Meteorology Senior Theses Undergraduate Theses and Capstone Projects 12-2016 Rainfall Estimation from X-band Polarimetric Radar and Disdrometer Observation Measurements Compared to NEXRAD Measurements: An Application of Rainfall Estimates Brady E. Newkirk Iowa State University, [email protected] Follow this and additional works at: https://lib.dr.iastate.edu/mteor_stheses Part of the Meteorology Commons Recommended Citation Newkirk, Brady E., "Rainfall Estimation from X-band Polarimetric Radar and Disdrometer Observation Measurements Compared to NEXRAD Measurements: An Application of Rainfall Estimates" (2016). Meteorology Senior Theses. 6. https://lib.dr.iastate.edu/mteor_stheses/6 This Dissertation/Thesis is brought to you for free and open access by the Undergraduate Theses and Capstone Projects at Iowa State University Digital Repository. It has been accepted for inclusion in Meteorology Senior Theses by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Rainfall Estimation from X-band Polarimetric Radar and Disdrometer Observation Measurements Compared to NEXRAD Measurements: An Application of Rainfall Estimates Brady E. Newkirk Department of Geological and Atmospheric Sciences, Iowa State University, Ames, Iowa James Aanstoos – Mentor Department of Geological and Atmospheric Sciences, Iowa State University, Ames, Iowa ABSTRACT This paper presents the comparison of rainfall estimation from X-band dual-polarization radar observations and NEXRAD observations to that of rain gauge observations. Data collected from three separate studies are used for X-band and rain gauge observations. NEXRAD observations were collected through the NCEI archived data base. Focus on the Kdp parameter for X-band radars was important for higher accuracy of rainfall estimations. The Unidata IDV was used to evaluate the rainfall estimates of NEXRAD observations. The rainfall estimation models were then evaluated versus gauge data using significance testing. It is shown that the X- band radar has a lower p-value thus indicating that the X-band radar is better in predicting rainfall accumulations. Average rain intensity (ARI) was then calculated approximately every 30 minutes to show that the NEXRAD radar is able to show periods of increase rain rate. The results show that the methodology works well at interpreting rain rates and rainfall estimations by X-band and partially by NEXRAD for the events analyzed herein. ______________________________________________________________________________ 1. Introduction estimates have been performed by radar Improving rainfall estimates from wavelengths at S band and C band. Longer radar measurements has been one of the radar wavelengths are the obvious choice for priorities of radar meteorology. The use of measurements in moderate to heavy rain radar polarization parameters in the linear because of low attenuation and backscatter polarimetric basis is generally accepted to phase shifts effects (Matrosov 2002). X- improve quantitative estimates of rainfall band radars have advantages over longer rate (Matrosov 2002). There are many wavelengths that make them a convenient different tools to collect rainfall estimations. tool and an appropriate choice for some Next-Generation Weather Radar practical applications. For a given (NEXRAD), X-band, C-band, S-band, rain transmitter power and antenna size, shorter gauges, and disdrometers are all tools that wavelengths offer greatly increased can be used to determine rainfall estimates. sensitivity for detecting weak targets Most research done in the field of (Matrosov 2002). radar polarimetry applied to rainfall 1 Next-Generation Weather Radar estimators and the high-resolution rain (NEXRAD) consists of a network of gauges. From this study we chose two cases Weather Surveillance Radar-1988 Doppler which were presented with both X -band and (WSR-88D) radars. Reflectivity rain gauge storm accumulations of rainfall observations from each WSR-88D are used (Table 1). to generate many operational products, In addition, Anagnostou et al. (2003) including estimates of precipitation used high-resolution X-band polarimetric developed with the NEXRAD precipitation radar (XPOL) data and the IIHR processing system (Young 2000). The gauge/disdrometer network located in Iowa precipitation estimates based by these radars City, Iowa. He then proposed a rainfall are used in NWS Forecast Offices by estimation technique that was based on meteorologists and hydrologists for algorithms that coupled along -the-ray guidance in forecasts and warnings. profiles of Z aH (r), Z aV (v), and ΦDP (r), where NEXRAD is combined with gauge data to ZaH (r) and Z aV (v) are the attenuated rad ar make multisensory precipitation estimates. reflectivity