About Canadian weather radar

Executive summary

Environment and Climate Change Canada (ECCC) uses advanced weather radar technology to track rain, snow, and other types of precipitation across Canada and parts of the United States. The system combines data from up to 200 radar stations into a single, high-resolution image that updates every six minutes. It provides detailed information on the type and intensity of precipitation, and indicates whether it is rain, snow, or another form such as freezing rain or hail.

Older C-band radars have been replaced with more powerful S-band radars, offering greater accuracy and wider coverage. These newer radars use dual-polarization technology, which improves the ability to distinguish between different types of precipitation and reduces false signals caused by birds, insects, or other interference.

The radar composite of precipitation rate (MM/HR and CM/HR)

The radar layer displayed on our platform is based on a mosaic calculated over the North American domain (Canada and the U.S.) with a horizontal spatial resolution of 1 km. This mosaic therefore includes all Canadian and American radars available in the network and can reach a maximum of 200 contributing radars. Figure 1 shows the extent of the area covered by the mosaic as well as the density of contributing radars (white circles). The coverage is marked by numerous areas of overlap between the radars, which is useful in the event of scheduled or sudden failures of certain radars. This composite is available every six minutes.

To better represent precipitation over the different seasons, this mosaic renders in mm/h to represent rain and in cm/h to represent snow. For the two precipitation types (rain and snow), we use two different mathematical relationships to convert the reflectivity into rainfall rates (mm/h rain cm/h for snow). Note also that these mosaics are available in a palette of 8 and 14 colors (intensity levels of radar echoes). Figures 2 and 3 illustrate two examples: one representing the precipitation rate of rain and the second representing the precipitation rate of snow.

This is a hybrid mosaic composed of different estimation precipitation products. Below are the technical description of each product and the context of their use:

1. DPQPE product: DPQPE stands for: Dual-Polarization Quantitative Precipitation Estimation. It is a two-dimensional representation of the precipitation rate estimated by the lowest sweep of the radar (elevation angle of 0.4 degrees for most S-Band radars). Therefore, the precipitation rate is estimated as close to the earth's surface as possible. The DPQPE product is based, among other things, on a series of polarimetric processing steps (quality control based on the dual-polarization technology) to eliminate non-meteorological artifacts from the raw data (volumetric scans). It renders in mm/h for rain and cm/h for snow. This product is calculated with a maximum coverage range of 240 km. When the DPQPE product is unavailable, the PRECIP-ET product is used as a contingency.
2. U.S. product: Environment and Climate Change Canada utilizes base reflectivity mosaics from the NOAA (National Oceanic and Atmospheric Administration) Multi-Radar-Multi-Sensor (MRMS) system, with quality control that removes ground clutter and non-meteorological echoes. These mosaics cover the contiguous U.S. as well as the Alaska and Hawaii domains. This product is calculated with a maximum range of 460 km for the majority of U.S. radars. We have adapted this composite to meet our specific needs and to ensure consistency with Canadian radar data and provide seamless integration between the two networks.

The Surface Precipitation TyPe (SPTP) Composite

The SPTP product is a radar composite with a resolution of 1 km covering the North American domain (Canada and the U.S.). For areas covered by radar, it can distinguish the presence, type of precipitation, and recognize multiple levels of intensity. The product distinguishes between five types of precipitation: rain, snow, freezing rain, hail, and mixed precipitation. For rain, snow, and rain/snow mix, three intensity levels are assigned to each (low, moderate, and heavy). As for freezing rain, two intensities are defined (low and heavy), while hail is assigned a single intensity level.

The product is based on:

• Two high-resolution North American radar composites: one for cleaned precipitation rates and another for particle classification, both using dual-polarization technology
• Surface temperature data from the HRDPS (High-Resolution Deterministic Prediction System) at 2.5 km for Canada and the GDPS (Global Deterministic Prediction System) at 10 km for the United States

The SPTP product is updated every six minutes, with a production and availability delay of approximately seven minutes after the valid time. This product is available with a maximum range of 240 km for Canadian radars and 230 km for U.S. radars.

