Today was the warmest day of the year so far for our region, with several stations getting into the 80s. And a short drive could result in a 20°F temperature change.
To illustrate, here are the high temperatures today (Thursday). 60s on the coast, mid-70s in central Puget Sound, lower 80s around Portland, 70s and low 80s in eastern Washington.
The contrasts were remarkable. Here in Seattle, temperatures ranged from the upper 50s and 60s near the water to the lower 80s on Mercer Island or on the eastside.
The high-resolution model forecasts for 5 PM Thursday pretty much picked up on most of this pattern.
But now the problem.
The warmth and sun have clearly shown that a significant number of temperature sensors in our region are broken, with those temperatures being way too warm.
The National Weather Service forecast office, adjacent to cool Lake Washington, reported a high of 90°F, while nearby stations only rose to the upper 70s. It's wrong. A little embarrassing....😂
Or consider Friday Harbor in the San Juans. It has a high of 79°F, while nearby stations were in the 60s. Clearly wrong.
Next, 87F in Darrington, with nearby stations in the low 70s and 60s.
Finally, it was 96°F east of Pasco, with no station within 10°F of this extreme value.
I could show you a dozen more of these. Temperature errors are widespread, and most problematic stations are too warm. There are many reasons for this type of error, including broken fans or poorly sited temperature instruments (e.g., direct sun exposure).
Such errors can enhance warming trends unrealistically.
Now the shocker. I checked with the head of a local National Weather Service office whether there was any automatic monitoring of temperature sensor performance at local observing sites. Software that would make all kinds of temperature quality checks, including comparison with other local sites.
Surprisingly, the answer was no. Doubly surprising since writing such software would be easy.
Not good.
So my pro-tip for the day: if you learn of an impressive temperature record at a location, make sure the reading is reliable.
________________________
I will be talking at the UW Olympic Natural Resources Center at 7 PM on Tuesday, May 6.
Wow, those are significant temperature variations at the official stations. That coupled with changes around the sensors such as the SeaTac could change how much climate change there is.
ReplyDeleteSetting sensors in a realistic setting is such basic science. How in the world, or who in the world, set those stations up
ReplyDeleteIn the 1980's I ran a company that published large natural resource data sets on CD-ROMs. One of those sets was the daily values from the National Climate Data Center in Asheville, NC. It was clear at that time that there were many anomalies in the record and we had the intention to troll for those mistaken values and offer a "correction" best-guess layer. Alas it didn't happen but as you say it would be an easy thing to do. Urbanization in general has the effect of skewing the data higher.
ReplyDeleteWorked in software dev for decades. The errors and/or missing specs rarely get fixed. It's always on the The Next Great Thing.
ReplyDeleteCliff, would these wrong readings still be used in overall temperature trend data? Or when they do those reports do they throw them out?
ReplyDeleteIt's disappointing that the NWS would accept inaccurate data, no matter how insignificant. You'd think and hope that the one bedrock, essential duty would be maintained.
ReplyDeleteGiven the cutbacks in funding and staff that the NWS has been expierencing in recent months, they may not have the money or staff to repair broken weather stations, which is very concerning because forecasting the weather is an essential public service that must be maintained and must be as accurate as possible. Yes, the private sector can do this as well, but they aren't obligated to do so and could stop at any time if they wanted, unlike the NWS. As for accepting the data, they may be figuring that its better to have some data, even if its wrong, than no data at all. Personally, I'd rather have no data rather than inaccurate data, but then I'm not the one running the NWS.
DeleteCliff, there are some issues with this post that I’d like to point out:
ReplyDeleteFor instance, the NWS Seattle WFO recorded an official corrected high temperature yesterday of 74F.
The Darrington station you point out is owned by the Washington State Department of Ecology and does not produce measurements that are considered official by NOAA.
The station just south of the Columbia River to the east of Pasco is owned by Union Pacific Railroad and, again, does not produce measurements that are considered official by NOAA.
The Friday Harbor Airport ASOS, which really should be reliable especially compared to the non-NOAA affiliated stations, also does not produce measurements that are considered official by NOAA.
I think you’ll find that, among the observing sites the data for which can be accessed via noaa.gov or xmacis, none are actually recording obviously erroneous data despite occasional incorrect measurements which are quickly identified and discarded.
My dad installed his weather station in a place that gets affected by direct sunlight, especially in the early afternoon. As a result, the temperature reading on sunny days prematurely peaks in the early afternoon. I keep on having to tell him that the high temperature readouts he gives me are inflated, but I don't think he really cares, otherwise he would've moved the sensor to a well-shaded area.
ReplyDelete"Scientists also make adjustments to account for station temperature data that are significantly higher or lower than that of nearby stations. Such out-of-the-ordinary temperature readings typically have absolutely nothing to do with climate change but are instead due to some human-produced change that causes the station readings to be out of line with neighboring stations. By comparing data with surrounding stations, scientists can identify abnormal station measurements and ensure that they don’t skew overall regional or global temperature estimates....While such data adjustments can substantially impact some individual stations and small regions, they barely change any global average temperature trends." NASA
ReplyDeleteI live just north of Sedro woolley at about 250 feet elevation and the 85 temp shown on your high temperature map seems realistic to how it felt outside. Driving into town 1 mile south 70 feet elevation was significantly cooler. Could there have been inversions or some other explanation than poor temperature readings yesterday.
ReplyDeleteA few years ago the station at the Ellensburg airport (KELN) appeared to be reporting warmly. Your Mark Albright worked to have it investigated. The NWS responsible for the station is in Pendelton OR, roughly a 3 hours drive – one way.
ReplyDeleteIt took a while to get a fix, but I think those warm readings remain in the record. Mark would know more than I do. I just live 5.25 miles NNE of the site and a bit higher and frequently compare my readings with theirs. As I write KELN says 79° and I'm getting 77°.
FYI, the OMB released its proposed 2026 budget to Congress today, May 2. In the accompanying letter to Senator Susan Collins, Vought singles out “climate” and writes that "The recommended funding levels result from a rigorous, line-by-line review of FY2025 spending, which was found to be laden with spending contrary to the needs of ordinary working Americans and tilted toward funding niche non-governmental organizations and institutions of higher education committed to radical gender and climate ideologies antithetical to the American way of life." The word "climate" appears 30 times in the proposed budget. For example, the $1.5 billion budget cut to NOAA "cancels contracts for instruments designed primarily for unnecessary climate measurements rather than weather observations" and "terminates a variety of climate-dominated research, data, and grant programs." My interpretation of Executive Branch philosophy is that they want to eliminate climate research while preserving weather research. In my opinion that’s not wielding a scalpel. That is hacking away at the body of atmospheric and oceanographic research.
ReplyDeleteGotta say? The above thread of comments from May 1 and 2 are well-written, without twisting to fit a particular political agenda. Thank you folks for sharing your thoughts without trying to work your political beliefs in there too!
ReplyDeleteApproximately 20 years ago Anthony Watts organized a citizens' group to check out the siting of the 1221 stations comprising the USHCN program. My wife and I checked 30 of the stations and found only 2 to be in compliance with NOAA standards; many in gross nonconformity.
ReplyDeleteWatts recently rechecked many of the stations and found that they had deteriorated further. Ironically, the USCRN network was created in 2005 which is state of the art and properly sited. However. the USCRN data is used to correct the degraded USHCN stations.
The original Watts study should only 15% of the stations to be compliant. The recheck recently showed that only about 7% were compliant.