Attachment 5 - SummitProjectionsFinal.pdf

Stephen Saunders Tom Easley

Melissa Mezger Updated December 2021

CLIMATE PROJECTIONS IN

SUMMIT COUNTY, COLORADO

MOUNTAIN

ROCKY

CLIMATE

the

Organization

A report to Summit County, Town of Breckenridge, Town of Frisco, and NWCCOG Water Quality/Quantity Committee

By Stephen Saunders, Tom Easley, and Melissa Mezger

A report by the Rocky Mountain Climate Organization to Summit County, Town of Breckenridge, Town of Frisco, and Northwest Colorado Council of Governments Water Quality/

Quantity Committee December 2021

Climate Projections in Summit County, Colorado

MOUNTAIN

ROCKY

CLIMATE

the

Organization ii

The Rocky Mountain Climate Organization works to reduce climate disruption and its impacts to help keep the Interior West the special place we love. We do this in part by spreading the word about what a disrupted climate can do to us here, such as through reports like this, and also about what we can do to protect our climate.

www.rockymountainclimate.org

Acknowledgements

The authors wish to thank for providing assistance on this report: first and foremost, Torie Jarvis, Northwest Colorado Council of Governments Water Quality/Quantity Committee (QQ), via Dynamic Planning + Science, for her leadership in bringing together the local governments that with QQ sponsored this project and in managing this project; Alex Krebs, also of Dynamic Planning + Science; Jessie Burley and Haley Littleton, Town of Breckenridge; former County Commissioner Jill Hunsaker Ryan, John Gitchell, Maureen Mulcahy, and Justin Patrick, Eagle County; Mayor Hunter Mortenson, Don Reimer, Gilly Plog, and Ryan Thompson, Town of Frisco; former County Commissioner Karn Stiegelmeier, Nicole Valentine, and Michael Wurzel, Summit County; Kristen Bertuglia and Suzanne Silverthorn, Town of Vail;

Mike Nathan, Arapahoe Basin Ski Area; and John Fielder.

Cover photos, clockwise from the top:

Breckenridge and wildfire smoke over Peak 8, provided by Summit County, and Gore Range creek, provided by and © John Fielder.

Text, figures, and tables © 2021 the Rocky Mountain Climate Organization.

Permission is granted to reproduce and republish text, figures, and tables from this report if properly credited.

MOUNTAIN

ROCKY

CLIMATE

the

Organization

December 2021 update

This version of this report is an updated version of the report originally released in August 2021. The changes are the addition of a new figure (Figure 4), an update with final information on summer 2021 temperatures (on page 11), and miscellaneous minor corrections.

Dedication

This report and a companion report, Climate Projections in Summit County, Colorado, are both dedicated to Adam Palmer, who was the sustainability director for Eagle County until his untimely death in February of this year. Adam played an instrumental role in getting underway and shaping the project leading to these two reports—just one example of his many efforts to preserve and protect the quality of life in Colorado’s mountains.

Executive Summary

1. Introduction

2. Future Temperatures

Tables of High Temperature Projections

Tables of Low Temperature Projections

3. Future Precipitation

Tables of Precipitation Projections

4. Consequences

5. Methodology

Notes iii

CONTENTS

Executive Summary

This analysis of climate change projections for Summit County, Colorado, shows how much this area has at stake as human activities continue to change the climate, and how much difference it can make locally—how much the local climate would be protected—if heat-trapping emissions are sharply reduced.

This report, and a parallel one covering Eagle County, Colorado, identify in detail what climate models project for future local temperature and precipitation. Each report analyzes 24 million individual projections of daily high and low temperatures and precipitation amounts:

Hotter summers

The most striking projections are for how much hotter Summit County summers could become unless heat-trapping emissions are sharply reduced.

The following examples are with high future heat-trapping emissions for the Frisco/Dillon Reservoir area, including Frisco and the other communities around Dillon Reservoir and also representative of Breckenridge.

The hottest days of the year in this area are projected, according to the median projections from multiple models:

So an average summer day in mid-century could be three degrees hotter than the single hottest day of the late 20th century.

Even more telling might be how much more common hot days could become. Days 80° and hotter per year, according to the median projections:

• In typical mid-century years, would average 12 days—a thirty-fold increase over the baseline years, which averaged 0.4 such day every year.

• In the extreme year in mid-century, would occur 37 times—more than a full month’s worth. That compares to a high of four such days in 1970–1999.

The above projections are for average conditions across a grid of 18 miles by 14 miles, within which temperatures would vary with elevation. Frisco and Breckenridge are at lower elevations than the average for this grid, so their actual temperatures would be higher than the grid's averages. High temperatures in Frisco might be 3° hotter than the average for the grid, and Breckenridge might be about 1° hotter.

Sharp emission reductions halt temperature changes

There is good news, too, from this analysis—it also shows how completely the above changes can be avoided if global emissions of heat-trapping pollution are sharply reduced. With low future emissions (as well as with the other emission scenarios), temperatures in 2020–2039 would be higher than they were in 1970– 1999. But with low future emissions, they would not continue climbing after that.

