Home » Climate Change » Weather Patterns (Page 4)

Category Archives: Weather Patterns

Heavy Precipitation Events on the Increase

Climate predictions say there will be an increase in heavy rain events. In Missouri the average yearly precipitation may or may not change significantly (see series of posts beginning here). But even if the yearly average doesn’t change, more of it will come during heavy precipitation events, between which there will be less precipitation.

Figure 1. Source: GlobalChange.gov.

Figure 1. Source: GlobalChange.gov.

It is already happening. Figure 1 shows the change in the frequency of heavy rain events in the contiguous 48 states. As you can see, the trend shows a clear increase over time. They occur at least 50% more frequently than they did in the 1920s and 1930s.

Here’s how they measured “heavy precipitation event” for this chart: for the years 1901-1960 they computed the two-day precipitation total that is only exceeded once every 5 years. They then looked at each decade and counted how many times there had been a two-day precipitation total that exceeded that amount. They presented the results as a percentage deviation from the expected number. During the 1910s, 1920s, and 1930s, there were fewer than expected heavy rain events. During the 1990s and 2000s, there were 30-40% more than expected.

Figure 2. Source: GlobalChange.gov.

Figure 2. Source: GlobalChange.gov.

Figure 2 shows something a little different. Here, they compared the amount of precipitation falling in the heaviest 1% of precipitation events from 1958-2012. The figure shows that it increased 71% in the Northeast, 37% in the Midwest, and 27% in the Southeast. The only region where it decreased was Hawaii.

These charts show that not only are heavy precipitation events occurring more frequently, but more precipitation is falling during them.

“Heavy Precipitation Event” is a term that doesn’t convey much. Let’s look at two examples. One occurred in Missouri almost a year ago, in late December, 2016. Storms occurred over all of Missouri, with the heaviest precipitation roughly paralleling Interstate 44’s route through the state. Over 3 days these storms dumped up as much as 10 inches of rain in a band running from Branson to St. Louis. The result was record flooding along the Meramec River (Figures 3 and 4).

Figure 3. Source: U.S. Weather Service.

Figure 3. Source: U.S. Weather Service.

Flooding in Valley Park, December, 2016. Source: Unknown photographer.

Figure 4. Flooding in Valley Park, December, 2016. Source: Unknown photographer.

 

 

 

 

 

 

 

 

In August, 2016, similar flooding occurred in Louisiana, but this was even worse (See Figure 5).

Figure 5. Source: Department of Agriculture.

Figure 5. Source: Department of Agriculture.

Up to 31 inches of rain fell over a period of 4 days, and record flooding followed. It was the worst U.S. disaster since Hurricane Sandy in 2012, with 146,000 homes damaged. The dollar amount of damage represented is still being figured out.

Climate change modelers, of course, predicted an increase in heavy precipitation events. It’s already happening, and these floods shows just what it means.

Sources:

“2016 Louisiana Floods.” Wikipedia. Viewed online 11/13/2016 at https://en.wikipedia.org/wiki/2016Louisiana_floods.

National Weather Service. December Historic Rainfall and Flooding Event Review. St. Louis Forecast Office. Downloaded 2016-11-13 from http://www.weather.gov/lsx/12_26_2015.

Unknown photographer. Meramec River at Valley Park. Advanced Hydrologic Prediction Service. Downloaded 11/13/2016 from http://water.weather.gov/ahps2/hydrograph.php?gage=vllm7$wfo=lsx.

GlobalChange.gov. Broadcast_Heavy_Precipitation-map_V2. National Climate Assessment 2014. Downloads, Graphics (Broadcast). Downloaded 11/13/2016 from http://nca2014.globalchange.gov/downloads.

GlobalChange.gov. Broadcase_Trends-in-heavy-precip_V2. National Climate Assessment 2014. Downloads, Graphics (Broadcast). Downloaded 11/13/2016 from http://nca2014.globalchange.gov/downloads.

Hot First Half of 2016

Figure 1. Source: Centers for Environmental Information.

Figure 1. Source: Centers for Environmental Information.

The first 6 months of 2016 were the third hottest ever across the United States, according to data from the Centers for Environmental Information (See Figure 1). The average temperature was 50.75°F, which is 3.22°F above the average for the 20th Century.

(Click on chart for larger view.)

 

.

.

.

Figure 2. Source: Centers for Environmental Information.

Figure 2. Source: Centers for Environmental Information.

During the same period, precipitation across the country was slightly above average, at 15.58 inches (Figure 2), which is 0.27 inches above the average for the 20th Century.

.

.

.

.

.

.

Figure 3. Source: Centers for Environmental Information.

Figure 3. Source: Centers for Environmental Information.

Combined, the temperature and precipitation resulted in moister than average soil conditions as measured by the Palmer Drought Severity Index. For the first half of 2016, the PDSI was 2.61, which is 2.23 above the average for the 20th Century.

.

.

.

.

.

