Tuesday, 29 August 2023: 5:15 PM
Great Lakes BC (Hyatt Regency Minneapolis)
Quantitative Precipitation Estimation (QPE) from radar measurements for snow is challenging due to large variability in particle size distributions (PSDs), densities, shapes, sizes and orientations of falling snow. Almost all current snowfall retrieval algorithms are based on horizontal reflectivity (Z) and are related to snowfall rate or liquid water equivalent by power law relationships. The Sekhon and Srivastava 1970 S(Z) relationship has been used in Canada to convert snow horizontal reflectivity to liquid water equivalent precipitation rate (mm/h) for decades and is still being used for many operational products.
Environment and Climate Change Canada (ECCC) is nearing the completion of the upgrade to a polarimetric weather radar network and the renewed network provides polarimetric measurements in diverse weather conditions and in varied precipitation phases across the country. New relationships for snow water equivalent combining differential reflectivity (ZDR), specific differential phase (KDP) and Z are explored to determine whether they can improve overall snow QPE in an operational environment. Snow liquid water relationships utilizing ZDR in an operational environment are limited by biases in ZDR at individual radars in the network and by the stability of ZDR over long periods. Though we have implemented bias estimation and corrections in post processing software, snow liquid water estimates that employ ZDR are deferred. At ECCC a hybrid algorithm comprising of S(Z) and S(Z, KDP) from Bukovcic et al. 2018 for winter precipitation has been recently implemented in operations. S(Z) is applied for relatively low Z and low KDP, and S(Z, KDP) is applied for larger values. A heavy snowfall case in Atlantic Canada illustrates the performance of the new composite snow QPE algorithm, where the estimates compared better with surface measurements than the conventional method. However, uncertainties in surface measurements must also be considered.
Surface snow measurements are obtained from ECCC climate station networks and they provide hourly precipitation liquid amounts from automated precipitation gauges. A secondary source of surface snow measurements is obtained from the Community Collaborative Rain, Hail & Snow Network (CoCoRaHS) for melted snow amounts. Gauges tend to measure low precipitation amounts, as light snow can be deflected away from the collector by wind flow around the gauge. These observations require systematic correction depending on gauge and shield configuration, precipitation phase, temperature, and wind speed.
The presentation will highlight the ongoing analysis and performance of the snow QPE algorithms in use across Canada for winters 2021 and 2022 in comparison with the surface observations. The analysis will provide insights to improve radar liquid water estimates and allow for tuning of thresholds and algorithm coefficients.
Environment and Climate Change Canada (ECCC) is nearing the completion of the upgrade to a polarimetric weather radar network and the renewed network provides polarimetric measurements in diverse weather conditions and in varied precipitation phases across the country. New relationships for snow water equivalent combining differential reflectivity (ZDR), specific differential phase (KDP) and Z are explored to determine whether they can improve overall snow QPE in an operational environment. Snow liquid water relationships utilizing ZDR in an operational environment are limited by biases in ZDR at individual radars in the network and by the stability of ZDR over long periods. Though we have implemented bias estimation and corrections in post processing software, snow liquid water estimates that employ ZDR are deferred. At ECCC a hybrid algorithm comprising of S(Z) and S(Z, KDP) from Bukovcic et al. 2018 for winter precipitation has been recently implemented in operations. S(Z) is applied for relatively low Z and low KDP, and S(Z, KDP) is applied for larger values. A heavy snowfall case in Atlantic Canada illustrates the performance of the new composite snow QPE algorithm, where the estimates compared better with surface measurements than the conventional method. However, uncertainties in surface measurements must also be considered.
Surface snow measurements are obtained from ECCC climate station networks and they provide hourly precipitation liquid amounts from automated precipitation gauges. A secondary source of surface snow measurements is obtained from the Community Collaborative Rain, Hail & Snow Network (CoCoRaHS) for melted snow amounts. Gauges tend to measure low precipitation amounts, as light snow can be deflected away from the collector by wind flow around the gauge. These observations require systematic correction depending on gauge and shield configuration, precipitation phase, temperature, and wind speed.
The presentation will highlight the ongoing analysis and performance of the snow QPE algorithms in use across Canada for winters 2021 and 2022 in comparison with the surface observations. The analysis will provide insights to improve radar liquid water estimates and allow for tuning of thresholds and algorithm coefficients.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner