70°F Road Trip version 2.0 (2023)

In all my (many) years of making maps, the most popular by far is the 70°F Road Trip map that I originally produced in 2015. It's been shared over 10 million times on various platforms if I had to guess. Maybe 20 million. That map was a hypothetical trip through the U.S. with the route tracking where the normal daily high temperature was approximately 70°F (21°C). I have been contacted numerous times over the years since from people who want to bike or drive the trip. It has resonated with people in a way I never imagined.

The map was based on the 1981-2010 climate normals produced by the National Center for Environmental Information (NCEI). Every 10 years, new climate normals are produced. In May 2021, the 1991-2020 normals were released by NCEI. Since then, I have been meaning to update the 70°F road trip map with the new normals. Several people have asked about it over the last two years. It might surprise people that as fast as the globe is warming up due to human activity, the shift in 70°F temperatures between 1981-2010 and 1991-2020 is very slight. It's just too short of a time period to capture the distance change. Instead, I decided to come up with new routes. The original route was interesting, but it left out a lot of options. This time around, there are multiple routes to choose from. Enjoy!

Continuous U.S.:

There are two routes this time (and a combined route), 1) Coastal Route, and 2) Interior Route. 

1) Coastal Route - This route begins in Florida, goes up counter-clockwise to the Pacific Northwest before winding its way through the Intermountain West before reaching San Diego. (7,468 miles)

• January - Begin in Tampa, Florida, and slowly head east toward Orlando. (151 miles)
• February - Head up the east coast of Florida to Jacksonville. (81 miles)
• March - Continue northward along I-95 to just past the N. Carolina border. (341 miles)
• April - Keep on I-95 northbound to Philadelphia area. (447 miles)
• May - The last segment of I-95 takes you to the Boston area. (300 miles)
• June - Things change really fast in the summer. It's nearly impossible to beat the 70°F isotherm. That said, you work your way from Bost over to Buffalo, then Cleveland, then Detroit, through the Upper Peninsula of Michigan, and all the way to central Montana. Most of this section needs to be done in early June. (2,388 miles)
• July - Most of the July segment follows I-90 through western Montana, Idaho, and then to Seattle. The end of month point is along the coast. (965 miles)
• August - Head south along the coastal highway to Newport, Oregon. (289 miles)
• September - This month involves a lot of lesser traveled mountain highways though eastern Oregon. Then you pick up I-84 in Idaho and eventually I-15 in Utah. (962 miles)
• October - Follow a series of rural highways southeast to Albuquerque. (546 miles)
• November - Go west along I-40 to Flagstaff and then south along I-17 to Phoenix. (467 miles)
• December - Take I-8 from Phoenix to Los Angeles and then I-5 to San Diego. (531 miles)

2) Interior Route - This route begins in south Texas and works its way northward to near the Canadian border. It then heads east before working through the northern Rockies, before heading south through the western Great Plains and then heading through the Desert Southwest. (7,064 miles)

• January - Start in Brownsville, Texas, and head north to Corpus Christi, Texas. (140 miles)
• February - Follow the Texas coastal bend to Houston. (197 miles)
• March - Take US 59 north to Dallas and then I-35 to Oklahoma City. (408 miles)
• April - Head north and then northeast to Kansas City. (343 miles)
• May - Continue north to Des Moines, east to Rockford, then northwest to Minneapolis. (765 miles).
• June - Go north to Duluth and then head east toward central Montana. Most of this section needs to be done in early June. (1,062 miles - 923 are shared with the Coastal Route)
• July - Head southeast, and then at Butte, Montana, go to Yellowstone National Park and continue through western Montana, northwestern Colorado, and back into south central Wyoming. We have to take this circuitous route to stay in the high country. (1,255 miles - 202 are shared with the Coastal Route)
• August - Slowly move from south to north through central Wyoming and back into extreme southern Montana. (384 miles).
• September - Go due east until reaching the southwestern corner of North Dakota and turn south - until the Nebraska border. (533 miles)
• October - Continue nearly due south to the Texas panhandle and turn southwest to head toward Cloudcroft, New Mexico. (1,047 miles)
• November - Head nearly due east on state and U.S. highways until Phoenix. (399 miles)
• December - This last segment from Phoenix to San Diego is 100% shared with the Coastal Route. (531 miles).