The following median projections for the same Frisco/Dillon Reservoir area, but now with low future emissions, illustrate that.

The hottest days of the year:

82° in both 2060–2079 and 2080–2099.

Also in this area and also with low emissions, days 80° and hotter are projected:

• In 2020–2039 in typical years, to average three such days per year (compared to 0.4 day per year in

1970–1999).

• Then to stay the same thereafter, also averaging three such days in each of the next 20-year periods.

Warmer winters and earlier spring warm-up

Winters are also projected to get warmer. Temperatures were analyzed for the stretch from November 15 of one year to April 15 of the next year, chosen to approximate the core season for both snowpack accumulation and skiing and other snow-dependent sports. In the Breckenridge/Quandary area (including the Breckenridge Ski Resort) with high emissions, the median projections are that days with high temperatures above 32°:

• In mid-century typical years, would average 23 percent of the days in that snow/ski season—compared to 12 percent in 1970–1999.

• In the hottest mid-century year, would be 41 percent of those days—compared to a high of 25 percent in 1970–1999.

• In late-century typical years, would average 38 percent of those days, and in the extreme year would be 58 percent.

Days in late winter and early spring are also projected to warm up earlier. This analysis identifies projected changes in days that are 40° or hotter, warm enough to cause snowmelt and slushy skiing. For the stretch of March 16 through April 15, the last month of the core snow/skiing season, days that warm are projected for the Breckenridge/Quandary grid with high emissions:

• In mid-century typical years, to average 48 percent of the days in that stretch of 31 days—compared to a historical average of 30 percent.

• In the hottest year in mid-century, to be 73 percent of those days—compared to a high of 55 percent in 1970–1999.

• Late in the century, in typical years to average 63 percent and in the extreme year to be 79 percent of those late-season days.

Precipitation

Projections about precipitation are less certain than those about temperatures. Also, projections about summer precipitation are especially uncertain here because the models do not well represent this region’s summer monsoons. But the models are consistent in suggesting two changes in precipitation changes that are particularly noteworthy.

First, the models generally project that for the cold months of the year (November through April) continued heat-trapping emissions will lead to increased precipitation. For example, in the Vail Pass/North Tenmile Creek area with high emissions, the median projections are that the amount of cold season precipitation is projected:

Second, the models project that days with a modest amount of precipitation (less than one-quarter inch) will become less frequent but that days with heavier precipitation amounts will become more frequent. Again using as the example the Vail Pass/North Tenmile Creek area, the median projections are that with high emissions:

• Wet days with less than a quarter-inch of precipitation are projected to average four percent less frequent in mid-century, and nine percent less frequent late in the century, compared to 1970–1999.

• Days with a quarter-inch to a half-inch of precipitation are projected to average fifteen percent more frequent in mid-century, and 23 percent more frequent late in the century.

• Days with a half-inch or more of precipitation are projected to average eighteen percent more frequent in mid-century, and 33 percent more frequent late in the century.

Consequences

Assessing the impacts of these and other climate changes is beyond the scope of this report, but it is well documented in the scientific literature that the climate changes of the type projected here are likely to lead to a wide range of impacts, including the following.

• Higher temperatures increase the acreage burned in wildfires and the length of the wildfire season.

Projections range up to a nearly seven-fold increase in this region in area burned with only a modest increase in temperatures.

• Increases in wildfires obviously threaten people’s safety and property, particularly as development expands in fire-prone areas. More wildfire smoke also increases the risk of respiratory disease and mortality, and heavy precipitation on burned areas leads to more debris flows, such as the mudslides this summer in Colorado that have repeatedly closed mountain highways.

• The season for skiing, snowboarding, and other snow-dependent sports could be shorter and the snow slushier—reducing enjoyment for skiers, profits for skiing-dependent businesses, and tax revenues for state and local governments. If ski areas do not experience long enough stretches of sub-freezing temperatures, it is conceivable they will not be able to maintain snowy slopes, regardless of whether they have snowmaking equipment or the water supply, shortening the length of the ski season.

• Increased temperatures, especially the earlier occurrence of spring warmth, have already altered the water cycle across the West, with changes that include decreases in snowpack and its water content, earlier streamflows, and shifts in precipitation from snow to rain.

• Higher temperatures decrease water availability, by increasing evaporative losses from water bodies, soils, and plants, and increase irrigation requirements for crops and other outdoor plants.

• Higher temperatures and reduced river flows can reduce opportunities for fishing and rafting. Other impacts to summertime recreation and tourism could include losses of visitation and visitor enjoyment, for reasons ranging from temperatures too high for outdoor activities to disrupted transportation systems.

• Higher temperatures, especially if combined with drier summers, can increase tree mortality. In Colorado, tree mortality in subalpine forests has increased in recent decades, with the greatest increases occurring during hot, dry periods.

• Hotter and drier conditions can drive outbreaks of insects such as bark beetles as trees lose their resistance to infestations and as winters no longer have enough deep cold to limit beetle populations, allowing them to reach epidemic levels.