MO Temp 2016 Jan-Jun

Figure 4. Source: Centers for Environmental Information.

In Missouri, the temperature for January-June was the 7th highest on record, at 52.9°F. It was 3.1°F hotter than the average for the 20th Century.

.

.

.

.

.

Figure 5. Source: Centers for Environmental Information.

Figure 5. Source: Centers for Environmental Information.

Precipitation in Missouri was low, however, at 16.86 inches, a shortfall of 3.70 inches compared to the average for the 20th Century. For those of us here in St. Louis, this may come as a bit of a surprise, as our local rain has been the 5th highest on record for the first half of the year.

.

.

.

.

.

Figure 6. Source: Centers for Environmental Information.

Figure 6. Source: Centers for Environmental Information.

The PDSI in Missouri during January-June was 2.90, which is 2.67 above the average for the 20th Century.

I’m not sure why significantly above average temperature and below average precipitation should result in above average soil moisture. Generally, high temperature and low precipitation is thought to resulting in dry soil conditions. If anybody knows, please write a comment and let us all know.

I also follow the Palmer Hydrologic Drought Index for the Northern Rockies and Plains. The PHDI measures long-term trends in soil moisture, which are thought to affect river and reservoir levels. I follow this region because it is the watershed for the Missouri River, the most important source for drinking water in Missouri. For the first half of 2016, the PHDI in the Northern Rockies and Plains was 1.29, which ranks exactly in the middle of the years measured (since 1895).

In California, which I have also been following, the temperature for January-June was 57.1°F, the 3rd highest on record, 3.8°F above average. Precipitation has been 15.04 inches, which is 0.86 inches above average. The precipitation came primarily in the first 3 months of the year, and the last 3 have been slightly drier than average. Since the bulk of California’s water comes during the winter, it is good that those months were the above average ones. The PDSI is -2.22, which is 2.20 below average. The soil in California is still very dry.

Source:

Centers for Environmental Information. Climate at a Glance. Data accessed and downloaded 7/25/2016 at http://www.ncdc.noaa.gov/cag/time-series/us.

Final California Snowpack Reading: Below Average

I’ve been keeping a watch on the snowpack in California this winter. The snowpack is California’s most important source of water. It matters for Missouri because of the important role California plays in the national economy and in our food supply.

Figure 1: Statewide California Snowpack Water Content, 3/30/16. Data source: California Department of Water Conservation

Figure 1: Statewide California Snowpack Water Content, 3/30/16. Data source: California Department of Water Conservation

On March 30 this year, the water content of the snowpack was 87% of its historical average for that date. Last year on April 1 it was virtually nonexistent (Figure 1). The period around April 1 is when the snowpack is at its peak, and the amount of water in it now determines how much water California will have during the dry summer and fall months.

(Click on chart for larger view.)

Eighty-severn percent is obviously better than 5%. However, it is still below average. Water officials were hoping for a larger than average snowpack to begin reversing the multi-year drought the state has experienced. It was a winter with a very large El Niño, and those years typically bring California lots of precipitation, including lots of snow in the Sierra Nevadas.

Figure 2. Data source: Mammoth Mountain Ski Resort

Figure 2. Data source: Mammoth Mountain Ski Resort

Snowfall data from Mammoth Mountain, one of California’s largest ski resorts, indicates an above average snowfall through March 31: a cumulative 342 inches of snow, vs. 304 in an average year. (Figure 2) Statewide data from Climate at a Glance suggests that precipitation throughout the state was 0.54 inches above average.

.

.

.

.

Figure 3. Data source: National Centers for Environmental Information

Figure 3. Data source: National Centers for Environmental Information

If California received above average precipitation, yet has a below average snowpack, then one of the causes must have been a warm winter, causing the snow to melt. Indeed, as Figure 3 shows, California had a warmer than usual winter, 2.4°F above average.

.

.

.

.

Figure 4. Data source: California Department of Water Resources

Figure 4. Data source: California Department of Water Resources

California’s man-made reservoirs are in better shape than last year. As Figure 4 shows, Lake Shasta and Lake Oroville water levels are above historical averages. (The red lines show the historical average for this date, the blue bars the current level of the reservoir. The yellow bars show the reservoir’s total capacity.) Some reservoirs, especially ones farther south, are still quite low.

Notice how near the top of the yellow bar some of the red lines are. California’s reservoirs, especially the big northern ones, are typically pretty full at this time of year. Typically, California does not receive very much rain from now through the end of November. As water is drawn out of the reservoirs to service California’s needs, snowmelt from the snowcap flows into the reservoirs to recharge them. This process is largest during the spring, but it continues at reduced rates all summer and into the fall. A small snowpack means that there is less water to recharge the reservoirs when California needs it most.

Lake Mead is at its lowest level in the last 10 years for this time of year.