2B) Combined Interior Route that connects with Coastal Route - In mid/late March, head northeast to connect with the Coastal Route and follow that route the rest of the way. (8,466 miles)

• January - Start in Brownsville, Texas, and head north to Corpus Christi, Texas. (140 miles)
• February - Follow the Texas coastal bend to Houston. (197 miles)
• March - Take US 59 north to Dallas and then turn east on I-30 to Little Rock. (553 miles)
• April - Head east-northeast to Memphis, Nashville, and eventually Washington D.C. (1,128 miles - 91 are shared with the Coastal Route)
• May - The last segment of I-95 takes you to the Boston area. (300 miles)
• June - Things change really fast in the summer. It's nearly impossible to beat the 70°F isotherm. That said, you work your way from Bost over to Buffalo, then Cleveland, then Detroit, through the Upper Peninsula of Michigan, and all the way to central Montana. Most of this section needs to be done in early June. (2,388 miles)
• July - Most of the July segment follows I-90 through western Montana, Idaho, and then to Seattle. The end of month point is along the coast. (965 miles)
• August - Head south along the coastal highway to Newport, Oregon. (289 miles)
• September - This month involves a lot of lesser traveled mountain highways though eastern Oregon. Then you pick up I-84 in Idaho and eventually I-15 in Utah. (962 miles)
• October - Follow a series of rural highways southeast to Albuquerque. (546 miles)
• November - Go west along I-40 to Flagstaff and then south along I-17 to Phoenix. (467 miles)
• December - Take I-8 from Phoenix to Los Angeles and then I-5 to San Diego. (531 miles)

All U.S. and Canada:

The U.S. - Canada version is quite a bit longer (of course). Since Canada still (!!!) has not published 1991-2020 normals, I took the liberty of computing them myself using the "old method" of fitting a 3rd-order polynomial spline to the monthly averages. It's not perfect, but it is pretty close.

The big difference with the U.S. - Canada map is that the route stays on non-interstate highways as much as possible. Also, they cross each other in Arkansas. This may be a little confusing. Otherwise, the map should be straightforward.

• January - Start in San Diego and stay on I-8 to Phoenix. (357 miles - 575 km)
• February - Follow I-10 to El Paso (384 miles - 618 km)
• March - From El Paso, we stay on local state roads to about Wichita Falls, TX, and then through extreme southern and extreme eastern Oklahoma (991 miles - 1,596 km)
• April - Head east through Arkansas, all of southern Tennessee, into N. Carolina and up the Blue Ridge Parkway, then through western Virginia, W. Virginia, and finally to Pittsburgh. (1,328 miles - 2,138 km)
• May - Start eastward for about 100 miles before looping back westward toward Youngstown and Canton, OH, turn northeast to Chicago, IL, then north into Wisconsin, and finally west to central Minnesota. (1,440 miles - 2,318 km).
• June - Continuing northward, you reach the international border at International Falls, MN. Follow the highway north to Longbow Lake and then west to Winnipeg and Edmonton. Keep heading northeast and then north on smaller highways to Hay River on the shore of Great Slave Lake. From there, head generally west toward Dawson, YT, and the border crossing at the Top of the World Highway. Then make your way to Fairbanks, AK.  (3,424 miles - 5,513 km)
• July - Head south from Fairbanks to just north of Anchorage and continue eastward to the Canadian border. Take the western route to the Lower 48. By the end of the month find your way to Prince George, BC. (1,735 miles - 2,793 km)
• August - From Prince George, head southeast to Calgary and cross back into the U.S. near Glacier National Park. (643 miles - 1,035 km).
• September - Stay on smaller highways that head south through parts of western Wyoming and eastern Idaho and then northeastern Utah. Briefly take I-70 east to the Colorado Border. (1,056 miles - 1,700 km)
• October - In Colorado, head southeast on state and U.S. highways into extreme northern New Mexico near Taos. Then head north and northeast to Denver and then east into the plains of eastern Colorado. Then continue eastward through most of southern Kansas before turning southeast to Little Rock, AR. This is where the route crosses itself from early April.  (1,632 miles - 2,628 km)
• November - Head nearly due east to Jackson, MS, then up to Birmingham, AL, and then southeast to Lake City, FL. (732 miles - 1,179 km)
• December - This last segment is from Lake City, FL, to Daytona, FL. (187 miles - 301 km).

Trends in Daily Record Temperatures

The Earth is warming. Period. Humans are responsible for most of the warming in the last 200 years. Period. Let's get that out of the way. These are uncontroversial statements in the climate science community. The people who doubt these facts are appropriately know as climate deniers. 

What I am interested in knowing is how daily temperature record have changed over time. It's common to hear the phrase, "a new record high temperature was set today." Less frequently you will hear, "a new record low temperature was set today." What I am interested in knowing is how often this occurs and are they occurring more frequently or less frequently with time.

