Happy Father's Day - Chris's Corner 19 June 2022

Well 4 pm was 101 at lambert. So official I believe the 5th 100 degree day of the year

What's happening with the NWS products. Nothing is updating. Is there and outage?

I was wondering if the metropolitan area reporting of temperatues was being exploited to over exaggerate the world temperature myself. I know climate changes, but how much of those pretty linear charts going up over the years are due to urban sprawl. I mean we know at lambert, its a lot warmer at night than warrenton.

This is an incredibly complicated topic. I was going to post a lengthy explanation of what is going on, but I deleted it all because it was way too nerdy. The short answer is no, the UHI effect does not translate into a positive bias on the global average temperature trend. If anything it may actually be causing a negative bias at least after WWII. That's right...negative! Wickham et al. 2013 assess the bias as -0.10 C/century from 1950 to 2010. Others have arrived at a similar result. If you're interested in the reasoning why let me know and I'll try to retype a more terse explanation. BTW...fun fact that is completely irrelevant...one of the authors on that paper (and other Berkeley Earth publications) is Saul Perlmutter. The massively geeky among you will recognize this name. That is the guy who figured out the universe's expansion is accelerating and won a Nobel Prize for his achievement. Though he may be better known for his role in a Big Bang Theory episode playing himself in a cameo appearance and to dis out insults to Sheldon. I guess processing supernova and other astronomical data got too boring for him.

Urban areas most definitely are warming faster than rural areas. But that's not a bias. That is a real effect.

The UHI Bias is different than the UHI Effect though they are definitely related topics. The UHI Effect is only positive. In other words, it always leads to higher temperatures in urban areas. But depending on a lot of different factors this bias on the global average temperature can and probably is (at least over some periods) be negative. To understand why you have to understand how the global average temperature is calculated. It is calculated by gridding the planet into cells. Observations are then aggregated to form a value for each cell. The global average temperature is then the area weighted average of all the cells. The key to understanding the bias is to understand that the cells are often large enough to represent both rural and urban areas. For example, imagine a cell that is 75% rural and 25% urban. If 75% of the observations for that cell are from urban stations AND those urban stations are experiencing an increase in the magnitude of the UHI effect then the bias on that cell will be positive. But, if the official observation site got moved from a city center to the outskirts of town (this was common after WWII) the station would exhibit a changepoint with overall lower temperature reporting precisely because the UHI effect is positive. The bias that this causes on that cell is now negative. Similarly if the ratio of rural-to-urban stations increases (common after WWII) it will cause a negative bias precisely because the UHI effect is positive. There are many other configurations that can lead to both a positive and negative bias. And there are a lot of details and caveats here I'm not going to even attempt to get into. The salient point here is that the UHI effect is different than the bias it causes during gridding and averaging operations. It is important not to conflate these concepts.

BTW...there is an effect opposite the UHI as well. Agriculture suppresses daily highs. Being in St. Louis and near arguably the most fertile croplands in the entire world both of these effects are competing simultaneous for us. One of the big advancements in weather modelling over the last couple of decades is land use changes; not just in urban areas, but rural areas as well. Right now I think the GFS (and others) use static cropland parameterizations. This is part of the reason why its temperature forecasts can be off. It doesn't know if the corn and soy crops around us are wilting or how aggressively the farmers are irrigating from day to day both of which have a dramatic effect on evapotranspiration, soil moisture, albedo, etc.

How many rural obs were in a particular cell in 1950

Sparseness of observations is a much bigger problem. Actually, the UHI bias is probably the least of the problems. Instrument/shelter change biases, time-of-observation bias, observation procedure biases, grid cell infilling bias, etc. are all much worse. The biggest problem of them all, like by far, is the switch from bucket measurements to ship/buoy measurements after WWII. This introduces an extreme positive bias on the warming trend since bucket measurements are biased low and ship/buoy measurements are biased high relative to each other. The bias here is at least an order of magnitude higher than that for the UHI bias.

The number of rural and urban observations varies both by cell location and by time. One year the same cell might be 50/50 while the next it could be 60/40 or 40/60 or whatever. That's where the problem comes in. And there are still record digitization efforts ongoing. If you calculate the temperature today and compare it to a calculation tomorrow you'll almost certainly get a slightly different result because the GHCN (and like) repository is continually accepting observations that are years, decades, and even 100 years old as those hand written records finally get digitized and uploaded.

Dewpoints have been mixing out to 61 or 62 the last couple of days and no reason to think it won't be the same or perhaps a degree or two lower. Clearly less water in the soil to suck up some of the insolation so stage is set for another superadiabatic surface temp response. Gained 23F degrees yesterday and today is already starting 6F warming. Thinking airport will make it above the 103F season high so far.

That’s what I keep focusing on as well! It won’t be a dramatic cool down (never is this time of year), but I will take 85-90 with storm chances and clouds ALL day long over this blast furnace.

105/60 for sure

Those mid 70 DPs are just suffocating

105/60 feels significantly better than 95/75 imo

105/60. At that point my sweat can provide me some evaporative cooling. With the other one I sit in a puddle of sweat.

Ok lambert 95 already. Im going with 10 degree rise from the noon hour. I feel like we will hit 108 today. The clouds that we had are gone and it is heating up fast. Just my thoughts letsnsee how close. 104 tp 105 for sure at lambert just like chris mentioned.

Those U70/L80 dewpoints were no joke. HIs are significantly lower this round. It's still scorching hot!