The goal of this project is to identify the sources and effects of some of the systematic biases that occur in radar measurements of neutral and plasma dynamics in the atmosphere and provide a means to correct for their effects. This will be done by using simulated radar backscatter in high-resolution numerical simulations of turbulence processes for realistic atmospheric sources. In previously funded research, a method was developed to assess radar backscatter from a modeled turbulence field that was computed by direct numerical simulations (DNS) of Kelvin-Helmholtz (KH) instability at high Reynolds numbers. The method makes no assumptions about the character of radar backscatter (turbulent, specular, or other). The method was tested with, and applied to, very high resolution DNS of KH billow evolution, turbulent breakdown, and restratification in order to aid in the interpretation of, and the potential measurement biases induced by, radar measurements of such dynamics throughout the lower and middle atmosphere. These potential biases in radar estimates may have important implications for the radar measurements of both the small-scale dynamics and the mean quantities, among them mean horizontal and vertical velocities, wind shears, and turbulence intensities and dissipation rates. This project will extend the radar backscatter computations to a number of other cases, including both zenith and off-zenith measurements of KH instability for a range of Richardson numbers, given their very different dynamics, turbulence evolutions, and likely implications for mixing and measurements, and both zenith and off-zenith measurements accompanying turbulence due to gravity wave (GW) breaking for a range of GW parameters, superpositions, and mean wind environments. The KH instabilities are expected to exhibit specular reflections that bias horizontal and vertical velocity estimates and depend on radar frequency and beam width, and spectral widths (often used to estimate energy dissipation rates) that are difficult or impossible to quantify due to turbulence impacts on refractive index fluctuations and distributions. GW applications will examine the impacts of turbulence and refractive index fluctuations that are known to vary strongly within the GW field and with time on radar measurements of GW parameters, including amplitudes, momentum fluxes, dissipation rates, mixing, and heat fluxes, all of which have substantial, but poorly quantified, impacts on atmospheric structure and variability, as well as implications for modeling and parameterization. Undergraduate and graduate students will participate in the project as well as postdoctoral scholars. The broader impacts of the research are that it will provide an improved understanding of radar backscatter and will enable correction of measurement biases. This may lead to improvements in quantifying the dynamics underlying turbulence generation, spectral evolution, and mixing and transport processes in geophysical fluids and better constraints on parameterizations of small-scale GW and turbulence effects in atmospheric (and oceanic) circulation, weather, and climate models.
Most people think of atmospheric phenomena in terms of the weather. Hurricanes, tornados, fronts, high and low pressure zones - all terms used by the professional meteorologists to describe well-defined conditions of temperature, pressure, cloud-cover and type, humidity, wind velocity, etc... Predicting the weather has become very important to our modern society. The meteorologists, who make the predictions, often turn to computer models these days to help them. Several models are used to come up with a forecast because each has different strengths and weaknesses. Another group of scientists, climatologists, try to forcast long-range trends in the average behavior of the weather, or climate. They too turn to computer models for their predictions of climate change based on physical models, past trends, and human factors. The problem that all weather and climate models have in common is mapping the entire earth, or a portion of it, including terrain, land features, water features, and the entire height of all of the atmosphere. Computer power and memory have incressed in recent years but it is still impossible to simulate all but fairly crude approximations the earth-atmosphere-ocean system. Compensation for the lack of spatial and temporal resolution is done by including models of the smaller scale effects, which is where this project fits into the bigger picture. The earth's atmosphere actually extends hundreds of kilometers (the troposphere only included the bottom 10-15 kilometers). Understanding the regions of the atmosphere above the troposphere is very important because over longer time-scales the upper atmosphere is tightly coupled to the troposphere. Energy and momentum are exchanged by wave motions generated at lower height regions that propagate upwards. At high enough altitudes the wave break, much like a wave breaking on the beach, heating, and/or cooling the surrounding atmosphere in the process, as well as altering the general circulation. Theorists, modellers, and experimentalists are all working to understanding this complex system. The experimentists use ground- and satellite-based instruments. Our study looks at how radars "see" the upper atmospheric motions. Most people are familar with the Doppler radars located usually at airports that provide wind estimates in the troposphere. The radars that look at the upper atmosphere are very similar except that much more power is needed. Clear-air turbulence - small variations in the index of refraction - scatter the radar signal back to the radar receiver, where the signal is interpreted into wind velocity. Understanding the nature of the clear-air turbulence and how its structure effects the accuracy of the derived wind vector is the focus of our work. Using the output of fluid simulations of different types of instabilities are generated and input into a virtual radar. Using this method we can compare the derived winds against the real values to understand the biases introduced by the structure of the turbulence itself. As this type of study merges with the actual radar data we can understand the processes that generate the turbulence in the upper atmosphere as well as improve the velocity estimates by understanding how the turbulence in the radar volume affects the backscattered signal and introduces biases into the derived quantities.