The first hurdle to get past is something called the period-of-record (PoR). This refers to how long records have been kept at a location. The longer the PoR, the harder it is to set records. Imagine a station with only 4 years of daily temperature data. In year 5, there is a 20% chance of setting a record on any given day based strictly on probabilities. If a station had 100 years of data, that chance drops to 1% in any given year. How do we reconcile this? The strategy I employed was to fix the period of record. In this case, I used the 1943-2022 period (80 years) and assume that Jan 1, 1943, is the first day that records were kept. In this manner, every station has exactly 80 years of records - no more, no less.

Methodology

For each day of the year, Jan 1, Jan 2, and so on, I identified the year that the record high/low temp was set. That year gets a tally of 1 for each daily record that occurred during that year. For example, if the record high for Mar 1 was set in 1980, the record high for Apr 1 was set in 1980, and the record high for May 1 was set in 1980, and no other days in 1980 are identified as having any record highs, the year 1980 gets a tally of 3. If in 1981, the record set on May 1 in 1980 was tied, the May 1 record is assigned 50% to 1980 and 50% to 1981. This means the tally of 3 for 1980 is now reduced to 2.5. In this way, every day of the year counts toward a record value of 1. If the record is only in one year, that year gets the entire value. If the record is shared by two , three , or more years, each of those years gets a fractional share of that record. For an 80-year period, each year should have 4.56 records on average. Of course some will have more and some will have less. This study looks at the trend in records.

The Global Historical Climatology Database - Daily (GHCN-D) is used to assess daily temperature values. For inclusion, a station must be 95% complete for the 1943-2022 period and each year must have at least 250 valid temperature observations. At this point, readers are probably asking why 1943 was chosen as the start year. There are two primary reasons. First, the longer time period you look at, the fewer stations there are with sufficient data. Going back 90, 100, or more years greatly reduces the number of station that would qualify. Second, I very intentionally wanted to avoid the Dust Bowl years since there is a large contribution from human activities to the temperatures in large parts of the Lower 48.

WBAN and Cooperative Stations

The analysis includes both WBAN (Major or 1st Order) stations and Cooperative (Coop) stations. WBAN stations are typically found in big cities and always restrict the time period of high/low temps to midnight-to-midnight. Coop stations record the daily min/max temps in the morning (frequently 7 am) or sometimes in the evening (frequently 7 pm). An individual station is exclusively a morning station or an evening station. Major cities have a well-know propensity to be warmer than their surrounding Coop counterparts. This effect, the urban heat island effect, is real and because of it, we separate the analysis into two sets - WBAN and Coop.

Trends

The time series in Fig. 1 shows the number of daily high maximum temperature records assigned to each year for Nome, Alaska. A linear regression trend line is shown on the chart as a dotted line. As you can clearly see, record high temperatures are much more frequent in recent years and much less frequent in earlier years. The trend in daily records for each station is assessed using a Mann-Kendall (MK) test for significance. This test is well suited to data sets with large internal variability. The Mann-Kendall test for this time series indicates that it is increasing (duh!) at greater than 99th percent confidence interval. That means there's less than a 1% chance the distribution has the shape it does purely by chance.

Fig 1. Time series of count of daily record high temps assigned to each year between 1943 and 2022 for Nome, Alaska. Ties prorated. Trend show with dotted line.

Types of Records

Four types of temperature records are evaluated in this analysis: high maximum, high minimum, low maximum, and low minimum. The first two are warm records and the latter two are cold records. The maps for the records show the statistical significance. The categories are:

• Increasing/Decreasing at a rate that is statistically significant at the 90th confidence interval.
• Increasing/Decreasing at a rate that is not statistically significant (No Trend).

The Maps

The maps are organized by a) all stations, b) major stations, and c) Coop stations. Within each of those groupings, there are maps for high 1) high maximum, 2) high minimum, 3) low maximum, and 4) low minimum. 

Maps for All Stations (WBAN & Coop)

High Maximum - All Stations

Trend in frequency of high maximum temperatures records 1943-2022, for 537 stations.

High Minimum - All Stations

Trend in frequency of high minimum temperatures records 1943-2022, for 537 stations.

Low Maximum - All Stations

Trend in frequency of low maximum temperatures records 1943-2022, for 537 stations.

Low Minimum- All Stations

Trend in frequency of low minimum temperatures records 1943-2022, for 537 stations.

Maps for Major (WBAN) Stations

High Maximum - Major Stations

Trend in frequency of high maximum temperatures records 1943-2022, for 121 major (WBAN) stations.

High Minimum - Major Stations

Trend in frequency of high minimum temperatures records 1943-2022, for 121 major (WBAN) stations.

Low Maximum - Major Stations

Trend in frequency of low maximum temperatures records 1943-2022, for 121 major (WBAN) stations.

Low Minimum - Major Stations

Trend in frequency of low minimum temperatures records 1943-2022, for 121 major (WBAN) stations.