Especially when considered with additional scientific information on these and other possible impacts, the local climate projections analyzed in this report can help local governments, stakeholders, and the general public assess the possible future extent of these projected changes and their impacts in Summit County, and guide public and private decisions about taking actions both for climate protection and for climate change preparedness.

This analysis of climate change projections for Summit County, Colorado, shows how much this area has at stake as human activities continue to change the climate, and how much difference it can make locally—how much the local climate would be protected—if heat-trapping emissions are sharply reduced.

Analysis overview

This report, and a parallel one covering Eagle County, Colorado,1 identify in detail what climate models project for future local temperature and precipitation. Each report analyzes 24 million individual projections of daily high and low temperatures and precipitation amounts:

The analysis draws on the methodology developed in two parallel 2015 Rocky Mountain Climate Organization reports focused on Boulder and Larimer counties, funded by the Colorado Department of Local Affairs using federal disaster recovery funds, to help those localities and others prepare for wildfire and flooding threats as they become more extreme from continuing climate change, and expanded upon for two 2017 reports on projections for extreme heat and for precipitation in the Denver metropolitan area, funded by the City and County of Denver’s Department of Environmental Health.2 For the current reports, this methodology was adapted to focus on the climatic conditions most important in Colorado’s mountains. The Denver reports and the two new reports are, as far as the authors are aware, the most detailed analyses yet done of climate model projections for any locality. This is in part because these three reports analyze not just future average conditions but also future temperatures in extreme years—in the projected hottest year in each succeeding 20-year period over the rest of this century.

The analyses for Summit and Eagle counties certainly are the most detailed analysis of the details of climate model projections for Colorado mountain locations, and should be of interest not only in those counties but also in other locations in the Southern Rocky Mountains that could experience similar changes.

The projections analyzed in this report are from global climate models that have been downscaled to produce local results and made available on an online archive created by federal agencies and others.3 Similar downscaled projections have been used in many previous analyses, notably in this state Climate Change in Colorado: A Synthesis to Support Water Resources Management and Adaptation (Second Edition—August 2014), a report prepared by the Western Water Assessment (WWA) program at the University of Colorado Boulder for the state government’s Colorado Water Conservation Board.4 This WWA report remains the primary overall source focused on what climate change may be like in Colorado. In its report, WWA reported on many ways that climate change could be manifested in the state, including that with continued high increases in future heat-trapping emissions, statewide average temperature could increase by 2035–2064 by 2.5° to 5.5°* compared to 1971–2000 under the emission scenario identified in this report as "medium #1", or by 3.5° to 6.5° under the "high emissions" scenario (see page 7 for details on those scenarios). (These temperature ranges cover the middle 80 percent of the available projections from climate models, as illustrated by Figure 3 on page 9, but without a median projection).

The analysis done for this report differs from and adds to the WWA report in several ways:

• First and most importantly, this report analyzes model projections for conditions that could occur on individual days—projected high and low temperatures and precipitation amounts for every individual day from January 1, 2020, through December 31, 2099 (plus retrospective projections for daily

1. Introduction

*All temperatures presented here are in degrees Fahrenheit.

conditions in the 1970–1999 baseline). Projections for individual days have no particular reliability, but an analysis of the average and extreme daily conditions projected to occur over the 20-year periods used here identifies in detail how the models suggest that the future climate may change. Analyzing daily data in this way is rare, as it requires analyzing literally millions of more individual projections than the few thousands of projections for future decadal, annual, seasonal, and monthly averages typically analyzed by scientists (including by WWA in its report). The far more laborious process of analyzing daily projections makes it possible to identify such details as how often the models project that daily temperatures could reach certain thresholds.

• Second, this report considers projections from all four current scenarios for future levels of heat-trapping emissions (described on the next page). Considering all scenarios shows the full range of possibilities embodied in the models and shows how much difference will be made by the extent to which global emissions are reduced.

• Third, this analysis covers the full century, with results presented for four 20-year time periods—2020–2039, 2040–2059, 2060–2079, and 2080–2099. The first of those periods is a time span we have just entered, and covers the immediate planning horizon for local governments and others. Together, all four periods cover the expected lifetimes of children alive today.

For more, see the Methodology section on pages 47–48.

Geographic area

This analysis covers three separate, contiguous grids in Summit County, each one-quarter of a degree of latitude by one-quarter of a degree of longitude, or about 18 miles by 14 miles. Grids of this size are the smallest units for which the projections from the global climate models have been downscaled to yield local projections. These three grids, shown in Figure 1 on the next page, are identified in this report as:

• The Frisco/Dillon Reservoir grid, with elevations from about 9,000 to about 12,900 feet above sea level, with the lowest average elevation of these three grids. The grid includes Frisco, the northern portion of Breckenridge, portions of Silverthorne and Dillon, and unincorporated areas around Dillon Reservoir. The average high temperature in this grid in 1970–1999 was 48°, the average low temperature was 19°, and precipitation averaged 20 inches per year (see the tables at the end of the temperature and precipitation sections). Because the above named towns, including all of Breckenridge, are lower than the grid's average elevation, they would have higher temperatures than the grid's average (see the "caveat" below). This grid, however, would come the closest of the three grids in representing the temperatures of all these towns.