California has seen a partial easing of the very severe drought of the previous two years, but nothing to signal the drought’s end. This was supposed to be a wet winter leading to an above average snowpack, but because of increased temperatures, the snowpack was below average. The El Niño is now weakening and is forecast to end by summer. What will happen then, nobody knows. Lake Mead continues to lose water. Long term, California continues to face serious challenges regarding its water supply.

Sources:

California Department of Water Resources. 2016. Conditions for Major Reservoirs: 05-APR-2016. Downloaded 4/6/16 at http://cdec.water.ca.gov/cdecapp/resapp/getResGraphsMain.action.

Water-data.com. 2016. Lake Mead Water Database. Database accessed 4/6/16 at http://lakemead.water-data.com.

National Centers for Environmental Information. 2016. Climate at a Glance. This is a data portal. Accessed 4/6/16 at http://www.ncdc.noaa.gov/cag/time-series/us.

Summary of Climate Projections for St. Louis

In the last 3 posts, I’ve reported on three studies making climate projections for St. Louis or Columbia by Hayhoe, VanDorn, Naik, and Wuebbles (2009), by Posey (2014), and by Anderson, Gooden, Guinan, Knapp, McManus, and Shulski (2015).

Because there is uncertainty associated with all climate projections, and because that uncertainty is magnified when making projections at the local level, there is value in having multiple projections from multiple studies. It can help to reduce some sources of uncertainty, though it cannot eliminate the uncertainty, by any means.

However, it is not always common for different climate studies to replicate methods precisely. Differences in method affect the resulting projections, and it means that the projections from these three studies should only be compared with great care. Some of the methodological differences include studying slightly different regions, making projections for different time periods, and using different climate scenarios.

Perhaps a bit of an amplification on the climate scenarios is in order here. It is thought that changes in climate will be sensitive to how humans respond to climate change, especially how we change or don’t change our greenhouse gas emissions. Climate change is already occurring, and most climate scientists agree that it is going to cause negative effects, no matter what we do – it is already too late to prevent them entirely. If GHG emissions rise from current levels, however, climate change is expected to be more severe. If we significantly curtail GHG emissions, climate change is expected to be less severe. If we follow a middle path, climate change is expected to follow a middle path.

Both climate change itself and mitigating climate change are expected to have implications for a variety of socio-economic factors – economic growth, poverty, social justice, migration, etc. To try to take these factors into account, climate scientists developed what they called scenarios. Each scenario outlined a path the world could take that included GHG emissions and a variety of socio-economic factors.

Figure 1: Total cumulative carbon dioxide emissions under different climate scenarios. Source: IPCC 2000.

Figure 1: Total cumulative carbon dioxide emissions under differing climate scenarios. Source: IPCC 2000.

Three scenarios, the A1fi, the A2, and the B1, were used to make climate projections by these authors. They are shown on Figure 1. This chart is now 16 years old, but it shows the original conceptualization. The lines depict the expected cumulative amount of carbon dioxide emissions projected to occur under various scenarios. The A1fi scenario was projected to result in the highest emissions in the group shown, though by no means the highest possible emissions. The A2 scenario was projected to have somewhat lower emissions, and the B1 lower still, though by no means the lowest possible.

(Click on chart or table for larger view.)

You don’t have to look at this chart for very long before you understand that, other things being equal, climate projections using the A1fi Scenario might be expected to result in larger changes in climate, while those using the A2 scenario might result in smaller changes, and those using the B1 scenario might result in smaller changes yet.

Table 1 summarizes some of the projections from the three studies I’ve been reporting on.

Summary Table

You can’t make direct comparisons between the studies, but some trends can be seen:

  • All three studies project temperature to increase by mid-century. Those that made projections for the end of the century projected it to increase even more by then.
  • The two studies that made projections for mid-century for each of two scenarios found that temperature would increase more under a higher emission scenario than a lower emission scenario.
  • The two studies that made projections related to heat waves agreed that the number of very hot days would increase.
  • All three studies projected that average annual precipitation would increase by a small amount.
  • Four out of five projections were for decreased precipitation in summer, with an increase during other seasons of the year.
  • All three studies projected an increase in the number of days with heavy rain. Both studies that projected the maximum amount of rain over multi-day periods projected an increase.
  • Both studies that projected the number of frost-free days per year projected that they would increase.

Only in two areas were there significant disagreements between the studies. First, the season in which the greatest change in temperature would occur differed, with one study projecting summer, one spring, and one winter. Second, of the two studies that projected seasonal changes in precipitation under the low emission scenario, one predicted a summer decrease and one predicted a summer increase.

Temperature projections using the A1fi, A2, and B1 scenarios followed the trend that would be expected from GHG emissions under those scenarios: higher, middle, and lower. Projections were made for 30-year periods centered on 2035, 2050, 2065, and 2084. The temperature was projected to increase more with each passing period.

Thus, despite their methodological differences, there appears to be a good deal of consistency between the results of these studies. They point to a warmer, slightly wetter local climate, with an increased frequency of heavy rain events.