Maps for Cooperative (Coop) Stations

High Maximum - Coop Stations

Trend in frequency of high maximum temperatures records 1943-2022, for 416 Cooperative (Coop) stations.

High Minimum - Coop Stations

Trend in frequency of high minimum temperatures records 1943-2022, for 416 Cooperative (Coop) stations.

Low Maximum - Coop Stations

Trend in frequency of low maximum temperatures records 1943-2022, for 416 Cooperative (Coop) stations.

Low Minimum - Coop Stations

Trend in frequency of low minimum temperatures records 1943-2022, for 416 Cooperative (Coop) stations.

Supplemental Plots

Fig. S1. This scatter plot shows the relationship between the 80-year trend in max temps versus the 80-year trend in min temps for 392 Coop stations. It's a modest, but mostly unremarkable relationship. Some instances of max temps trending cooler with min temps trending warmer. Really all combinations are present, but some are more favored.

Coldest Days Trends

Changes in Cold Days

This is a companion post to the Hottest Days Trends analysis. 

Is the frequency of the coldest days of the year changing? The tl;dr answer is of course yes. There are any number of ways to measure this. In this blog post, I identify the frequency of cold days during a base period (1951-2000), and then evaluate the trends of those values. The methodology and results are presented below.

Methodology:

1) Find the temperature that occurs on average 20, 15, 10, and 5 days per year during a baseline period for all long-term climate stations in the U.S. (I chose 1951 to 2000 - 50 years). For a 50-year period, the 10-day per year temperature value is found by ordering all 18,263 low temperatures (assuming no missing obs) and identifying the 500th lowest value.

2) For each year between 1950 and 2021, count the number of days that met or exceeded the temperatures identified in Step 1. Also identify the first and last occurrence of these temperatures each year. For example, if 20F is the temperature that occurred an average of 10 days per year during the 1951-2000 baseline period, count the number of days each year that 20F occurred and find the first/last 20F day each year.

3) If a station contained 65 or more complete years of temperature data (no more than 10 missing days in a year), it was kept for the final analysis. There were 609 U.S. stations that successfully met the criteria, 15 Canadian stations, and 0 Mexican stations. 

4) Fit a linear regression line to the 72-year (1950-2021) time series of the number of annual occurrences (and the first/last dates) for each station. The different between the fitted (not observed) 2021 value and the fitted (not observed) 1950 value represents the change,

Example:

In Fig. 1, we see the change in days per year for the temperature that historically occurred 20 and 10 days per year for Boise, Idaho. Those temperatures are 17F and 1F respectively. A regression line is fitted to each time series and the begin and end points of the regression lines represent the change. In the Boise example, the 17F temperature, which occurred an average of 20 days per year during the 1951-2000 period, trended from 26 days per year to 12 days per year. An decrease of 24 days per year over the 72-year period.

Fig 1. Change in the number of days per year for the temperature that historically (1951-2000) occurred 10 and 20 days per year in Boise, Idaho.

Fig. 2. Change in the length of time (days) between the first/last occurrence of the temperature that historically (1951-2000) occurred 10 and 20 days per year in Boise, Idaho.

In Fig. 2 we see the same information for Boise, except it notes the length of time between the first and last occurrences of the temperature that historically occurred 20 and 10 days per year. A low temperature of 17F historically occurred 20 days per year in Boise. The trend in the length of time between the first and last 17F day of the year decreased from 85 to 57 days per year. We interpret this as a shortening of the coldest time of year by 28 days. Specifically, the typical first occurrence for Boise of 17F changed from November 30th to November 29th (1 day later) and the typical last occurrence of 17F changed from February 21st to January 25th (27 days earlier).

Maps: Modern Days per Year Count

Figs 3-6 below show the current number of days per year for the temperatures that historically occurred 20, 15, 10, and 5 days per year. If you refer back to Fig. 1, the value at the end of the trend (dotted) lines are what we are mapping below. 

Fig 3. Current number of days per year that the historically (1951-2020) coldest 20 days per year now occur. This is the end point of the 72-year regression line.

Fig 4. Current number of days per year that the historically (1951-2020) coldest 15 days per year now occur. This is the end point of the 72-year regression line.

Fig 5. Current number of days per year that the historically (1951-2020) coldest 10 days per year now occur. This is the end point of the 72-year regression line.

Fig 6. Current number of days per year that the historically (1951-2020) coldest 5 days per year now occur. This is the end point of the 72-year regression line.

Maps: Change in the Length of the Cold Portion of the Year

As Fig. 2 shows for Boise, we can track the length of time between the first/last occurrence of particularly hot days and see how that length changed over time. Figs 7-11 below show how the length of time of peak heating has changed. 