• The Breckenridge/Quandary grid, ranging from about 9,600 to 14,300 feet. The grid includes most of the town of Breckenridge—but, as stated above, because of its elevation the town itself is better represented by the Frisco/Dillon Reservoir grid (also see below). The grid includes the Breckenridge Ski Resort and other slopes and peaks of the Tenmile Range, to Quandary Peak at the range's southern end. The average high temperature in this grid in 1970–1999 was 43° (five degrees cooler than the Frisco/Dillon Reservoir average), the average low temperature was 16°, and the average annual precipitation was 31 inches.

• The Vail Pass/North Tenmile Creek grid, from about 9,200 to about 12,500 feet in elevation. The grid includes the drainage for North Tenmile Creek, the water supply for the Town of Frisco. The average high temperature in this grid in 1970–1999 was 44°, the average low temperature was 15°, and the average annual precipitation was 31 inches—all similar to the averages for the Breckenridge/ Quandary grid.

Caveat: Temperatures vary within grid

In the database of climate projections analyzed for this report, the values for temperature and precipitation are average values across a particular grid. All of these three grids include many different elevations, and temperature varies with elevation—decreasing by an average of about 3.4° with every 1,000 feet of additional elevation.5 This means that a grid's average temperature best represents the conditions at a grid's

Summit County grids analyzed in this report

Vail Pass

Frisco

Breckenridge

Quandary Peak

1 2

North Tenmile Creek

Breckenridge Ski Resort

Figure 1. The three grids for which climate projections were separately analyzed for this report: (1) the Vail Pass/North Tenmile Creek grid, (2) the Frisco/Dillon Reservoir grid, and (3) the Breckenridge/Quandary grid.

average elevation, and that a particular location at an elevation lower than that would be warmer than the grid's average and one at a higher elevation would be cooler.

A weather station near the shore of Dillon Reservoir, just inside the Frisco/Dillon Reservoir grid, provides an example that can be documented. The station is at 9,081 feet, one of the lowest elevations within that grid. Its average summertime (June-July-August) high temperature in the 1970–1999 baseline period was 71°.6 That is 3° hotter than the grid's comparable baseline value, reflecting the weather station's lower elevation. The towns of Frisco, Silverthorne, and Dillon are all at essentially the same elevation as the weather station, and could all be expected to also have summertime highs about 3° above the grid's average. The town of Breckenridge, at 9,600 feet, would also be warmer than the grid's average, probably

A. Annual Emissions of Carbon Dioxide

B. Atmospheric Concentrations of Carbon Dioxide

C. Radiative Forcing

Scenarios of Future Heat-Trapping Emissions

Figure 2. Key values for the emissions scenarios used in this analysis: A, annual global emissions of carbon dioxide in gigatons of carbon; B, atmospheric concentrations of carbon dioxide, in parts per million; and C, radiative forcing, or the average warming at Earth’s surface resulting from heat-trapping pollution, in watts per square meter. In all three parts of the figure, the blue lines represent the scenario identified as “high” in this report; the black lines, “medium #1”; the red lines, “medium #2”; and the green lines, “low.” Figures provided by D. van Vuuren.11 by about 1°. So the projections for the Frisco/Dillon Reservoir grid, while the best available for representing all four of these towns, likely understate the temperatures that would occur in the towns.

Emissions scenarios

Projections of the future climate vary depending on what is assumed about future levels of heat-trapping emissions, because the magnitude of future climate change depends primarily on cumulative emissions of heat-trapping gases and how sensitive Earth’s climate is in responding to those emissions.7

The four emission scenarios illustrating possible futures used to drive the climate projections analyzed in this report8 are:

• What is called here the high scenario. Officially known as Representative Concentration Pathway or RCP 8.5, a label chosen because it represents an increase in energy of 8.5 watts per square meter at Earth’s surface at the end of the century. To result in that level, it assumes little future change in emission reduction policies. It is not a true business-as-usual scenario, but is the closest to that of the four scenarios.9 In Figure 2 below, the high scenario is represented by the blue lines.

• A medium #1 scenario. Officially known as RCP 6.0, again based on the net energy change at the end of the century, it initially assumes the lowest emissions levels of all scenarios but then sharp increases.

From the 2060s to the end of the century, it assumes the second highest atmospheric concentrations of heat-trapping gases. It is represented by the black lines in Figure 2.

• A medium #2 scenario, or RCP 4.5. It starts out with higher emissions than medium #1 but then has major reductions after mid-century, as shown by the red lines in Figure 2.

• A low scenario, RCP 2.6. It assumes emissions cuts of more than 70 percent from current levels by 2050 and an elimination of net human emissions by about 2080. This likely would keep the average global temperature increase under 3.6° (2° Celsius) compared to pre-industrial levels, but not enough to achieve the official international goal of keeping it “well below” 3.6°, preferably to 2.7° (1.5° C).10 The low scenario is represented by the green lines in Figure 2.

