Nowcasting Canadian Wildfire Smoke and Improving Air Quality Forecasting with Ceilometers

Wildfire smoke continues to be an health hazard for Canadians and wildfire events are increasing in frequency with climate change. PM2.5 concentrations are currently monitored 24/7 by surface sensors, and satellites are used to detect the presence of smoke, with high spatial but low temporal cadence. However, there is a significant gap in information between the surface and the top-most layers observed by satellites. This lack of information hampers air quality and aviation forecasting as the height of smoke plumes determines if,when, and where they will reach the surface. Ceilometers and lidars can bridge this gap by providing the location and thickness of smoke layers throughout the boundary layer and into the free troposphere. Lidar networks, such as the Canadian Micropulse Lidar Netowrk, part of NASA’s global Micropulse Lidar Network, and the European E-PROFILE network, have already proven extremely useful for detecting and tracking smoke and other particulates in near-real time

Environment and Climate Change Canada’s (ECCC) Meteorological Service of Canada (MSC) has begun testing of the next-generation Vaisala CL61 ceilometer in Edmonton, Alberta. This new instrument improves on the CL51 ceilometer by providing, in addition to the elastic return, the depolarization ratio, which allows for the differentiation of particulate type. The CL61 has been operating alongside a Vaisala CL51 since September 2022. We present a summary of the CL61’s performance over the last two years. We find that the CL61 measurements have remained stable over the entire period, with no significant drift and agree well with the CL51 measurements.

Additionally, several wildfire events over the last two years are used to assess ECCC’s operational wildfire smoke forecast model’s (FireWork) performance. We compare model forecast performance for surface PM2.5 against co-located Purple Air sensors, as well as its performance at forecasting elevated smoke plumes with the CL51 and CL61 ceilometers. In some cases, we find that the forecast model under-estimates surface concentrations and may miss some elevated plumes. We then investigate the relationship between ceilometer/model elevated plume agreement and model surface PM2.5 performance.