in the signal phase at horizontal (H) and vertical ( V) polarization for the 2. Background XPOL measurements, ΦDP (r) is the The best way to get accurate data in differential phase shift between H and V X-band rainfall rates is to compare them polarization (degrees), and r symbolizes the with gauge and disdrometer data. In range bin along the ray. An experimental Matrosov (2002), 15 observed rain events stu dy by Matrosov in 2002 provided a were quantitatively compared by rainfall quantitative error analysis of the various rain accumulations from different radar estimators based on field data (Anagnostou 2 2003). This study is based on a field shown in Matrosov (2004) that the estimates experiment which collected data from an X- of storm accumulation from a nearby band polarimetric radar, XPOL, over a NEXRAD (KMUX) were much lower than densely instrumented site with three actually observed for these events. This can disdrometers and several tipping-bucket be mainly attributed to the much higher dual-gauge (0.25 mm) platforms located altitude of NEXRAD scans (compared to the between 5 and 10 km from XPOL X-band scans) over the area of interest in (Anagnostou et al. 2003). The observed this study. The NOAA X-band rainfall events occurred for the months of transportable polarimetric radar has shown October and November 2001 and ranged in that it is capable of providing high- intensity from moderate stratiform resolution and accurate estimates of rainfall precipitation to high-intensity (>50 mm h -1) parameters. While Lim et al. (2013) convective rain cells. From this study we presented new methods for rainfall chose one event on October 22-23 (Table 1) estimation from X-band dual-polarization where each rain gauge cluster is indicated as radar observations. Dual-polarization radars A, B, or C (Figure 1). have become an important tool for X-band radar has been traditionally meteorological applications. This has limited in its applicability for quantitative occurred over the last decade and it is precipitation estimations (QPE). This is due important for tools such as quantitative to a relatively high attenuation rate of radar precipitation estimations (QPE) and signals in rain. The use of polarimetry hydrometeor classification. However, provides a new tool for correcting various methods for hydrometeor attenuation effects, which in turn greatly classification have been proposed during the increases the utility of X-band radars for last decade but these methods were mainly QPE. In Matrosov (2004), there were five for C and S band weather radars. The significant landfalling storms at Fort Ross proposed hydrometeor classification (FRS), California, near the mouth of the described in Lim et al. (2013) is a robust Russian River, in an area that has poor technique that can be applied to S band, C coverage by the NWS WSR-88Ds. These band, and even higher frequencies such as X storms were observed by the NOAA X-band and K u bands. Also, the rainfall estimation radar. Also, during this study there were included a comparison of both instantaneous three high-resolution (0.01 in.) tipping- and cumulative rainfall using the bucket rain gauges used for validating conventional, filtered specific differential rainfall retrievals. These were located at the phase (Kdp ) and the proposed method with Salt Point State Park, Goat Rock, and surface rain gauge and disdrometer data Bodegy Bay. They recorded the number of (Lim et al. 2013). Data collected from the 0.01-in. tips every 2 min. The Hydrometeorology Testbed (HMT) in the accumulations collected in this study by the orographic terrain of California are used to X-band radar were then compared to the demonstrate the methodology. The study three high-resolution (0.01 in.) tipping- this paper presents is a further enhancement bucket gauges. Collecting rainfall to a class of applications in improving the accumulations for given time intervals robustness of the rainfall estimation process. especially over the entire course of a storm This works well especially for complex present very important hydrological terrain applications due to reduced ground information (Matrosov 2004). From this clutter contamination at K dp retrieval (Lim et study we chose two events (Table 1). It was al. 2013). 3 Throughout these studies a common to know which radar is best in determining theme seems to be multi-parameter rain accumulations to better our forecasts. estimates that include the Kdp parameter. To better the results of this paper we used this 3. Methods estimator for our X-band rain accumulations which can be seen in Table 1. The multi- a. Setup parameter estimator gave the best agreement with the rain gauge. The other important The locations used for this