No scenario is believed to be more likely than the others—instead, they were chosen to illustrate the range of plausible futures. Actual global emission levels in recent years have been lower than the high scenario’s pathway, but closer to it than to any other scenario.12 Which scenario turns out closest to reality depends on future public and private actions. (For more on the scenarios, see pages 47–48.)

Climate models

The climate projections used in this analysis were obtained by RMCO from an online archive created by the U.S. Bureau of Reclamation and other institutions.13 The projections are from the generation of climate models, known as CMIP5 models, that are the latest models have been downscaled to produce local projections. (For details, see pages 47–48.)

The climate projections obtained and analyzed for this report are of daily maximum and minimum temperatures and precipitation amounts for every day from 2020 through 2099, compared to actual observations for a baseline period of 1970–1999 (see page 47). One projection was obtained from each of the available models using the different emission scenarios—20 climate models for the high scenario, 12 for medium #1, 19 for medium #2, and 16 for the low scenario. In all, for this report more than 24 million individual projections of future weather were analyzed, covering the three weather values, 110 years, 67 pairings of climate models and emission scenarios, and the three separate grids.

In presenting results from each pairing of climate models and emission scenario, we report both the median of all relevant projections as well as the range from the 10th percentile to the 90th percentile of the projections—setting aside the lowest and highest values, as illustrated by Figure 3 below for the subsequent figures used in this report. As an example of how this is represented in text, the values shown by the sample column in Figure 3 below would be written as "20 (15–25)," with 20 being the median projection, 15 the 10th percentile, and 25 the 90th percentile.

Figure 3. Illustration of how individual projections are represented in the figures on temperature projections on the following pages. On the left, of all projections (hypothetically here from 10 models), the middle 80 percent are combined into a single column (on the right), with that range shown by the lighter color of the upper portion of the column. Another way to express this is that the brighter portion of the column shows the range from the 10th percentile to the 90th percentile of all projections. The numeral in the column shows the median of all projections.

Median of all projections

Range of the middle 80% of projections

How this report’s figures represent the projections

Combining multiple projections . . . into one summary column

Ninety percent of the models project at least as much change as is represented by the dividing line between the brighter and the darker colors in a column. That line shows the 10th percentile of the projections—meaning that ten percent of the projections are below there, so ninety percent are above that level.

2. Future Temperatures

The model projections obtained for this report were analyzed to consider 74 different measures of how local temperatures may change, including 39 measures of high temperatures and 35 of lows. This report focuses just on the most telling of those projections, and for simplicity the text uses examples from only the high and low emission scenarios and primarily from the mid-century (2020–2039) and late century (2080– 2099) time periods. The figures and tables throughout the report, though, cover all four emission levels and all four time periods. Spreadsheets with the results from the analysis of all 74 temperature values (as well as of all precipitation values) are available online at www.rockymountainclimte.org/extremes/summit.

The tables at the end of this section document the statements made here, unless another source is indicated.

Baseline temperature values (and baseline precipitation values in section 3) used as comparisons to the model projections are gridded observed values for 1970–1999. These are derived from records from weather stations that have been converted into averages for each grid (see page 47).

Average temperatures

To begin with, the average annual high temperature in, for example, the Frisco/Dillon Reservoir grid is projected with high emissions to increase by mid-century by 5° (3–7°)* and by late in the century by 9° (8–12°), compared to 1970–1999 (see the online data). Somewhat larger temperature increases are forecast for hotter months than for colder ones, by a margin of about 1–2° in mid-century, depending on the grid.

For example, in the Frisco/Dillon Reservoir grid in mid-century, temperatures in the hot months (May through October) are projected to average 6° (5–8°) hotter than in the baseline years, and the temperatures in the cold months (November–April) are projected to be 4° (3–6°) warmer. In this case, that is about a two-degree difference. (Throughout this report when one grid is used as an example in text or figures, the other grids show similar patterns, as shown by the data in the appropriate tables.)

But what, exactly, does an increase of a few degrees in average temperature mean across a year or six months? To understand how daily life could be different for people living in and visiting Summit County, it helps to look at daily data. Analyzing millions of projections of hot and cold temperatures on individual days makes it possible to extract more vivid pictures of the possible futures before us.

Hotter summers

The most striking projections are for how much hotter Summit County summers could become unless heat-trapping emissions are sharply reduced.

The following two sets of examples are for the Frisco/Dillon Reservoir grid with high emissions. The tables at the end of this section show comparable values for the other grids.

The hottest days of the year in the grid are projected:

1970–1999.

*Here, 5° is the median projection, 3° is the 10th percentile, and 7° is the 90th percentile (see the previous page).

In the Frisco/Dillon Reservoir grid, an average summer day in mid-century could be three degrees hotter than the single hottest day of the late 20th century.

Note that, because their elevations are lower than the grid's average elevation, Frisco and other communities around Dillon Reservoir could be expected to have temperatures about 3° hotter than the above projections, and Breckenridge could be expected to be about 1° hotter (see pages 5–6).