Precipitation intensity composite versus SPTP

A pixel-by-pixel comparison between precipitation intensity and SPTP product is not advised currently since they are processed differently. This is due to the requirements of the data type for each product. The key differences between the two composites are:

• The rain (mm/h) and snow (cm/h) precipitation intensity composites Maximum Value approach to handle overlapping radar coverage. For each pixel, the system selects the highest precipitation value among all overlapping radars. This method is applied specifically in areas covered by multiple Canadian radars, as well as in border regions where both Canadian and U.S. radars overlap. It ensures that the strongest signal is retained in overlapping zones and that precipitation estimates can still be generated even if the nearest radar is unavailable by leveraging data from other radars across Canada and the U.S.
• SPTP uses multiple radar data only from the nearest radar and model information to produce a precipitation type near the surface. As a result, some echoes may be classified as non-precipitation and removed but not replaced with precipitation type from another radar.

ECCC is aware of the processing differences and working on a new approach to integrate the techniques, especially for combining U.S. and Canadian radars.

Impact of the new S-Band radars on the quality of the mosaics

In 2017, ECCC launched the Canadian Weather Radar Replacement Program, a seven-year project that replaced  the aging Meteorological Service of Canada’s weather radar network with a modern dual polarization system. This upgrade allows ECCC to deliver more accurate and timely meteorological information, which helps Canadians make better informed decisions regarding their health, safety, and protection. The result is a more reliable, affordable, and sustainable weather radar network with expanded coverage. More information on the progress of the installation of these new radars can be found here: RADAR replacement schedule.

The table below summarizes the main differences between the new S-Band radars and the previous C-Band radars:

(*): This maximum range of 330 km is available externally only for raw data (volumetric scans) in ODIM-H5 format. The maximum range for processed products from Canadian radars in this composite is limited to 240 km.

Double polarization and improvement of product quality

The new S-Band radars use dual polarization (polarimetry) technology. The basic premise behind radar weather is that, in addition to know estimate reflectivity, it determines hydrometeor shapes (measurements of the microphysical properties of hydrometeors). This is possible because dual polarization allows the radar to measure both the vertical and horizontal aspects of the targets. By measuring the returns of these two polarizations, the shape and the effects of the different shapes of hydrometeors can be estimated.

An added value of polarimetric radars is the additional information coming from the vertical polarization rather than horizontal polarization. This information makes it possible to distinguish the different types of precipitation and greatly improves the observations. This supports better quality forecasts and issuing warnings to the public in extreme weather conditions.

The benefits include:

• Significantly improve the precision of quantitative estimates of precipitation;
• Identify with precision the different types of precipitation and improve the forecasts of liquid or equivalent solid water (differentiate rain, snow, etc.) - better treatment of the bright Band;
• Simplify (or even eliminate some of them) the quality control algorithms used by forecasters to target meteorological signals in single polarization radar;
• Increase confidence in radar data, which is the basis for preparing warnings in the event of extreme weather conditions;
• Provide relevant and critical information for estimating precipitation, which helps predict flooding and flood warnings caused by rivers;
• Provide information useful to help municipal authorities in relation to water management;
• Be able to distinguish non-meteorological echoes, which can be dangerous for meteorological aviation such as birds, for example;
• Facilitate and assist the detection of icing conditions in aircraft.

Figures 5 and 6 show the positive impact of dual polarization technology on product quality. This is a comparison between the conventional CAPPI product, which does not use the dual polarization information, and the DPQPE using dual polarization. In both cases (with and without precipitations) for the same radar (Blainville radar covering the greater Montreal area) and for the same date, it demonstrates how the non-meteorological echoes over the southwestern region caused by radial interferences are cleaned when applying the dual polarization information of the DPQPE product.

Accessibility of cleaned DPQPE images

The single S-Band images, which are based on the cleaned DPQPE product and use information from the dual polarization, are available in real-time on the public site Datamart, accessible here: Single images DPQPE.

Dynamic radar coverage

Radar coverage is provided to dynamically display the zones covered by the radars every six minutes. It provides information on the availability (or not) of the contributing radars and on the areas of overlap as shown in Figure 7.