Winnipeg and the Red River Valley will get a bit of a reprieve from the unsettled weather of late today with some sunshine and daytime highs in the mid-teens for Winnipeg and closer to 20°C for regions in the Southern Red River Valley. Conditions will deteriorate tonight, however, as Southern Manitoba deals with another low pressure system that will move through on Thursday.

Clouds and showers will push into the Red River Valley tonight as a low pressure system makes it’s way through North Dakota. A very slight risk for a thundershower exists along the US border, however the odds aren’t very likely. For everyone else, just plain old rain will be pushing your way. Models disagree on how far northwards the precipitation will push: the American models keep the northern edge on a line that cuts from Morden to Bisset, south of Winnipeg, while the Canadian models push the precipitation as far north as Gimli. It seems fairly certain that the southern half of the Red River Valley will see showers tonight and Thursday; Winnipeg will likely see at least a few passing showers out of this next system, with a chance that we’ll see rain tonight through much of tomorrow.

For areas that do see steady rain, total accumulations should be in the 5-10mm range with localized areas that see some convective enhancement seeing as much as 15-20mm. My current feeling is that Winnipeg will see at least 5-10mm through Thursday evening.

Rainfall in systems like this is typically hard to predict because it is so dependent on the initiation of elevated convection at night; something that models do not have a great handle on and humans struggle with not because we don’t understand it, but because there’s simply so little data to work with. Small errors in the location or strength of initiation of the nocturnal convection can result in huge errors in the model for later times, which can cause the model to end up forecasting tons of rainfall for one location that ends up seeing nothing and completely missing other areas that might see plenty of rain. For this reason, working off of real data is often the best way to go; satellite & RADAR imagery can be a forecaster’s best friend in dealing with nocturnal convection. A drawback to this, though, is that you can’t make your forecast as early. As soon as things begin to take shape, a forecaster can act quickly and make a fairly accurate forecast for the next 12-24 hours. It’s crucial to make sure that the starting point is correct, though, which is why when dealing with convectively driven situations, it’s important to talk in terms of probabilities and likelihoods while still a ways away from the event actually happening.