Even more telling might be how much more common hot days could become. Also in the Frisco/Dillon Reservoir grid with high emissions, days 80° and hotter per year are projected:

• In typical mid-century years, to average 12 (6–27) days. That median projection is for a thirty-fold increase over the baseline years, which averaged 0.4 day per year.

• In the extreme year in mid-century, to be 37 (20–55) days. That median projection is more than a full month’s worth. By contrast, the highest number in 1970–1999 was four such days.

• In typical years late in the century, to average 54 (38–86) days. That median projection, nearly two months’s worth.

• In the extreme year late in the century, to occur 89 (74–115) times. That median projection, a full summer’s worth.

Again, because their elevations are lower than the grid's average elevation, for Frisco and other communities around Dillon Reservoir (around 9,000 feet in elevation), the above projected number of days per year would correspond to about that many days 83° and hotter, instead of 80° and hotter. For Breckenridge, the above projections would correspond to days 81° and hotter. (See pages 5–6).

Following are the corresponding projections for the Breckenridge-Quandary grid (at a significantly higher elevation, and so with a much cooler climate), also with high emissions.

The hottest days of the year are projected:

Days 75° and hotter per year are projected:

• In typical mid-century years, to average seven (4–20) days, compared to 0.6 day per year in the baseline period.

• In the extreme mid-century year, to number 31 (19–51) days, compared to seven in the hottest baseline year.

Caveat: Projected high temperatures could be higher

It is worth pointing out that the climate models appear not to actually capture how hot the future extremes could be, as the federal government's climate science programs says in the following quote. So the values stated here for future hot days and extreme years could well be underestimated.

“Climate models are more likely to underestimate than to overestimate the amount of long-term future change; this is likely to be especially true for trends in extreme events.”

U.S. Global Change Research Program14

Comparison to summer 2021 The summer of 2021 has been hot enough to prompt the question, how does this summer compare to the projections for typical years and the extreme year for the period 2020–2039, which has just begun? Data from the Dillon weather station described on pages 5–6 offers the best comparison. Reducing the station's recorded daily high temperatures by 3° to compensate for its lower elevation and higher temperatures than the averages for the grid (see pages 5–6) produces approximate average values for the grid. Allowing for that 3° difference, the station's records for summer (June–August) 202115 are very close to the median projections with high emissions for conditions averaged across the grid for a typical year in 2020–2039. Key comparisons are:

• The average high temperature at the weather station in summer 2021 was 74°, which would correspond to a 71° average across the grid—one degree shy of the 72° projected for a typical year's average high temperature across the grid in 2020–2039.

• The station's hottest temperature for the year was 85°, arguably corresponding to a grid-average high of 82°—or one degree below the 83° projected for a typical year's hottest day across the grid in 2020– 2039.

• The station's daily high reached 83° or hotter six times in 2021, corresponding to that many days 80° or hotter across the grid—or two more than the four 80°-plus days projected for a typical 2020–2039 year.

Compared instead to the extreme year projected with high emissions in the 2020–2039 period, the summer of 2021 fell short:

• The 85° temperature of the hottest day of 2021 at the weather station would correspond to an 82° hottest day across the grid—four degrees short of the 86° projected for the extreme year in that period.

• The six 83°-plus days at the weather station in 2021 would correspond to that number of 80°-plus days—nine fewer than the fifteen such days projected for the extreme year.

This suggests that the summer of 2021 in this grid resembles a typical 2020–2039 year—but is not close to the hottest year projected to occur in that period.

Sharp emission reductions halt temperature changes There is good news, too, from this analysis. The most important is that it shows how completely the above changes can be avoided if global emissions of heat-trapping pollution are sharply reduced. The following examples also are for the Frisco/Dillon Reservoir grid, but now with the scenario of low emissions.

With low future emissions, the hottest days of the year are projected:

• In 2020–2039 in typical years, to average 82° (81–83°)—about three degrees higher than in the baseline.

• Then to remain essentially the same for the rest of the century—in 2040–2059, 83° (81–84°); in

2060–2079, 82° (81–84°); and late in the century, again 82° (81–84°).

Also with low emissions, days 80° and hotter are projected:

and in 2020–2039, again 3 (1–7) such days.

These examples show, as do all other data in the figures and tables in this section, that with the low emission scenario, the period 2020–2039 will have higher temperatures than the recent past, but after then essentially no further temperature increases are projected for later in the century.

Figure 4 on the next page shows projections for the hottest days in the Frisco/Dillon Reservoir grid in typical years for each 20-year period. Figure 5 on the following page shows projections for the number of 80°-plus days in the same grid for typical years, and Figure 6 on page 14 shows the number of projected 80°-plus days in the gird in extreme years (the single hottest years) in those 20-year periods.

If global emissions are brought down to low levels, essentially no further temperature increases are projected for later in the century beyond those that could occur in 2020–2039.