Fig 7. Change in the length of time that the historically (1951-2020) coldest 20 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Fig 8. Change in the length of time that the historically (1951-2020) coldest 15 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Fig 9. Change in the length of time that the historically (1951-2020) coldest 10 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Fig 10. Change in the length of time that the historically (1951-2020) coldest 5 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Maps: Change in the Length of Consecutive Cold Days

This is where we will look at the changing length of coldwaves. As with before, we're using the temperature that historically occurred 20, 15, 10, and 5 days per year. Now we are looking at each year and identifying the maximum number of consecutive days that those temperatures occurred. If the 5-day per year temperature occurred 8 times in a specific year but there was at least 1 day between each occurrence, then the maximum consecutive streak would be 1 in that example. Figs. 11-14 show the change in the length of these streaks. The change is the end trend point minus the beginning trend point divided by the beginning trend point times 100. 

Fig 11. Change in the number of consecutive days at or below the temperature that historically (1951-2020) occurred 20 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Fig 12. Change in the number of consecutive days at or below the temperature that historically (1951-2020) occurred 15 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Fig 13. Change in the number of consecutive days at or below the temperature that historically (1951-2020) occurred 10 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Fig 14. Change in the number of consecutive days at or below the temperature that historically (1951-2020) occurred 5 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Mapping

The map in Fig. 15 shows the stations used in this analysis. The maps in Figs 3-10 were generated from station data using an inverse distance weighted surfacing algorithm with 50km grid cells. There is often a lot of variability in small geographical regions; therefore, a 5x5 smoothing filter was applied to all maps. No adjustment for topography was made. 

Fig 15. Map of stations used in the analysis presented in this blog post. There are 609 U.S. stations that successfully met the criteria, 15 Canadian stations, and 0 Mexican stations.

Conclusion

The frequency and duration of the coldest days of the year are dramatically decreasing in the U.S. Period. Unlike the heatwave analysis, there are no locations that have escaped this trend.

Hottest Days Trends

** Update: there is a companion Coldest Days Trends blog post. **

There's a lot of discussion about the increase in heatwaves and hot days in general in a warming world. Unfortunately there is no standard definition of a "heatwave". Many places in the northeastern U.S. describe heatwaves as multiple days in a row with a high temperature above 90F. In the southern Great Plains, days above 100F are more relevant. From a public health perspective, nighttime low temperature may be most meaningful. 

In 2021, the EPA published a widely-shared blog post and map set of the trends in heatwaves from 1961-2019 for the 50 largest cities in the U.S. They defined heatwaves as: 

[A] period of two or more consecutive days when the daily minimum apparent temperature (the actual temperature adjusted for humidity) in a particular city exceeds the 85th percentile of historical July and August temperatures (1981–2010) for that city.

The 85th percentile equates to the 9 highest values per summer (July-August). I have no real objection to their criteria. If I were to critique it though, I would say that most people informally characterize heatwaves as a measure of the peak heat, not the overnight minimum (apparent) temperature. Again, from a public health point of view, nighttime recovery is crucially important. Their metric is more heat stress than heatwave. In my opinion, heatwaves are about high temperatures. Not low temperatures. Not apparent temperatures. In most cases, the very hottest temperatures occur with locally dry airmasses. But again, there's plenty of room to slice and dice this any number of ways.

Changes in Hot Days

In May 2021, I wrote a blog post on the change in days with heavy precipitation. The USA Today newspaper did a full-length write-up on the findings. For this analysis of change in hot days, I used a similar methodology.

Methodology:

1) Find the temperature that occurs on average 20, 15, 10, and 5 days per year during a baseline period for all long-term climate stations in the U.S. (I chose 1951 to 2000 - 50 years). For a 50-year period, the 10-day per year temperature value is found by ordering all 18,263 high temperatures (assuming no missing obs) and identifying the 500th highest value.

2) For each year between 1950 and 2021, count the number of days that met or exceeded the temperatures identified in Step 1. Also identify the first and last occurrence of these temperatures each year. For example, if 90F is the temperature that occurred an average of 10 days per year during the 1951-2000 baseline period, count the number of days each year that 90F occurred and find the first/last 90F day each year.

3) If a station contained 65 or more complete years of temperature data (no more than 10 missing days in a year), it was kept for the final analysis. There were 706 U.S. stations that successfully met the criteria, 18 Canadian stations, and 0 Mexican stations. 