Sources:

Anderson, Christopher, Jennifer Gooden, Patrick Guinan, Mary Knapp, Gary McManus, and Martha Shulski. 2015. Climate in the Heartland: Historical Data and Future Projections for the Heartland Regional Network. Downloaded 3/15/16 from http://www.marc.org/Government/GTI/pdf/ClimateintheHeartlandReport.aspx.

Hayhoe, K, J VanDorn, V. Naik, and D. Wuebbles. 2009. “Climate Change in the Midwest: Projections of Future Temperature and Precipitation.” Technical Report on Midwest Climate Impacts for the Union of Concerned Scientists. Downloaded from http://www.ucsusa.org/global_warming/science_and_impacts/impacts/climate-change-midwest.html#.VvK-OD-UmfA.

Intergovernmental Panel on Climate Change. 2000. IPCC Special Report: Emissions Scenarios: Summary for Policymakers. https://www.ipcc.ch/pdf/special-reports/spm/sres-en.pdf.

Posey, John. 2014. “Climate Change in St. Louis: Impacts and Adaptation Options.” International Journal of Climate Change: Impacts and Responses. Vol 5, #2. Downloaded 1/15/2016 from http://ijc.cgpublisher.com/product/pub.185/prod.233.

Climate Predictions for Columbia, MO

This is the third in a series of posts about climate prediction for Missouri. This post reports on a study by Anderson, Gooden, Guinan, Knapp, McManus, and Shulski (2015) that made projections for 5 midwestern cities: Columbia MO, Iowa City IA, Lawrence KS, Lincoln NE, and Oklahoma City OK. I will look only at the results for Columbia.

The Anderson group used a method that is different from what I have seen previously. I am not a climate scientist, so this doesn’t mean the method is improper, it only means that, in trying to report the results using ordinary English, I’m unsure how to describe them. I noted in my previous two posts that climate change is expected to be very sensitive to how humans respond to it. In particular, if we sharply reduce our GHG emissions, it is expected to reduce the amount of climate change, mitigating its effects. If we continue to emit high levels of GHG, or even increase them, it is expected to increase the amount of climate change, exacerbating its effects. Climate scientists have developed a standard suite of scenarios to describe how the world might respond. Some envision reduced emissions, some envision unchanged emissions, and some envision increased emissions.

Every climate projection I have seen has modeled more than one scenario. Typically, scientists have used a high emissions scenario and a low emissions scenario. They have presented results for both scenarios. The Anderson group proceeded differently. They modeled an increased emission scenario (A1fi, the same one used by the Hayhoe group), a less high emission scenario (A2, the same one used by Posey), and a low emission scenario (B1, used by both the Hayoe group and Posey). Then, instead of presenting the results of each, they averaged them. Their rationale in doing so was to present only the “signals that are strongest across a range of circumstances.”

I’m not sure how to describe what this means. Perhaps this reflects the fact that I’m not a climate scientist, but to me it seems a bit like saying that the high temperature during the summer in Columbia averages 83.0°F, while the high in winter averages 41.2°F, therefore the predicted temperature is the average of the two, 62.1°F. While your prediction has a certain methodological validity, what does it mean? Does it represent what it is going to be like in winter? No. In summer? No. If you are planning to visit, does it tell you what clothing to bring? No. If you are planning to build a house, does it tell you how much heating or air conditioning to put in? No, you need to put in enough to handle the coldest and hottest days, not the average day.

It also doesn’t tell you the most likely outcome. There is an implication that somehow the average of the three scenarios represents the most likely eventuality. But it doesn’t. Nobody knows which scenario the future will follow, that’s why studies usually report the results of several.

So, I don’t really know how to describe these results, and they won’t match those of the Hayhoe group or of Posey, as reported in the previous two posts.

Okay, with that said, here is what the Anderson group found. In Columbia, climate change is already underway. The following changes have already occurred:

  • There have been changes in daily high and low temperatures, though they vary according to season;
  • During heat waves it is staying hotter at night;
  • There have been fewer cold waves; and
  • There has been an increase in heavy rain events.

In the future:

  • Figure 1: Current Climate and Climate Change Projections for Columbia, MO. Source: Anderson et al 2015.

    Figure 1: Current Climate and Climate Change Projections for Columbia, MO. Source: Anderson et al 2015.

    Columbia’s annual average temperature is projected to increase 2.7°F by 2035 (actually a 30-year average centered on 2035), and 5.5°F by 2065 (actually a 30-year average centered on 2065) (See Figure 1 at right);

  • The temperature change is projected to be larger during spring and winter and smaller during summer and fall;
  • Annual precipitation is projected to increase by 4.2 inches, but summer precipitation is projected to decrease by half-an-inch.
  • Heat waves are expected to get hotter by 6.1°F, with both highs and lows during the heat waves expected to be warmer;
  • The number of days on which more than 1.25 inches of rain falls are not projected to increase significantly; however
  • The maximum amount of rain that falls during 5-day and 15-day periods is expected to increase by over an inch.