Hottest day of the year in the gridFrisco/Dillon Reservoir Typical years: Averages for each 20-year period

83° 82°

85°

83° 84°

83°

85°

82°82° 82°82°

High LowMedium #2Medium #1

Future emissions

1970–1999 2080-20992040–2059 2060–20792020–2039

79° 80°

85°

90°

95°

87°

90°

86°

84° 85°

Figure 4. Observations and projections for the highest daily high temperature of the year, averaged across the Frisco/Dillon Reservoir grid for the indicated time periods. Observations for the 1970–1999 baseline period are from observed/gridded data (see page 480). For each future period, the four columns represent different projections based on the emission scenarios identified on page 7. For each such column, the brighter color on the top of the column shows the range of the middle 80 percent of the projections (from the 10th percentile to the 90th percentile); the numerals are the medians, as illustrated in Figure 3 on page 8. For the data illustrated here, see tables 1 and 2 on pages 21 and 22.

In the Frisco/Dillon Reservoir grid, an average summer day in mid-century could be three degrees hotter than the single hottest day of the late 20th century, which was 82° (see Table 1 on page 21).

The four emission scenarios illustrate possible futures. Which one turns out to be closest to reality depends on future public and private actions.

In a future with continued high emissions, the Frisco/Dillon Reservoir grid in mid-century could average twelve 80°-plus days a year— compared to 0.4 per year in the recent past. Late in the century, the average could be 54 days.

Figure 5. As Figure 4 on the previous page, but for the number of days per year with high temperatures (averaged across the grid) of 80° or higher.

Days 80° or hotter in the Frisco/Dillon gridReservoir Typical years: Averages for each 20-year period

3 33

High LowMedium #2Medium #1

Future emissions

2080-20992040–2059 2060–20792020–20391970–1999

0.4

Figure 6. As Figure 5 on the previous page, but for the hottest individual year in each period—on the left, the highest number of 80°-plus days in 1970–1999, and on the right the highest projected number of such days in each of the 20-year periods. For the data illustrated here, see tables 1 and 2 on pages 20 and 21.

With high emissions, the hottest year in mid-century could have more than a month's worth of days 80° or hotter, and late in the century a whole summer's worth.

With low emissions, the hottest year in each period could stay about the same.

Days 80° or hotter in the Frisco/Dillon gridReservoir Extremes: The hottest year in each 20-year period

15 16

High LowMedium #2Medium #1

Future emissions

1970–1999 2080-20992040–2059 2060–20792020–2039

The other two Summit County areas, the Breckenridge/Quandary and Vail Pass/North Tenmile Creek grids, are at higher elevations, where 80° days are historically unknown and begin to appear in the projections only in extreme years after mid-century. The figures below and on the next page illustrate instead the number of days 75° and hotter in the Breckenridge/Quandary grid.

Figure 7. As figure 5 on page 13, but for days 75° and hotter in the Breckenridge/Quandary grid. For the data illustrated here, see tables 3 and 4 on pages 23 and 24.

The Breckenridge/Quandary grid, similar in elevation to the Vail Pass/North Tenmile Creek grid, has similar projections for future 75°-plus days.

Days 75° or hotter in the Breckenridge/Quandary grid Typical years: The average year in each 20-year period

2 1

3 4

2 22

High Low Medium #2Medium #1

Future emissions

1970–1999 2080-20992040–2059 2060–20792020–2039

0.6

Days 75° or hotter in the Breckenridge/Quandary grid Extremes: The hottest year in each 20-year period

High LowMedium #2Medium #1

Future emissions

1970–1999 2080-20992040–2059 2060–20792020–2039

Figure 8. As figure 6 on page 14, but for days 75° and hotter in the Breckenridge/Quandary grid. For the data illustrated here, see tables 3 and 4 on pages 23 and 24.

Warmer winters

Winters are also projected to get warmer. This analysis considered temperatures from November 15 of one year to April 15 of the next, chosen to approximate the core season for both snowpack accumulation and skiing and other snow-dependent sports. The examples on this page are for the Breckenridge/Quandary grid (including the Breckenridge Ski Resort) with high emissions.

High temperatures above freezing.—The number of days with highs above 32° in the November 15 through April 15 stretch (which is 152 days long, or 153 in leap years) are projected:

• In typical years in mid-century, to average 54 (48–64 percent) of those days (or a count of 82 (74–98) days). That compares to an average of 38 percent in 1970–1999.

• In the extreme year in mid-century, to be 70 (57–78) percent of those days—compared to a high of 52 percent in the baseline period.

Figure 9 on the following page illustrates these projections.

On the other hand, some other good news: Over the same snow/skiing season, no nights are projected to have low temperatures above freezing, even in extreme years late in the century.

Earlier spring warm-up

Days in late winter and early spring are also projected to warm up earlier. As winter turns to spring, days appreciably above freezing start melting snowpacks and turning ski slopes slushy. As a measure of changes in this transition, the projected number of days 40° or hotter, warm enough to cause snowmelt and slushy skiing, were analyzed for the stretch of March 16 through April 15, chosen to represent the last month of the core snow/skiing season. In the Breckenridge/Quandary grid with high emissions, the percentage of 40°-plus days in that 31-day stretch is projected:

• In mid-century in typical years, to average 48 (41–55) percent of those days—compared to 30 percent in 1970–1999.