4) Fit a linear regression line to the 72-year (1950-2021) time series of the number of annual occurrences (and the first/last dates) for each station. The different between the fitted (not observed) 2021 value and the fitted (not observed) 1950 value represents the change,

Example:

In Fig. 1, we see the change in days per year for the temperature that historically occurred 20 and 10 days per year for Boise, Idaho. Those temperatures are 96F and 98F respectively. A regression line is fitted to each time series and the begin and end points of the regression lines represent the change. In the Boise example, the 96F temperature, which occurred an average of 20 days per year during the 1951-2000 period, trended from 14 days per year to 34 days per year. An increase of 20 days per year over the 72-year period.

Fig 1. Change in the number of days per year for the temperature that historically (1951-2000) occurred 10 and 20 days per year in Boise, Idaho.

In Fig. 2 we see the same information for Boise, except it notes the length of time between the first and last occurrences of the temperature that historically occurred 20 and 10 days per year. A high temperature of 96F historically occurred 20 days per year in Boise. The trend in the length of time between the first and last 96F day of the year increased from 65 to 89 days per year. We interpret this as a lengthening of the hottest time of year by 24 days. Specifically, the typical first occurrence for Boise of 96F changed from June 24th to June 13th (11 days earlier) and the typical last occurrence of 96F changed from August 26th to September 8th (13 days later).

Fig 2. Change in the length of time (days) between the first/last occurrence of the temperature that historically (1951-2000) occurred 10 and 20 days per year in Boise, Idaho.

Maps: Modern Days per Year Count

Figs 3-6 below show the current number of days per year for the temperatures that historically occurred 20, 15, 10, and 5 days per year. If you refer back to Fig. 1, the value at the end of the trend (dotted) lines are what we are mapping below. 

Analysis

Most readers are familiar with the great heatwaves of the 1930s during the Dust Bowl era. Poor agricultural practices led to a drying of the soils, which meant much more efficient solar heating (higher temperatures). In the 1950s, there were pronounced droughts in the central portion of the U.S. too. These droughts meant more sunshine and greater solar heating efficiency. I note this because the time period we begin with is 1950. This results is a muting of the trends for many regions. If we started in 1960, the change in hot days per year would be more dramatic. That said, starting in 1950 allowed for a maximum number of stations to be included in the analysis; and we should never arbitrarily pick a start date for the analysis to maximize (or minimize) the trends.

The obvious question that arises from looking at the maps are the blue areas in the Great Plains and Mississippi River valley. My assessment is that these regions had large values in the 1950s (flattens the curve) and also have seen a dramatic increase in precipitation during this same time period - which causes less solar energy to be used for warming the surface (more solar energy is used to evaporate water). Conversely, there has been a dramatic drying of the western U.S. This allows for more efficient solar heating.

Let's not lose sight of the fact that the net values are dramatically higher - even with the blue areas. The orange and blue areas do not cancel each other out. Looking at Fig. 6, we now see 7.8 days per year nationwide of the temperature that historically occurred 5 days per year. That is a greater than 50% increase!

Fig 3. Current number of days per year that the historically (1951-2020) hottest 20 days per year now occur. This is the end point of the 72-year regression line.

Fig 4. Current number of days per year that the historically (1951-2020) hottest 15 days per year now occur. This is the end point of the 72-year regression line.

Fig 5. Current number of days per year that the historically (1951-2020) hottest 10 days per year now occur. This is the end point of the 72-year regression line.

Fig 6. Current number of days per year that the historically (1951-2020) hottest 5 days per year now occur. This is the end point of the 72-year regression line.

Maps: Change in the Length of the Hot Portion of the Year

Ultimately I was interested in how the length of summer has changed with respect to peak heating. As Fig. 2 shows for Boise, we can track the length of time between the first/last occurrence of particularly hot days and see how that length changed over time. Figs 7-11 below show how the length of time of peak heating has changed. 

Analysis

In general, the patterns are consistent with the maps showing the change in the number of days (Figs 3-6). I would say that the primary difference is a sharper peak in max summer heating. Using the example of the historically hottest 10 days per year, Fig. 7 shows a nationwide average increase of 44% (now 14.4 days per year vs 10 days per year). The length of time that the historically warmest 10 days per year occurs increased from 59.6 days to 67.7 days - a length increase of 13.6%.

Fig 7. Change in the length of time that the historically (1951-2020) hottest 20 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Fig 8. Change in the length of time that the historically (1951-2020) hottest 15 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Fig 9. Change in the length of time that the historically (1951-2020) hottest 10 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Fig 10. Change in the length of time that the historically (1951-2020) hottest 5 days per year now occur. This is the difference between end point and beginning point of the 72-year regression line.

