June 05, 2018

Fixing the Warm Temperature Bias At Yakima

The accuracy of temperature sensors used by the National Weather Service, FAA, and others is very important.   Agricultural interests depend on reliable temperatures to make critical decisions about planting and other issues.  And the evaluation of global warming can be undermined by biased temperature sensors.

This blog has described a number of examples of problematic surface temperature observations at a number of sites, with a recent blog noting issues at Yakima Airport in central Washington, where temperatures have consistently been too warm compared to other observing locations in the region.  These problems have been obvious for several years. 

The problem had to be the temperature sensor, since there was little urbanization in the area (see satellite map below).

The good news is that the National Weather Service has finally replaced the Yakima temperature sensor on May 8 and the problems seem to have been ameliorated.  Let me show you the impact of the new sensor and show you more about this long-standing problem.

Here is the temperature anomaly (the difference from average) for daily average temperature over Washington State for the last three years.  Repeat: three years.   Temperatures have generally been above normal for the past three years over Washington, with one location being crazy warm (red color, 3-5F above normal).   Which location?  Yakima Airport.

Last two years?  Yakima was too warm.

Last year?   Yakima is again standing out as an anomalously warm location.

Last three months?  Yakima is still very warm.

The last month, which only includes a short period with the bad sensor?  The Yakima problem is almost non-existent.

 The last week?  The warm anomaly is gone!   

The bottom line:  it is clear Yakima had a bad temperature sensor for years and that replacing the unit corrected the issue. 

But why did it take so long to identify this problem and swap out the sensor?

You might have noticed another problem station...a cool one..over the northeast side of Washington State.  Perhaps a topic of a future blog.


  1. This is part of two much bigger connected issues.

    1. The urban heat island effect has skewed readings throughout the industrialized world. I have seen articles (which I cannot cite, because I don't bookmark everything) that cite longstanding rural stations in the U.S., and show that average temps in this country have actually declined for a long time.

    2. The measurement methodology of that often-reported "global temperature" is rarely discussed. Given the documented history of the AGW lobby having introduced phony upward "adjustments" to historic temperature data, I don't trust the global "record high temperatures" that are frequently "reported" by the one-sided media.

  2. This is off today’s topic... I’m noticing today that in the mid level clouds some of them appear to have lenticular clouds on top of them... what would cause that?

  3. I'd like to know whether there is a scheduled maintence/testing program for all stations that report hourly obs.It seems that at least once or twice a year there are malfunctions,such as reporting rain when the skies are clear.

  4. This is why I run a PWS which I compare with gladstone.net to isolate abnormalities. The data from my PWS is used by my cloud irrigation controller, providing hyper local weather data that I know I can trust. Any indivdual relying on accurate forecast data should be investing in such technology as the cost barriers are minimal.

  5. I am amazed that this problem was allowed to go on so long.

  6. Cliff

    Has the defective sensor been left in place so it could be calibrated to the accurate one and Yakima records eventuall be corrected?

    Will Stewart

  7. I doubt the anomaly in NE Washington is due to a bad sensor as much as a reflection of an unusual micro-climate. I have property in the Kettle River mountains and my whole life we have received much more rain and cloudy weather than anywhere else in Eastern Washington. It is reflected in the tree species- we have very little pine on my property but mostly fir with some tamarack and quite a good amount of western red cedar. Where you find western red cedar you will find more precipitation and more moderate temperatures.

  8. I find myself in a rare situation where I almost agree with something Placeholder said.

    When assessing the reality, pace and impacts of climate change I rely hardly at all on the global record high temperatures. Metrics obtained by instruments created by humans are much less convincing to me than the many manifestations of global warming in the physical and biological world.

    The ongoing decreases in Arctic sea ice and glaciers worldwide coupled with the increases in sea level and droughts are well documented. There are also the poleward shifts in the range of subarctic and temperate species and the major decreases in less mobile species as their historic ranges warm. Farmers and others whose livelihood is dependent on somewhat predictable annual weather patterns are experiencing the rapid pace of climate change - which in some case can lead to major political unrest.