• In mid-century in the extreme year, to be 73 (57–84) percent of those days—compared to 55 percent, the high number in 1970–1999.

The first month of the core snow/skiing season, by contrast, does not have as many 40°-plus days to begin with, nor a similar projected growth in such days. For instance, 40°-plus days in the Breckenridge/Quandary grid in November 15 through December 14 with high emissions are projected to average seven (6–10) such days in typical mid-century years, compared to four such days in the baseline period (see the online data).

This and related data posted online suggests that temperature changes may be more likely to have an effect at the end of the snow/skiing season than at the beginning.

With high emissions, temperatures above freezing could occur in twice as many days in the snow/ski season in typical mid-century years, compared to 1970–1999.

Figure 9. As figure 5 on page 13, but for days 32° and hotter from November 15 of one year through April 15 of the following year in the Breckenridge/Quandary grid. The number of days in that stretch is 143 (144 in leap years), and the actual average number of days in 1970–1999 was 57. For the data illustrated here, see tables 3 and 4 on pages 23 and 243. (In those tables, the data shown above are presented as the number of such days in the stretch of November 15 through April 15; the figure above instead shows the data as the percentage of such days in that stretch.)

Percent of Nov 15-Apr 15 days w/ highs above 32° in Breckenridge/Quandary grid Typical years: The average year in each 20-year period

18 17

20 21

27 27

High LowMedium #2Medium #1

Future emissions

1970–1999 2080-20992040–2059 2060–20792020–2039

With high emissions, the percentage of days in the core snow/skiing season that get above freezing in the Breckenridge/Quandary grid could double by mid-century and triple by late in the century.

Percent of Nov 15-Apr 15 days w/ highs above 32° in Breckenridge/Quandary grid Extremes: The hottest year in each 20-year period

44 43

33 3333

High LowMedium #2Medium #1

Future emissions

1970–1999 2080-20992040–2059 2060–20792020–2039

With low emissions, in the hottest year in the 2020–2039 period, one third of days in the core snow/skiing season could get above freezing in the Breckenridge/Quandary grid. That would be higher than in the recent past, but the number could stay at one third of the days even in the hottest years over the rest of the century.

Figure 10. As Figure 9 on the previous page, but for the hottest individual year in each period—on the left, the highest percentage of days above 32° of any single year in 1970–1999, and on the right the highest projected percentage for any single year in each 20-year period. For the data illustrated here, see tables 3 and 4 on pages 23 and 24.

Tables of High Temperature Projections Tables 1 through 6 that follow (through page 26) present key results of the analysis of the climate models’s projections of future high temperatures. After those, tables 7 through 12 (on pages 27–32) do the same for low temperatures. The tables are in pairs, with the first of each pair showing projections for the first two 20-year periods (2020–2039 and 2040–2059) for a particular grid and the next table showing projections for the two subsequent periods.

Full results from the analysis of all 74 temperature values considered in this project can be found at www.

rockymountainclimate.org/extremes/summit.

Typical years

Cold months avg high (Nov thru Apr), change

Hot months avg high (May thru Oct), change

Average high in Jun-Jul-Aug

Days per year 75° or hotter

Days per year 80° or hotter

High temp of year’s hottest day

Days/yr Nov 15–Apr 15 32° or hotter

Days/yr Mar 16–Apr 15 40° or hotter

Extreme years

Average high in June-July-August

Days per year 75° or hotter

Days per year 80° or hotter

High temp of year’s hottest day

Days/yr Nov 15–Apr 15 32° or hotter

Days/yr Mar 16–Apr 15 40° or hotter

35°

53°

68°

0.4

79°

71°

82°

+3° (+1 4°)/+

+4° (+2 5°)/+

72° (71–73°)

(25–45)

(2–6)

83° (82–83°)

(90 107)–

(16–21)

74° °(72–77 )

(38–73)

(8–33)

86° (84– )88°

(109–131)

(23–27)

+2° (+1 3°)/+

+3° (+2 4°)/+

71° (70–72°)

( –32)21

(1–5)

82° (81–82°)

(91–101)

(16–21)

73° (72–7 )5°

(31–63)

(4–19)

84° (83–86°)

(107–130)

(20–26)

+3° (+1 4°)/+

+4° (+2 4°)/+

72° (70–72°)

( –40)22

(1–4)

82° (81–83°)

(90–104)

(16–22)

74° (72–76 )°

(39–61)

(9–25)

85° (82– )88°

(107–130)

(21–28)

+2° (+2 4°)/+

+3° (+2 4°)/+

72° (70–73°)

( –40)24

(1–5)

82° (81–83°)

(92–105)

(16–23)

74° (72–75 )°

(38–62)

(8–22)

84° (83–8 )6°

(109–130)

(22–29)

+4° (+3 6°)/+

+6° (+5 8°)/+

74° (73–76°)

(45–73)

(6–27)

85° (83–87°)

(98–117)

(17…