Maps: Change in the Length of Consecutive Hot Days

This is where we will look at the changing length of heatwaves. As with before, we're using the temperature that historically occurred 20, 15, 10, and 5 days per year. Now we are looking at each year and identifying the maximum number of consecutive days that those temperatures occurred. If the 5-day per year temperature occurred 8 times in a specific year but there was at least 1 day between each occurrence, then the maximum consecutive streak would be 1 in that example. Figs. 11-14 show the change in the length of these streaks. The change is the end trend point minus the beginning trend point divided by the beginning trend point times 100. As with the other maps, there is a decline in the Mississippi River valley, but a marked increase everywhere else. A quick summary is that heat waves are longer in most places, except for the Mississippi River valley.

Fig 11. Change in the number of consecutive days at or above the temperature that historically (1951-2020) occurred 20 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Fig 12. Change in the number of consecutive days at or above the temperature that historically (1951-2020) occurred 15 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Fig 13. Change in the number of consecutive days at or above the temperature that historically (1951-2020) occurred 10 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Fig 14. Change in the number of consecutive days at or above the temperature that historically (1951-2020) occurred 5 days per year. This is the difference between end point and beginning point of the 72-year regression line as a percentage.

Mapping

The map in Fig. 15 shows the stations used in this analysis. The maps in Figs 3-10 were generated from station data using an inverse distance weighted surfacing algorithm with 50km grid cells. There is often a lot of variability in small geographical regions; therefore, a 5x5 smoothing filter was applied to all maps. No adjustment for topography was made. 

Fig 15. Map of stations used in the analysis presented in this blog post. There are 706 U.S. stations that successfully met the criteria, 18 Canadian stations, and 0 Mexican stations.

IPCC Comparison

The maps presented above are broadly consistent with the IPCC analysis of trends in hot extremes (see Fig. 16). They also looked at trends since the 1950s and found parts of central North America were the only region in the world not experiencing an increase in extreme hot temperatures.

Fig 16. IPCC 6 figure (SPM.3) showing the trend in extreme

Conclusion

The frequency and duration of the hottest days of the year are dramatically increasing in the U.S. Period. The rate and magnitude of increase differs across space. Some areas have even seen a decrease. Why does this spatial variation exist? My guess is that the 1950s were an especially warm and dry decade in the central U.S. and a dramatic increase in precipitation in this region in recent decades make achieving hot temperatures especially difficult. Conversely, the western CONUS has dried substantially in recent decades. This has worked to magnify the increase in hot days. Remember that the orange and blue areas on the maps do not cancel each other out.

The blue colors in Figs 3-14 (the analysis maps) unfortunately draw attention away from the strong warming signal in the rest of the maps. It provides "deniers" with a false narrative to latch on to. Such is life. Scientists should always publish data and analysis and accept that bad faith actors will willfully misinterpret. It is incumbent upon us to explain when the bad faith arguments are incorrect and why the people pushing those arguments are bad actors. 

Chugach 5,000' Peaks

The Chugach Mountains outside of Anchorage, Alaska, are a vast, unpopulated wilderness full of wildlife, glaciers, and more mountains than you can hike/climb in a lifetime. The portion of Chugach State Park closest to Anchorage boasts a handful of peaks above 5,000'. Specifically, the region south and west of Ship Creek and west of Indian Creek has a dozen (12) peaks that are identified on USGS topographic maps as having a summit more than 5,000' above sea level. Many people have identified these peaks as goals to summit in their hiking excursions. A now famous endurance challenge involves summiting all 12 peaks in a single outing (see: https://www.stockalpine.com/posts/chugach-front-linkup.html ). The routes (and there are several routes) have an elevation gain of ~20,000' and a distance of ~40 miles.

But are there really 12 peaks above 5,000'? Again, these are what the USGS topographic maps indicate. The maps were generated from stereo aerial photographs taken in the late 1950s and early 1960s, and were published in the early to mid 1960s.

The image above is a screenshot of the Anchorage A-7 USGS 1:63,360 topographic map (1960) that shows several of the peaks above 5,000'. It is important to note that the elevations were determined from photogrammetric estimates - not cadastral surveys. The operative work is "estimate". Few, if any, of the highest peaks in this region were actually surveyed. I have been to all of these summits (identified as 5,000'+) and do not recall seeing any survey monuments. The Bureau of Land Management's land survey website indicates no formal survey activities in the area except for section corner and witness points, and the Powerline utility easement.

How accurate are USGS elevation estimates for mountain peaks? My assessment is that they are pretty good, but not perfect. Recall that a few years ago the quite famous Mt. Marathon elevation of 3,022' was determined to be 2,974' when a surveyor decided to test the USGS topographic map elevation estimate. This is a discrepancy of 48'. Let me be clear, USGS topographic maps are the most amazing feat of mapping in the history of humanity. Every American should be in awe of this map set. Full stop. But they are not intended to be a perfect, infallible representation of the landscape. 