While the specifics of the Anderson group’s projections can’t be compared to the studies by the Hayhoe group or by Posey because of the methodological differences discussed above, it is clear that the general trend of the changes go in similar directions: warmer temperatures, worse heat waves, decreased precipitation during the summer but increased precipitation during other seasons of the year, and an increase in heavy precipitation events.

I will summarize these 3 studies and draw conclusions in the next post.

Sources:

Anderson, Christopher, Jennifer Gooden, Patrick Guinan, Mary Knapp, Gary McManus, and Martha Shulski. 2015. Climate in the Heartland: Historical Data and Future Projections for the Heartland Regional Network. Downloaded 3/15/16 from http://www.marc.org/Government/GTI/pdf/ClimateintheHeartlandReport.aspx.

Hayhoe, K, J VanDorn, V. Naik, and D. Wuebbles. 2009. “Climate Change in the Midwest: Projections of Future Temperature and Precipitation.” Technical Report on Midwest Climate Impacts for the Union of Concerned Scientists. Downloaded from http://www.ucsusa.org/global_warming/science_and_impacts/impacts/climate-change-midwest.html#.VvK-OD-UmfA.

Posey, John. 2014. “Climate Change in St. Louis: Impacts and Adaptation Options.” International Journal of Climate Change: Impacts and Responses. Vol 5, #2. Downloaded 1/15/2016 from http://ijc.cgpublisher.com/product/pub.185/prod.233.

Posey Projects Climate Change for St. Louis

In my last post I reported on climate projections for St. Louis and the Midwest by Hayhoe, VanDorn, Naik, and Wuebbles. Here, I will report on climate projections for St. Louis by John Posey, Director of Research for the East West Gateway Council of Governments. His paper focused on temperature and precipitation changes by the middle of this century in the St. Louis Region, and on the types of socio-economic impacts that would be associated with such a change. Because future climate change is expected to be sensitive to how humans respond, he made projections for a high emissions scenario (A2) and a low emissions scenario (B1). Note that while he used a high emission scenario, the scenario he used does not envision emissions as high as the A1fi scenario used by Hayhoe et al.

Please remember that there is uncertainty associated with all climate change projections, and that with projections for the local level the uncertainty is magnified.

Posey’s results are shown in Table 1. He projects a 3.6°F temperature increase by mid-century under the low emissions scenario, and 4.9°F under the high emissions scenario.

Posey Temp Change

(Click on table for larger view.)

Posey projects that St. Louis will experience a slight increase in annual average precipitation, less than 10%. The analyses he conducted were uncertain about whether summer would see a decrease. Posey also projects an increase in heavy precipitation events, though he projects the increase to be small (1 additional heavy precipitation event per year).

Posey identifies 3 main socio-economic impacts that should be expected from climate change in the St. Louis Region. He expects an increase in flooding, heat stress, and energy consumption. In addition, he expects that other challenges might include agricultural stress, problems with roads (road buckling, etc.), and changes in infections disease vectors.

Given that Posey and Hayhoe et all projected for slightly different regions, and that they used different high emissions scenarios, their results seem more or less consistent. Where they differ, they differ in ways we would expect given the differences in their methods. (Posey’s projected temperature increase under the high emissions scenario should be less than Hayoe et al’s, and it is.)

Because of the differences in their methods, you shouldn’t really compare their projections directly, but I know that you’re gonna do it anyway. So, as you do, keep in mind that Hayhoe et al’s projections for extremely hot days were for one of the higher emission scenarios available, and they were for the end of the century, while Posey’s projections used a somewhat less extreme high emissions scenario, and only extended through mid-century.

In the next post, I will report on a study that made climate projections for Columbia, Missouri.

CORRECTION: In the original version of this post, the second to last paragraph stated that the A1fi Scenario used by Hayhoe et al was the highest emission scenario available. That has been corrected to read “one of the higher emission scenarios available.” See Hayhoe’s comment to the blog post for the specifics.

SECOND CORRECTION: In Paragraph 4, the original version of this post said that summer precipitation in St. Louis would decrease. That has been changed to indicate that projected change in summer precipitation is uncertain.

Sources:

Hayhoe, K, J VanDorn, V. Naik, and D. Wuebbles. 2009. “Climate Change in the Midwest: Projections of Future Temperature and Precipitation.” Technical Report on Midwest Climate Impacts for the Union of Concerned Scientists. Downloaded from http://www.ucsusa.org/global_warming/science_and_impacts/impacts/climate-change-midwest.html#.VvK-OD-UmfA.

Posey, John. 2014. “Climate Change in St. Louis: Impacts and Adaptation Options.” International Journal of Climate Change: Impacts and Responses. Vol 5, #2. Downloaded 1/15/2016 from http://ijc.cgpublisher.com/product/pub.185/prod.233.

