Abstract: It is well established that the Earth's water cycle is accelerating, and extreme precipitation events are becoming more common. While we cannot avoid this issue, we can be better prepared to handle it if we can obtain accurate observations of precipitation for use in short-term and long-term prediction models. Various remote sensing instruments are available to obtain precipitation data. In this research work, mapping precipitation at local, regional and global scales is studied. The technology of precipitation mapping at these scales is very different and elaborated. Examples of precipitation measurements from these scales are discussed.

At the local scale, rain gauges and disdrometers are two prominent instruments that are utilized for precipitation measurement. Precipitation observations captured from these two instruments are introduced. Millimeter wave radars have been previously used in various domains, and extensive research is currently in progress to improve this technology. In this research, the potential of using automobile class radars to obtain local surface precipitation is presented. Since the maximum range of an automobile radar is within a few hundred meters, the observations can be considered to be at a local scale. With the help of signal modeling, methods to obtain the rainfall rate at the millimeter wave band by using radar parameters such as reflectivity and attenuation are discussed. A simulation tool is developed that generates the radar signals at the millimeter wave frequency band. The various parameters which are used in the signal simulations are explained in detail, and the simulation results are presented. Experiments for mapping precipitation using a current state-of-the-art automobile radar are carried out, and the results are discussed.

Weather radars are remote sensing instruments that provide precipitation observations at a regional scale. They provide data at a large spatial extent; a typical X-band radar will provide precipitation data spanning up to 40 km in range. Weather radar observations obtained at various frequency bands for mapping precipitation are discussed. The current networks of instruments that provide precipitation information at a regional scale, along with their system architectures are discussed. The precipitation data obtained from individual automobile radars can be considered as a local data point, and precipitation maps at the regional scale can be constructed. The system analysis of using a network of automobile radars for mapping precipitation is discussed with the help of simulations. The Dallas-Fort-Worth urban region is considered for the simulation study, and the potential of using millimeter wave radars to create precipitation maps is presented. A system architecture for precipitation mapping using automobile radars is also discussed.

The attenuation of radar signals has to be addressed and corrected to obtain accurate precipitation information from radar data. The attenuation correction in weather radars for rain hydrometeors is well studied in the literature, but attenuation correction for snow is limited. This is due to the fact that snow does not attenuate much at lower frequency bands like S and C bands, the snow particles vary in their particle size distributions and have complex shapes. In this research, a new algorithm that corrects for radar signal attenuation in the rain, as well as snow is introduced. The attenuation correction method developed is applied to X-band and Ku-band radar data, and the results are discussed.

Mapping precipitation at a global scale is a challenging task. The Dual Precipitation Radar (DPR) is a spaceborne instrument deployed as a part of the Global Precipitation Measurement (GPM) mission. This radar is currently operational providing valuable precipitation information at the global scale, but the observations from this instrument suffer from poor spatial and range resolutions. Synthetic Aperture Radars (SAR) are well known for providing high spatial resolution data. In the past, SARs have been deployed on airborne and spaceborne platforms for mapping land cover and constructing surface elevation models. The potential of using SAR for mapping precipitation is not widely explored. In this research work, SAR signal simulations are carried out to observe precipitation from spaceborne platforms. The simulation results for two specific SAR architectures are discussed in detail.

This dissertation presents the advantages and challenges in observing precipitation at the three scales, with suggestions for future research.

Adviser: Dr. V. Chandrasekar
Co-Adviser: N/A
Non-ECE Member: Dr. Susan P. James
Member 3: Dr. Margaret Cheney
Addional Members: Dr. S. Ryan Gooch

Publications:
Please contact sjoshil@colostate.edu for details.

Program of Study:
ECE 549
ECE 521
ATS 652
STAT 511A
ECE 580B4
ECE 742
ATS 693
ECE 799