Can we do better than the USGS estimates? Yes. We could survey the summits. This is a costly and time consuming process. I imagine that like Mt. Marathon, perhaps someone might do it out of personal interest. What's the next best thing? Aerial or satellite systems that send pulses of energy to the ground and measure the return time. The most common systems are LIDAR (LIght Detection and Ranging) and InSAR (Synthetic Aperture Radar Interferometry). 

In 2010-2017, USGS flew the entire state of Alaska to generate a highly detailed elevation data (see: https://eros.usgs.gov/doi-remote-sensing-activities/2017/usgs/alaska-insar-elevation-data-status ). The gridded data set has 5m horizontal resolution, and from what I can tell has an estimated vertical accuracy of 1.5m. Using this data set, we can interpolate the elevation for all peaks in our area of interest. No longer are we limited to 1950s photogrammetric technology. 

So what does the InSAR data show? It shows that the 12 peaks are all indeed over 5,000' and that Mt. Williwaw is the highest and S. Suicide Peak is the lowest. But it also shows something interesting. It shows that Ptarmigan Peak is over 5,000'. Instead of 12 peaks over 5,000', the InSAR elevation data shows 13 peaks over 5,000'. 

The USGS map shows Ptarmigan Peak is 4,910'. However, when standing on the summit, it clearly is very close to the same elevation of S. Suicide Peak. Several people have reported that the GPS on their phone shows the summit to be well over 5,000', myself included. GPS is certainly more accurate in the horizontal direction than the vertical direction - but the > 5,000' GPS reading is a strong indicator that the topographic map value is far too low in elevation. 

Is the InSAR data to be trusted? An independent measure to cross reference would be nice. It turns out that we have a second data set for comparison. In 2015, the Municipality of Anchorage commissioned a LIDAR elevation mapping survey (also airborne). Unfortunately their area of study did not cover the 5,000' peaks in question; however, there are areas of overlap for many of the high peaks near Eagle River and Girdwood. A sample of peaks in the 4,000' to 5,500' range shows the 2015 LIDAR elevation data is within 3m of the 2010 InSAR elevation data. This is a very strong, independent verification of the InSAR accuracy. Even if the InSAR elevation for Ptarmigan Peak is 5m to 10m too high, it still is safely over 5,000'. 

The National Geodetic Survey has thousands of survey markers around the country that are used as control points for land surveys. In the part of the Chugach Mountains that we are discussing here, there are three monuments above 3,000' that were surveyed in the mid-1980s. One is at Near Point, one is at Rusty Point, and a third is on the McHugh Peak ridge line. Looking at the National Geodetic Survey data for these three sites and comparing it to the InSAR data, the National Geodetic Survey points are 3.8m higher in elevation than the InSAR. This may be a result of different vertical datums used by the National Geodetic Survey (NVD29) and the InSAR data (NVD88). In any event, the InSAR data is actually a little lower than the National Geodetic Survey data, confirming the 5,000' elevation of the peaks in question.

Finally, in 2021, the USGS published a new 1:24,000 topographic map series using the InSAR data (see map below). Those maps clearly show Ptarmigan Peak is over 5,000' (contours are 80'). Unfortunately they do not have Peak elevations noted like the previous maps did. 

Since the underlying data are provided, we can extract the highest grid cell value (5,052'). We expect the true peak elevations to be up to a few feet higher since the horizontal resolution of the underlying grid cells are 5x5 meters and the grid cell average value will not capture the highest elevation. This is why the nice rounded western summit of Ptarmigan Peak shows as 1 foot higher than the narrow, sharp, pointed eastern summit (true summit).

There is now overwhelming evidence that shows  Ptarmigan Peak is over 5,000' in elevation and therefore there are 13 peaks over 5,000' in this portion of the Chugach Mountains.

Postscript: On August 3, 2022, I hiked up the standard route of Ptarmigan with my drone in hand. I know, you are not supposed to have drones in the Park, but this was for science. Drones are very good at keeping a constant elevation. From the main summit, I calibrated the drone to the highest location and then flew it 360 meters to the summit immediately to the east. What I found is that the eastern summit is a 7.3 meters (24 feet) higher and is therefore the true summit of Ptarmigan Peak. In addition, my inReach and cell phone GPS both had elevations comfortably over 5,000'. For consistency, I will not edit the table or the map of the InSAR elevations since those reflect their true data source.

Google Earth File

Here is a Google Earth file that I made in GIS using the InSAR DEM and identifying peaks. It only covers the area south/west of Ship Creek and Indian Creek. I only left the peaks that have a page on the peakbagger.com site that are over 3,000' in elevation in InSAR. 

URL: http://www.61n150w.com/NamedPeaksPointsElevation.kmz