Climate Predictions for St. Louis: Dangerously Hot

Only in recent years have advances in climate modeling made it possible for climate scientists to start modeling climate on a regional or local level. Previously, projections could only be made for large areas, such as the United States or all of North America. Two studies have now modeled climate projections for St. Louis, and a third has modeled them for Columbia. When I discussed local climate change in my posts on the National Climate Assessment (see here), I was interpreting color gradations off of maps covering larger areas. These studies focus directly on areas in Missouri.

In the next couple of posts I’m going to report on 3 of these studies. Before I start, however, I want to note that each study uses a different method, and makes different assumptions about how people will respond to climate change. The result is that while each study provides information about how the climate will change in Missouri, their results are not directly comparable. Further, please note that all climate change projections contain some uncertainty, and with projections for local areas, that uncertainty is magnified.

Hayhoe, VanDorn, Naik, and Wuebbles (2009) studied climate change in the Midwest, including St. Louis. The effects of climate change are expected to be sensitive to how humans respond to climate change. If we curtail our emissions, they are expected to be less. If emissions remain high, they are expected to be greater. The Hayhoe group studied emissions using a low emissions scenario (B2) and a high emissions scenario (A1fi). The B2 scenario is a commonly used low emissions scenario, while the A1fi scenario is the highest emissions scenario available.

Table 1 shows the projected change in the annual average temperature (°F) for 3 time periods under the low and high emissions scenarios.

Temp Change Table

(Click on tables or charts for larger view.)

The projections in Table 1 are for the Midwest as a whole. They show that the increase is expected to grow as time passes, that the differences between the emissions scenarios increase over time, and that the increase in temperature is projected to be larger during the summer than the winter. Even under the low emissions scenario, the average annual temperature is projected to increase 5.0°F by the end of the century, and it is expected to increase by a whopping 10°F under the high emission scenario.

The Hayhoe group was interested in looking at more than just change in average annual temperature, however. The way in which climate change is most likely to directly impact Missouri is through a change in extremely hot days. In urban areas, extremely hot days can be killers. In 2003, an extended heat wave in Europe killed an estimated 70,000 people, and in Chicago, a heatwave in 1995 killed 500 people in just 4 days. What, the Hayhoe group wondered, was in store for the Midwest?

They looked at 9 Midwestern cities, including St. Louis. They found that currently St. Louis experiences 36 days over 90°F each year, and 2 days over 100°. But under the high emissions scenario, they projected that by the end of the century St. Louis would experience 105 days over 90° and 43 over 100°. I constructed charts to illustrate the change. They show the days occurring consecutively, centered on the end of July. Figure 1 shows the current situation, Figure 2 shows Hayhoe et al’s projection.

Hot Day Calendar Past

Figure 1: Current days over 90°F (orange) and over 100°F (red), 2070-2099. Data source: Hayhoe et al 2009.

Hot Day Calendar Future

Figure 2: Projected days over 90°F (orange) and 100°F (red), 2070-2099. Data source: Hayhoe et al 2009.

 

 

 

 

 

 

 

.

Under the Hayhoe projection, days over 90° begin June 10, and continue through September 22. That’s more than 3 months. Days over 100° begin July 11 and continue through August 22. That’s almost a month-and-a-half. Can you imagine 43 days with temperatures over 100°?

Figure 3: High temperature at Midway Airport, Chicago, Summer 1995. Data: Weather Underground

Figure 3: High temperature at Midway Airport, Chicago, Summer 1995. Data: Weather Underground

For comparison sake, Figure 3 shows the high temperature at Midway Airport in Chicago during the summer of 1995, the year of their heat wave. There were only 28 days where the high was above 90°, and only 2 days when the high was above 100°.

The Hayhoe group projected that climate change would bring Missouri no more than a slight increase in precipitation. They projected, however, that there would be a 22% reduction in summer precipitation, when it is most needed, and an increase during the remainder of the year. Further, they projected that more precipitation would occur during heavy precipitation events (more than 2 inches per day), with longer, hotter dry periods between. There have been two recent examples of heavy precipitation events: in late December, up to 8 inches of rain fell over 2-1/2 days in Missouri. In March, up to 20 inches of rain fell over 4 days in Louisiana. Both resulted in severe record flooding.

There are many ways in which climate change might bring indirect effects to Missouri. Hot, humid summers might cause a decrease in air quality, leading to an increase in asthma and other respiratory diseases. Hotter summers and warmer, wetter winters might lead to an increase in disease vectors, such as ticks and mosquitoes. Warmer temperatures may threaten water quality during parts of the year, etc. The list could go on.

The Hayhoe et al projections are the most dire of the 3 I will review, perhaps because they used the A1fi Scenario, the highest emission scenario. In the next post, I will report on climate projections for the St. Louis Region by local researcher John Posey.

Sources:

Hayhoe, K, J VanDorn, V. Naik, and D. Wuebbles. 2009. “Climate Change in the Midwest: Projections of Future Temperature and Precipitation.” Technical Report on Midwest Climate Impacts for the Union of Concerned Scientists. Downloaded from http://www.ucsusa.org/global_warming/science_and_impacts/impacts/climate-change-midwest.html#.VvK-OD-UmfA.

