Research

Climate change and severe convective storms

I work to try to reduce the uncertainty around how thunderstorms and convective hazards may be affected by climate change. I led a team of authors to write a review paper about the responses of hailstorms to climate change, published as Raupach et al. (2021) in Nature Reviews Earth and Environment. The article is available in free-to-read form. In the review we sum up the current state of knowledge and make recommendations for future study. This review was featured by the UNSW newsroom and subsequently covered by newspapers in Australia; I was interviewed for Australian radio outlets National Radio News, ABC radio Broken Hill, and Hope FM.

In Raupach et al., 2023a we developed an improved hail proxy, that indicates whether a given atmospheric environment is hail-prone or not, for Australia. In Raupach et al., 2023b, we applied this proxy to produce the first continental-scale estimates of trends in hail-prone environments in Australia for the past four decades.

I contributed to the ARC Centre of Excellence for Climate Extremes State of Weather and Climate Extremes 2022 report, which contains a section about some of the hailstorms that affected Australia in 2022.

I was an invited speaker at the 3rd European Hail Workshop in March 2021, where I gave a presentation titled “Hail in a warming climate”. I gave the keynote for the climatological studies session of the 11th European Radar Conference 2022 in September 2022. In July 2023 I was an invited speaker for a session on weather and climate extremes at IUGG Berlin.

High-resolution modelling of convection

I am currently working on using high-resolution models to study the changes to convection that occur with climate change. I was an early career researcher co-convenor for the session “Convection-permitting atmospheric modelling” (CL2.44) at the European Geosciences Union (EGU) General Assembly 2020.

Measuring the raindrop size distribution (DSD)

The raindrop size distribution (DSD) is a statistical description of the number and size of the water drops that make up rain. The DSD is important for measuring, modeling, and understanding rainfall, and it was the subject of much of my PhD work. The DSD is measured at ground level by instruments called disdrometers, and like any instrument there can be inaccuracies their measurements. In Raupach and Berne (2015), we proposed a method to “correct” DSDs measured using a common disdrometer (the Parsivel) using a measurements from a different instrument (the 2D-video-disdrometer or 2DVD). Updated correction factors and a list of updates are available online. In Raupach et al. (2019), we showed how an existing technique called double-moment normalisation could be used to take measurements and reconstruct DSDs to include the “drizzle mode” - the many small drops that are not very well measured by regular disdrometers. The reconstruction technique is superior to the earlier correction method, because it is more flexible and does not rely on the accuracy of a single reference instrument.

Small-scale variability of precipitation

Rainfall changes a lot in space and time — if it’s pouring in one neighbourhood, it may be only drizzling in the next. If both areas fall within a single “pixel” in radar or satellite data, which part of the rain should the pixel’s value represent? Some of my work has been on quantifying the effects of this so-called “sub-grid” variability in rain. In Raupach and Berne (2016a), we presented a geostatistical method for spatial interpolation of experimental DSD measurements. The technique can be used to take data from a network of disdrometers and estimate the complete DSD at unmeasured locations. We used the technique to stochastically simulate DSD fields in Raupach and Berne (2016b), in which we showed how sub-grid variability of the DSD affects the representativity of areal DSD estimations from satellites or models. In Raupach and Berne 2017a, we tested invariance of the double-moment normalised DSD through spatial displacement in stratiform rain, and concluded that in stratiform rain the double-moment normalised DSD can be considered spatially invariant for practical purposes.

Radar meteorology

Rain radars work by sending pulses of electromagnetic radiation into the atmosphere and measuring the signal that is reflected back by the falling rain. Radars are often used to estimate rain intensity, but more information about the rainfall — such as information on the DSD — can be gleaned by using polarimetric radar in which the beam is polarised into horizontal and vertical orientations. In Raupach and Berne 2017b, we proposed a flexible technique for estimating the full DSD from polarimetric radar data, that uses the double-moment normalised DSD. A modified version of this technique was used to generate high-resolution DSD fields that were used to compare urban rainfall monitoring techniques in de Vos et al. 2018. We showed a case study using an updated version of the retrieval method for radar retrieval of lower-order DSD moments in Bringi et al. 2020.