed and extracted from the National Hurricane Center (NHC) best-track dataset and the weekly National Oceanic and Atmospheric Administration (NOAA) global optimum interpolation weekly sea surface temperature dataset for the years of 1982-99. The climatology and persistence predictors used for both models include the previous 6-hour intensity change, latitude of the hurricane center, current intensity at the time of the observation, and sea surface temperature interpolated to the time and location of the observed hurricane center. The FLAW technique is based on the Standard Additive Method (SAM) developed by Kosko. The method adapts the weights given to predictors to decrease the difference between the forecast and observed value. Preliminary results suggest that the FLAW model produces errors comparable in magnitude to the MLR model. The bias is significantly less for the 36-, 48-, 60-, and 72-hour forecast periods. Several case studies show that the model is able to adapt the weighting appropriately when the one or more predictors become over-dominant. With more work and the inclusion of synoptic predictors, this method may eventually offer improved hurricane intensity change guidance for forecasters, thus reducing the loss of lives and property.
A global analysis of atmospheric refractivity anomalies using CHAMP data
A global analysis of atmospheric refractivity anomalies using CHAMP data
In early 2006, the US-Taiwan joint satellite mission known as the Constellation Observing System for Meteorology, Ionosphere and Climate (COSMIC) will launch six Low Earth Orbit (LEO) satellites. These satellites, each equipped with an advanced Global Positioning System (GPS) receiver, will use radio occultation (RO) limb sounding technology to profile the Earth's atmosphere with unprecedented accuracy and vertical resolution. The GPS RO soundings available from COSMIC will make significant contributions to global weather prediction, ionospheric research, and climate monitoring. The GPS receivers will measure the phase and amplitude. From that we can deduce the bending angles as a function of height, and obtain vertical profiles of refractivity using the Abel inversion under the local spherical symmetry assumption. To demonstrate the potential value of GPS RO data in climate research, we analyzed atmospheric refractivity obtained from GPS RO data in a recent single-satellite German mission, known as the CHAllenging Mini Payload for Geophysical Research and Application (CHAMP). This study examined the refractivity anomalies by altitude and latitude per season of CHAMP GPS RO data, provided by UCAR's COSMIC Data Analysis and Archive Center (CDAAC), from May 2001 through present. Refractivity anomalies across the globe were illustrated in color plots that identified any persistent anomaly patterns. A structure has been identified over the tropical stratosphere from 20-30km, which may have a possible relationship with the Quasi-Biennial Oscillation (QBO). Results show that GPS refractivity data can be used to identify specific trends between seasons as well as identify multi-year phenomenon such as QBO. This study highlights the usefulness of refractivity values from GPS RO data in climate research.
A higher order tracer transport scheme for icosahedral hexagonal grid
A higher order tracer transport scheme for icosahedral hexagonal grid
Weather and climate models require both efficient and accurate numerical methods to simulate tracer (e.g., moisture, salinity) advection. The distribution of the tracer used in transport equations is approximated by a Taylor expansion. A scheme is developed that builds upon a simpler, second-order convergent method. This original method describes tracer distributions with a first-order Taylor expansion, while the extension uses a second-order expansion to describe the distribution. The original method is conservative and defines a simple departure region, but violates monotonicity preservation. This scheme, due to its simplicity, is not very accurate with more complex tracer and velocity flow configurations. The extension of the method requires three specific modifications: Green’s Theorem is used to calculate these next order terms and minimize the computational stencil, Gauss Quadrature is employed to calculate the tracer advected in a departure region, and the cell-averaged value is re-normalized to correct for the addition of these higher order terms. Two tests are run on a planar, perfect hexagonal grid: a solid body rotational case and a time-dependent deformational-flow case. The extension of the tracer distribution function shows marked improvements over the original method, and this extended scheme is third-order convergent for the solid-body rotation case. The improvements, however, are not as obvious when in the deformational-flow test. Nonetheless, the results indicate that the scheme warrants further testing. The successful application of a flux limiter shows that the method can be prepared further for possible implementation into weather and climate models.
A modeling study on ozone formation in the upper troposphere in relation to thunderstorms
A modeling study on ozone formation in the upper troposphere in relation to thunderstorms
Tropospheric ozone is important because of its deleterious effects as urban smog at ground level, its role as a greenhouse gas in the upper troposphere (UT), and its control of the oxidizing capacity of the troposphere through its photochemical derivative hydroxide (OH). Thunderstorms can affect ozone in the UT through production of nitrogen oxides (NOx) by lightning and convective transport of carbon monoxide (CO), hydrocarbons, and other volatile organic compounds that produce peroxy radicals. The purpose of this study was to better understand the importance of specific reaction rates in UT ozone chemistry by analyzing data generated by a high-resolution WRF-Chem simulation in which thunderstorms were explicitly represented at a horizontal resolution of 4 km. Our aim was to determine if thunderstorms associated with the annual North American Monsoon influenced the fate of NOx and peroxy radicals in the UT. Using output for a typical summer day, we mapped the most important peroxy radical production reactions and compared these with maps of storms as well as boundary layer and lightning-NOx tracers. The results indicated that the oxidation of CO, methane and short-lived, intermediate species such as formaldehyde and organic peroxides were the most important reactions. Furthermore, higher reaction rates of formation of peroxy radicals were found in areas were the boundary layer tracer was transported by convection. Long term implications of the research include contributing to a growing body of knowledge on ozone formation as ozone has far reaching effects on the quality of life of plant and animal species.
A new approach to GPS multipath visualization
A new approach to GPS multipath visualization
Multipath is a condition where the transmitted radio signal is reflected by physical features or structures, creating multiple reflections of the same signal arriving at the receiver at different times. The result is degradation in signal strength of the transmitted signal from the satellite to the GPS antenna. Multipath occurs when transmitted signals do not go directly to the GPS antenna, but rather arrive from different parts of the environment. These additional reflected signals causes distortion of the direct signal to GPS antennas, but proper positioning can minimize multipath error. Reception of bounced signals at the antenna causes erroneous data from the GPS receiver which results in inaccurate measurement of position. The GPS receiver has trouble distinguishing between reflected signals from direct signals and that is some of the problems multipath produces. To minimize the multipath error, positioning the GPS antenna from a location that is less susceptible to multipath can help the receiver accept amplified signals. Furthermore, a MATLAB simulation was developed previously that predicts multipath based on site analysis data to generate the plot of vectors on a Digital Terrain Model (DTM).This work produces a three dimensional plot of ray paths when signals are being transmitted from a satellite. This ray path visualization enables a user to properly position a GPS antenna to minimize the multipath error.

Pages