Weather Underground. 2016. Chicago, IL. Data downloaded 3/8/16 from https://www.wunderground.com/us/il/chicago.

Missouri Weather Disasters 2015

Figure 1 source: Office of Climate, Water, and Weather Services, 2016.

Figure 1 source: Office of Climate, Water, and Weather Services, 2016.

Damage from sever weather in Missouri shows a different pattern than does damage nationwide. As Figure 1 shows, the cost of damage from hazardous weather events in Missouri spiked in 2007, then really spiked in 2011. Since then, it has returned to a relatively low level. I haven’t added any trend lines to the chart because they would not well describe the shape of the curve.

The bulk of the damage in 2011 was from 2 tornado outbreaks. One hit the St. Louis area, damaging Lamber Field. The second devastated Joplin, killing 158, injuring 1,150, and causing damage estimated at $2.8 billion.

.

.

.

Figure 2 data source: Office of Climate, Water, and Weather Services 2016.

Figure 2 data source: Office of Climate, Water, and Weather Services 2016.

Figure 2 shows deaths and injuries in Missouri from hazardous weather. Deaths are in blue and should be read on the left vertical axis. Injuries are in red and should be read on the right vertical axis. The large number of injuries and deaths in 2011 were primarily from the Joplin tornado. In 2006 and 2007, injuries spiked, but fatalities did not. The injuries mostly represented non-fatal auto accidents from winter ice storms. The fatalities in 1999 resulted from a tornado outbreak.

The Missouri data covers fewer years than the national data discussed in my previous post. It also covers all hazardous weather, in contrast to the national data, which covered billion dollar weather disasters.

While the national data shows a clear trend towards more big weather disasters, Missouri’s data does not.

Sources:

Office of Climate, Water, and Weather Services, National Weather Service. 2016. Natural Hazard Statistics. Data downloaded 2/10/16 from http://www.nws.noaa.gov/om/hazstats.shtml#.

InflationData.com. 2016. Historical Consumer Price Index (CPI-U) Data. Data downloaded 2/10/16 from http://inflationdata.com/Inflation/Consumer_Price_Index/HistoricalCPI.aspx?reloaded=true.

In addition, descriptions of specific weather events, if they are large and significant, can be found on the websites of the Federal Emergency Management Administration, the Missouri State Emergency Management Agency, and local weather forecast offices. However, in my experience, the best descriptions are often on Wikipedia.

Billion Dollar Weather Disasters 2015

Source: National Centers for Environmental Information 2016.

Source: National Centers for Environmental Information 2016.

Bad weather causes damage. The National Centers for Environmental Information keeps track of weather events that cause over $1 billion in damage. According to the most recent data, such events have increased dramatically over the last 35 years nationwide. The columns in Figure 1 shows the number of billion dollar disasters per year. From an average of 1.8 in 1980-1984, the number has increased to 11.25 for the years 2012-2015. The number peaked in 2011 at 16.

The colors show the kind of disaster. While there used to be roughly equal numbers of several different kinds of disasters, in recent years the number of severe storms has increased (shown in green). Over the last 4 years, 51% of all billion dollar disasters have been caused by severe storms. While one immediately thinks of hurricanes and tropical storms (shown in yellow on the chart), the last 4 years have actually been years of low hurricane activity. The billion dollar disasters are being caused by tornadoes, thunderstorms, and other kinds of severe weather.

The black line in Figure 1 shows the 5-year mean of the dollar value of the damage caused. In the mid-1980s, the annual damages were under $10 billion. Damages peaked at about $70 billion in 2008, and have been running between $50 and $60 billion in recent years. Cost assessments for 2015 have not yet been completed.

The disasters that cause the most dollar damage are those that affect a wide area. The most damaging of all were Hurricane Katrina ($152.5 billion) and Hurricane Sandy ($67.6 billion), mostly through flooding from storm surge.

Why have the number of billion dollar disasters increased so dramatically? There are two principal reasons. One is climate change. Heat is a form of energy. Warm the atmosphere, and you increase the amount of energy in it. Put more energy in the atmosphere, and there is more energy to fuel severe weather. One of the primary predictions of climate change is that the world will experience an increase in severe weather events, and that seems to be happening, in the United States at least.

Another reason is development. Over time, with our expanding population, more and more of the country has been developed, and more of the land has been covered with buildings and infrastructure. When severe weather strikes, it is more likely to hit developed land, causing more damage. Not only that, but we have put ourselves in harms way by moving into vulnerable areas. For example, in St. Louis County, millions of dollars of development has occurred in the area formerly known as Gumbo Flats, now called Chesterfield Valley. Most of the development happened after the Great Flood of 1993. If a flood were to occur in that area now, it would cause much greater damage than in 1993.

The vulnerable areas that have seen development include the seashore, river valleys, forests (fire hazard), and earthquake zones.

