A Novel Approach to Stratifying the Precipitation Time Series: A Precipitation Climatology for Montréal, Québec

How do we partition a precipitation time series? Most studies either define precipitation events based on 24-, 48- or 72-h accumulations, or, they endeavour to identify precipitation events based on continuous sub-daily precipitation observations. This latter method usually requires a maximum event duration (e.g. 72 hours), as well as a minimum event separation (e.g. 12 hours). Regardless of the merits of these techniques, they have at least one commonality: they subdivide the precipitation time series into temporally-based events. Our proposal is to utilize our understanding of atmospheric dynamics to develop a mechanism-based approach to stratify the precipitation time series.

The adiabatic and frictionless form of the quasigeostrophic (QG) omega equation provides meteorologists with a means to qualitatively diagnose vertical motion in the atmosphere. Knowing the positions of the low-level cyclones and anticyclones relative to the upper-level troughs and ridges, we can infer the contributions of the low-level temperature advections and upper-level vorticity advections towards omega. The QG-omega equation then enables us to qualitatively diagnose regions of: unambiguous ascent (where both temperature and differential vorticity advections are positive), unambiguous descent (where both temperature and differential vorticity advections are negative), and ambiguous vertical motion (where temperature and differential vorticity advections oppose one another). Consequently, we can break up an upper-level wave into four quadrants based on the locations of the 500-hPa trough and ridge axes, and the respective inflection points in order to qualitatively diagnose the vertical motion.

In this study, we use the North American Regional Reanalysis (NARR) 500-hPa geopotential heights to identify upper-level troughs, ridges, and inflection points. We then partition the 1979-2018 precipitation time series from Montréal, Québec into the four aforementioned quadrants. The end result is a 40-year precipitation climatology, where we investigate the following question: does precipitation in the midlatitudes conform to our expectations from the QG-omega equation?

Our results show that the quadrant where we expect unambiguous ascent (hereafter, quadrant 3) is indeed characterized by significantly higher precipitation amounts (and rates) compared to the other quadrants in all seasons. We also observe that the quadrant where we expect unambiguous descent (hereafter, quadrant 1) has the lowest mean precipitation amounts (and rates) compared to the other quadrants in all seasons. When comparing the two quadrants of ambiguous vertical motion, we note that precipitation amounts (and rates) are higher in the quadrant where cyclonic vorticity advection (CVA) contributes to upward motion (hereafter, quadrant 4) than in the quadrant where warm air advection (WAA) contributes to upward motion (hereafter, quadrant 2). To further explore these results, we also analyzed the 850-hPa equivalent potential temperatures (θ_e) and 700-hPa omega in each quadrant.

Quadrant 3 is characterized by significantly lower values of omega (higher values of ascent) compared to the other quadrants in all seasons except for summer. Moreover, quadrant 3 also features significantly higher θ_e anomalies than the other quadrants in all seasons. On the other hand, quadrant 1 has the highest mean value of omega (lowest values of ascent) in all seasons, and significantly lower θ_e anomalies compared to the other quadrants in all seasons. Quadrants 2 and 4 have similar θ_e anomalies in each season, but quadrant 4 distinguishes itself from quadrant 2 with significantly lower values (greater ascent) of omega in all seasons. If we refer to 850-hPa θ_e as a thermodynamic metric and 700-hPa omega as a dynamic metric, this last result may indicate that, in the mean (for Montréal), dynamics play a bigger role in quadrant 4’s heavier precipitation (compared to quadrant 2), as compared to thermodynamics.

Overall, the quadrant partitioning method adapted from the QG-omega equation inherently provides physical meaning to the stratification of the precipitation time series and allows for easier interpretation of the climatological results.