While it may sounds like a climatologists attempt to cover all the bases, it was a wet, dry, or very dry May depending on where you were in the Northwest. Nearly all of the inland NW was well below normal including a nearly two-week period mid-month with nominal precipitation across the region. Respectively, the inland NW receives more of their annual precipitation in May than locations further toward the coast.
By contrast, Seattle and western Washington continued their impressive comeback of the water year. Per the graph below on Feb 1st Seattle was at only 50% of their cumulative water-year precipitation and currently are just a tad above normal. Per NWS Seattle, precipitation from Feb-May has thus far not only beat the previous Feb-May record, but also the Feb-July record.
The comeback fell short in Spokane, where just over a half-inch of precipitation fell in May, over an inch below normal, with current water year precipitation of 73% of normal, the driest since the drought of 2001.
Finally, Pasco WA, where under 3″ of precipitation has fallen since October, making it the driest since the drought of 1977. The only bright side is that they are only 4″ below normal to date; however, the most precipitation they’ve ever had from June-Sep is 4″. Fortunately, irrigated cherry crops in the Columbia basin are doing quite well do to a lack of turbulent weather during the bloom and barring any substantial rain between now and harvest should be in good shape, particularly given the lackluster California crop.
Temperatures were a bit above normal for May, most notably for daily high temperatures given the relatively high number of sunny days sans precipitation during the month.
Climatological spring (March-May) across the inland northwest had near normal to slightly above normal temperatures and slightly below normal precipitation (with exceptions). As a Lagrangian observer of climate, this spring has been a pleasant surprise from five of the previous six springs which were all cooler and wetter than normal and led me to hypothesize that spring is among the least favorite seasons of NW citizens. A couple informal and unscientific polls in my classes (n=50) has yielded only a couple of folks that grew up in the NW that dared call spring their favorite season. This could be due in part to the recent string of crummy springs though. With tons of solar insolation (astronomical spring has same number of hours of daylight as astronomical summer), spring always has so much potential, but typically fails… (courtesy of the San Diego tourism bureau)
Gauging Precipitation Gauges
Precipitation measurements are incredibly valuable and are used in hydrologic monitoring and countless landscape scale assessments. As part of a senior capstone project in my department at the University of Idaho, Jet Johnstone collected daily precipitation observations from a few precipitation networks over the past four years including data from CoCoRAHS, Coeur d’Alene tribe and the Priest River Experimental Forest. We collected these data since they had not been included in the various gridded precipitation products include PRISM and Daymet and thus could be considered independent data for validation purposes. Jet compared monthly precipitation between each station and it’s co-located gridded precipitation estimate only on monthly where no more than 5 days of daily precipitation measurements were missing. He further eliminated any stations that didn’t have at least 12 months of data over the time period.
The maps below show the biases, calculated as the gridded estimate divided by the station observation times 100 minus 100, with blue colors showing where the gridded precipitation estimates exceed those observed, and red showing where the estimates are less than observed.
Of course, it is very tough to know what is “correct” when it comes to these measurements as observing precipitation is notoriously tough, and is made tougher when a better chunk of it falls as snow. Interestingly, we found that the gridded precipitation estimates had the most acute positive biases in winter which could very well be a function of observers having troubles (yours truly included) melting precipitation amounts. Also, there are issues of gauge undercatch that tend to be magnified during snowfall and in windy conditions. Nonetheless, there are some interesting patterns that emerge in the maps that Jet made. First, he tends to find that PRISM does a slightly better job than Daymet in this analysis in terms of biases with the overall PRISM bias of about 10% (close to what we’d expect for undercatch). Secondly, there are a few regions of decent station density that are routinely biased in the same direction. This includes the Moscow region, where the 5 stations all have a 10-30% positive precipitation bias. Data produced by the gridded precipitation datasets comes from official nearby stations like the Moscow COOP station. I’ve also been curious as to how the Moscow COOP station is so much wetter than what I’ve recorded (one of the CoCoRAHS stations in the analysis) even though I am <1 mile as the crow flies and at the same elevation. I’ve blamed it on crummy recording on my behalf, but apparently there is some consensus in the surrounding stations.
However, the preponderance of biases <10% suggests that (i) these observations probably are decent, and (ii) the algorithms used to deduce gridded precipitation are decent. Unfortunately, the best way to test this out would be to have a lot more observations in the mountains, which are horribly under-sampled.