    No need for Placeholder or anyone else to be concerned with any "AGW lobby" if they truly want to make an assessment about the reality of global warming. There are plenty of observations of its current impact on the world - and unfortunately there will be more to come - which are far more convincing and important than a biased temperature sensor.

  9. Sensors go bad all the time. But why it took so long to replace a (relatively) cheap sensor and no one looked at the bad data it was throwing remains a mystery. You can correlate data drift quite easily today too.

  10. @John Frankin, the AGW lobby (or as I often say, "cult") sits in the wreckage of predictions that have not come true. Not to mention all of the outright fraud, or an edifice built on a deeply flawed research foundation. The major media outlets have bought into the ideology, so they ignore the truth about any of it, and label anyone who so much as hints at the truth as a "denialist."

  11. There's a distinct reddish cast to the sun today and the high cirrus looks suspiciously dirty and definitely atypical, especially to the south. I hope it doesn't portend an early fire season... but what is it? Dust from Hawaiian or Guatemalan volcano, perhaps? Anyone?

  12. sldulin...... a very apt observation on micro climates. Your tree species amount to a long-term witness to local climate, more in perspective than most human sensors or record keeping could provide. Perhaps some would discount the tree record because it isn't in an 'official' graph or data base but their failure is demonstrative of the short-sightedness of some who are fixated on data in spread sheets in spite of the evidence. Perspective matters !

  13. Dr. Mass, any thoughts on the incoming cold front imminently descending on the PNW? Great to see some real precipitation falling in this typically dry month.

  14. Again I find myself agreeing with Placeholder, who is concerned about "the wreckage of predictions that have not come true" regarding climate change. We are not alone in our concern that reality has not matched model output. This 2012 post in Scientific American


    points out that "As recently as a few years ago, most models of the Arctic ice anticipated that summers would remain icy until the end of the 21st century, and well into the 22nd century" and goes on to say that recent model output predicts an ice-free Arctic could occur around 2030.

    That the earlier predictions were found to be wrong, regarding just how fast sea ice loss would occur, feeds the skepticism of people like Placeholder about what climate modelers really can say regarding how fast the world is responding to the unprecedented pace of global warming.

    It now seems likely that the 2012 model output will also be incorrect and that we will experience an ice-free Arctic before 2030.

  15. @John Franklin, nice try at moving the goalposts. But a lie is a lie, including yours.


  16. John Franklin - There are many opinions and conclusions about various elements of the climate change debate that are assumed to be correct simply because they enlist a kernel of truth to anchor an argument. One, as you point out, is the truth that computer models are inherently imperfect. The argument is that because they consistently and predictably prove to be imperfect, they are thus unreliable and of zero value.

    This, despite the fact that every single modern technological benefit that modern man enjoys, from weather forecasting to sewer systems to the efficiency of anyone's car is entirely a product of inherently imperfect computer modelling, making inherently imperfect predictions of the future that none the less seem to provide measurable improvements over whatever else was there before.

    Cliff Mass took a stab at examining the popular comprehension of the term "climate change denialism" to a revealing success. Maybe he should devote a blog article to the popular comprehension of the role and reliability of computer modelling in everything we do including predicting future climate?

  17. Placeholder - the world is warming and that warming is having and will have major impacts on the world's ecosystems and humans. The realities of that warming are clear now but you choose to label those who are documenting the warming and its effects as "cultists" and a "lobby".

    I wish your statement that "average temps in this country have actually declined for a long time" were true but it is not.

  18. @John Franklin, the world has been warming since the end of the Little Ice Age. Your cult says it's humans' fault, and that the answer is for Americans to pay higher taxes, which happens to be what you people have always wanted under one guise or another.

    This has never been about science. It's about your politics, and nothing else.


Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

Rain without Clouds, the Upcoming Cooling, and Strong Leeside Winds: All in My New Podcast

The radar image this morning at 5:30 AM showed rain...some heavy... offshore. As shown in the satellite image at the same time, much of that...