Inflation is probably not the major cause of the increase. The NCEI cautions that it is difficult to estimate the current cost of events that happened years ago. However, they have tried, and have adjusted for inflation. Their estimates may not be perfect, but I doubt that inflation is the major cause of the increase.

The next post will look at similar data for Missouri.

Source:

National Centers for Environmental Information. 2016. Billion-Dollar Weather and Climate Disasters: Summary Stats. Downloaded 2/9/16 from https://www.ncdc.noaa.gov/billions/summary-stats.

California Water Update

It is mid-February, and I promised that I would catch up with the water conditions in California. Over the summer of 2015 I ran a 13-post series on the drought in California, in which I attempted to estimate whether California would be able to cover its future water deficit, and if not, what it would do to California’s economy.

I concluded that California faced not only a severe current drought, but a future in which they would lose a significant fraction of their water supply, resulting in a severe water deficit. Various remedies would cover only part of the deficit. The combined consequences of the shortage and the remedies would throw the state into a recession, possibly even a depression. The series of posts starts here.

I’m neither a hydrologist nor an economist, and I learned a lot as I wrote the series. I still know of no other analysis that attempts such a comprehensive assessment of California’s water future.

About the time I finished the series, signs were increasing that a large El Niño event was starting. Large El Niños typically bring above average precipitation to California. The most important form of precipitation for California is the snowpack in the Sierra Nevada and Cascades. This forms a “reservoir” that melts slowly during the spring and summer, providing water during months when much of California receives virtually no precipitation at all.

The water content of the snowpack on April 1 is the most important, but measurements at the end of January can give some indication of how things are going. California measures two ways. First, they conduct “media oriented” events in which they physically go to specific locations and measure the water content of the snowpack at that location. Last April 1 when they did this, there was no snow on the ground at all. This year, on February 2 there was a snow water equivalent of 25.4 inches, which is 130% of the average for that date.

Source: California Data Exchange Center 2016.

Source: California Data Exchange Center 2016.

A second way they measure is via continuous electronic measurements at a couple of dozen locations around the state. Figure 1 at right shows the results for this year through February 12. The charts show the water content of the snowpack as percentages of the average amount on April 1. The top chart is for Northern California, the middle one is for Central California, and the bottom one is for Southern California. The blue line represents this year. The green line represents 1982-1983, the year with the biggest snowpack. The brown line shows 2014-2015, the year with the lowest snowpack on record. The light blue area represents average.

You can see that the average snow water content builds during the winter and reaches its maximum right around April 1 – that’s one reason the April 1 measurement is the most important. This year, in all 3 regions, the snowpack has built towards its April 1 average much better than it did last year. However, the data do not suggest that, statewide, it is an exceptional snow year. In this data, it looks like a pretty average snow year so far.

Somewhat worrying is the fact that since about February 1, the line for this year has leveled off. Might the snow season peter out and end up below average? I hope not.

Source: Mammoth Mountain Resort 2016.

Source: Mammoth Mountain Resort 2016.

Finally, I located one other source of data, this one more hopeful: the snow report at Mammoth Mountain, a large ski resort in the Central Sierra Nevadas. The resort tracks snowfall at the resort by year and by months, and it is available on the Web. Figure 2 shows total snowfall by year, and the different colors in the columns represent the amount for each month. The data suggest that at Mammoth Mountain it has been an above average snow year, 11th highest out of 47 years. February is, on average, the biggest snow month, however, and data for the first half of the month this year show the same tapering-off that the other data did (not shown on the chart).

I’m not sure how to reconcile these three contrasting data sources. Could above average snowfall have occurred, but above average temperatures melted more than usual, leaving about an average amount on the ground? Possibly, as January was 2.4°F warmer than average in California. That explanation would not, however, reconcile the difference between the electronic measurements and the media oriented measurement.

Now, one final word of caution: none of this changes my projection for California’s water future. It is surely good news that they are not having as dry a year as last year. However, my analysis was based on projected 30-year average precipitation levels. During any 30-year period, there are bound to be wetter years and drier years. If my analysis is correct, then while it is good news for right now, the real issue will be the average size of the snowpack over many years.

So, for the future, we’ll have to wait and see. For right now, it looks like a better snowpack year than last year: at least average, and perhaps above average.

In my next post, I will look at an analysis of water scarcity around the globe.

Sources:

California Data Exchange Center. 2016. California Statewide Water Conditions: Current Year Regional Snow Sensor Water Content Chart (PDF). Downloaded 2/13/16 from http://cdec.water.ca.gov/water_cond.html.

National Centers for Environmental Information. 2016. Climate at a Glance. Data retrieved 2/13/16 from http://www.ncdc.noaa.gov/cag/time-series/us.

Mammoth Mountain Resort. 2016. Extended Snow History. Data downloaded from http://www.mammothmountain.com/winter/mountain-information/mountain-information/snow-conditions-and-weather.

%d bloggers like this: