Skip Navigation

November-December 2014

ResizeResize Text: Original Large XLarge

Weather Queries

It's July 4, 2014, a time of year when the weather here in Ann Arbor, Michigan, can vary, but tends toward warm and humid. As I write this, though, while Hurricane Arthur is moving up the East Coast, the weather here in the Great Lakes region happens to be sunny and pleasant, on the cool and dry side. It's something I've noticed in the past—a hurricane affecting the Gulf or Atlantic states, just when conditions here happen to be at their best. Is this just a random coincidence, something I'm imagining, or some kind of fair-and-foul distribution that weather patterns can explain?

Robert Ahronheim

Ann Arbor, Michigan

Most of the hurricane season is included within the months of July through October, with the peak coming in mid-September. You have the impression that, when hurricanes threaten the East or Gulf Coasts, weather in Ann Arbor is usually pleasant. The climatology of southeast Michigan supports this impression in September and October. The average daily low and high temperatures in September are 53°F and 74°F, respectively; in October, 42°F and 61°F. About nine days have measurable precipitation in both months. So it is pleasant most of the time, regardless of hurricane activity elsewhere.

In July and August, however, the weather can be warm and humid. In Ann Arbor, the average low and high temperatures in July are 62°F and 83°F, respectively; in August, 61°F and 81°F. Record high temperatures range from the mid-90s to the very low 100s. Sixteen hurricanes made landfall in July or August since 1990 (see Table 1).

Table 1.

Landfalling United States Hurricanes in July or August Since 1990

NameDate of landfallSaffir-Simpson categoryStates affected
BobAugust 19, 19912Rhode Island, Massachusetts, New York, Connecticut
AndrewAugust 24, 19925Florida, Louisiana
ErinAugust 2, 19952Florida
BerthaJuly 12, 19962North Carolina
DannyJuly 19, 19971Louisiana, Alabama
BonnieAugust 26, 19982North Carolina
BretAugust 22, 19993Texas
ClaudetteJuly 15, 20031Texas
CharleyAugust 13, 20044Florida, South Carolina, North Carolina
GastonAugust 29, 20041South Carolina
CindyJuly 5, 20051Louisiana
DennisJuly 10, 20053Florida, Alabama
KatrinaAugust 29, 20053Florida, Louisiana, Mississippi, Alabama
DollyJuly 23, 20081Texas
IsaacAugust 28, 20121Louisiana
ArthurJuly 4, 20142North Carolina


For each of these dates, I examined the weather in Michigan, and Detroit in particular (since it is close to Ann Arbor and daily records are more readily available). I consulted several sources: prior to 1999, historical daily weather maps, including surface and 500-mb charts, maximum and minimum temperatures, and precipitation at; for 1999 and later, surface and upper air maps at 0000 and 1200 GMT from the Storm Prediction Center at; for 2002 and later, daily temperature and precipitation data from the NWS office in Detroit at From these sources, I extracted the data shown in Table 2.

Table 2.

Weather Conditions at Detroit, Michigan, and the Surrounding Region on Days When Hurricanes Struck the United States Mainland in the Months of July and August, Back to 1990

DateDetroit weatherRegional weatherUpper air conditions
 Min/max temp. (°F)ΔT (°F)Dew point (°F)Precipitation (in)
August 19, 199163°-69°−6°Upper 50s1.91Stationary front from New Hampshire to southeast Texas; high pressure from Quebec to MinnesotaDeep trough from Quebec to west Tennessee with closed low over Lake Michigan. This low tracked across southeast Michigan
August 24, 199269°-82°+5°Upper 60s.01Muggy; high pressure ridge from Massachusetts to VirginiaAnticyclone centered on Virginia coast; weak southwest flow aloft Texas to Michigan
August 2, 199568°-84°+3°68°-72°.78Muggy; stationary front along southern Michigan border, surface high over QuebecBermuda high from western Atlantic to the Carolinas; weak westerly flow over Michigan
July 12, 199660°-83°−2°Low 60s0Surface low northwest Wisconsin; weak southerly winds over MichiganBroad upper trough, upper Midwest
July 19, 199764°-79°−3°55°-59°0Surface high over Ontario; 10-knot northerlies over MichiganMajor trough aloft from Maine to Maryland; anticyclone over central Great Plains
August 26, 199860°-81°−1°Near 60°0Surface high centered over southern WisconsinWesterly flow over Michigan
August 22, 199960°-80°−2°Low 60s0Surface high over Lake Erie, moving slowly eastBroad upper tropospheric trough along East Coast; weak anticyclone over east Texas
July 15, 200366°-81°62°-68°.15Weak southerly winds over MichiganVigorous trough passed over Michigan during the day; northeast-southwest trough at 250 mb from east of NJ to LA
August 13, 200456°-70°−9°Mid 50s0Weak northerly flow over MichiganDeep trough centered over Michigan
August 29, 200461°-69°−5°Low 60s.30Surface trough just east of Michigan, moving east during the day; light windsBroad trough approaching Mississippi River; 80-knot jet over Michigan
July 5, 200568°-83° on July 6: 63°-75°+3°61°-65°.82Surface high north of Lake Superior; weak pressure gradient, light winds over MichiganBroad tilted trough from Quebec southwest to Oklahoma
July 10, 200564°-88°+3°50s0Broad surface high, Michigan to Delaware; light surface windsAt 500 mb, ridge line from Wisconsin to South Carolina; in high troposphere, trough from eastern Oklahoma to Gulf off southeast Texas coast
August 29, 200565°-86°+6°Near 60°0Very weak surface pressure gradient; light windsTilted upper trough from south of Hudson Bay to eastern Oklahoma
July 23, 200861°-81°−3°Mid 50s0Surface high over Wisconsin and Michigan, light surface windDeep, north-south trough from Lake Huron to northeast Alabama
August 28, 201263°-82°+3°Near 60°0Surface high over Michigan; light surface windsTrough oriented NE-SW from western New York to Arkansas
July 4, 201455°-77°−7°Upper 40s0Surface high over Michigan; light surface windsFairly deep upper trough Lake Huronto Arkansas


In the table, ΔT is the departure of temperature from normal for the given date. Standard two-letter abbreviations for states are used, as are standard abbreviations for directions. Dates when hurricanes struck the East Coast north of Florida are shaded in gray. On all other dates, hurricanes struck Florida or the Gulf Coast.

This is admittedly a small sample of storms (16), partly because of the labor in examining weather conditions on each day. Nonetheless, one trend stands out. On all six days when storms struck the East Coast, weather in southeast Michigan was cooler than normal. In fact, four of the coolest days (relative to normal) in the entire sample of 16 were in this grouping. On five of the six days, surface high pressure was centered either over Michigan or lay to the north through west. This resulted in dew points no higher than the low 60s, which residents welcome in July and August. Finally, there is no hint of anticylcones in the upper troposphere east of the Mississippi River when hurricanes strike the East Coast. The reason is clear: anticyclones in this location would prevent a curving track into the East Coast and instead steer the hurricane westward toward the Gulf of Mexico.

Precipitation occurred on six of the 16 days. Climatology would predict four and a half. There is no evidence that precipitation is more frequent when hurricanes strike the East Coast than when they strike the Gulf Coast.

All days with positive temperature anomalies occurred when hurricanes struck Florida or Gulf Coast states, though three of the 10 days had small negative anomalies. On five of these 10 days, anticyclones or ridges aloft were present north of the hurricane track, but conditions aloft over Michigan were quite variable.

In summary, I'd say there's weak statistical evidence, backed up by regional weather patterns, that when hurricanes strike the East Coast of the United States in July and August, weather conditions in southeast Michigan tend to be cooler than normal and perhaps a little less humid.

Natural gas in a rock formation is under high pressure of 1,000 to 20,000 pounds per square inch. Some of it escapes to the atmosphere during drilling, fracking, extraction, and transmission. Once combusted, it releases carbon dioxide into the atmosphere. Doesn't this increase the total mass of the atmosphere appreciably and affect climate? Thawing permafrost and methane hydrates on the ocean floor also have to be considered.

John Mulhall

Clinton Township, Michigan

The short answer to the first part of the question is that the addition of carbon dioxide (CO2), methane (CH4), the main component of natural gas, and other greenhouse gases do increase the total mass of the atmosphere. However, the concentrations of these gases are so small compared with those of oxygen, nitrogen, and water vapor that their contributions to total atmospheric mass are not measurable as an increase in barometric pressure. Table 3 lists some of the most abundant atmospheric gases in dry air.

Table 3.

Relative Abundance of Gases in a Dry Atmosphere Expressed as Parts per Million (ppm)

GasChemical symbolConcentration (ppm)Rank
Carbon dioxideCO2~ 4004


Parts per million (ppm) is equivalent to the average number of molecules of a specific gas for every million molecules of air, including all gases except water vapor. Alert readers will notice that the numbers in the third column add to slightly more than 1,000,000. Ideally, the concentrations for all atmospheric gases (not just those listed here) would sum to 1,000,000, but these numbers have not been adjusted for gases with changing concentrations, such as carbon dioxide, whose concentration is steadily increasing.

Even carbon dioxide, whose concentration is more than 200 times that of methane, is not abundant enough to contribute appreciably to total atmospheric mass. The total mass of the atmosphere is estimated to be about 5.15 × 1018 kg. The mass of CO2 is about 3.2 × 1015 kg, which is more than three orders of magnitude less. During the 19th century, the mass of CO2 was about 2.2 × 1015 kg. The mass of CH4 is about 5.2 × 1012 kg, which is about 600 times less than that of CO2. It is little wonder, then, that changes in concentrations of CO2 and CH4 are not detectable in terms of surface atmospheric pressure.

Table 3 does not include water vapor, whose concentration varies greatly throughout the atmosphere and is very strongly limited by temperature. It ranges from less than 100 ppm at temperatures below -40°F to more than 20,000 ppm in humid air at 80°F and above. What matters, though, is water vapor's contribution toward total atmospheric mass. It is appreciable, about 1.27 × 1016 kg, which is four times the mass of CO2.

In a 2005 paper in Journal of Climate, Kevin Trenberth and Lesley Smith documented an annual cycle in total atmospheric mass, due mostly to an annual cycle in water vapor mass. This, in turn, is associated with the tropical and subtropical monsoons. The cycle is exhibited in the figure on the next page as periodic changes in surface pressure due only to changes in atmospheric water vapor. The oscillations cover a pressure range of a few tenths of a millibar (the same as a hectoPascal, hPa, the units used in the figure), which is enough to measure with a sensitive barometer, but, in practice, is very difficult because of all the day-to-day pressure variations due to other causes.

The contribution of atmospheric water vapor to global mean surface pressure. Note the annual oscillation, measured in hectoPascals (hPa), from before 1950 to 2000. The contribution of water vapor to global mean surface pressure was calculated from a reanalysis of all available weather observations during the indicated years with the help of three computer models, incorporating modern methods of data assimilation.

The contribution of atmospheric water vapor to global mean surface pressure. Note the annual oscillation, measured in hectoPascals (hPa), from before 1950 to 2000. The contribution of water vapor to global mean surface pressure was calculated from a reanalysis of all available weather observations during the indicated years with the help of three computer models, incorporating modern methods of data assimilation.

Does the injection of methane and carbon dioxide into the atmosphere as a result of human activity change the climate? Almost certainly. In support of this, I offer some findings in the recently released monumental report (1,535 pages) of the Intergovernmental Panel on Climate Change (IPCC): Climate Change 2013: The Physical Science Basis. The report states, “With a very high confidence, the increase in CO2 emissions from fossil fuel burning and those arising from land use change are the dominant cause of the observed increase in atmospheric CO2 concentration.” Fossil fuel burning (coal and natural gas) consumes oxygen and generates carbon dioxide. A slight decrease in atmospheric oxygen has been observed over the past two decades, more in the Northern Hemisphere, where there is more industry, than in the Southern Hemisphere. With respect to CH4, again from the report: “Total anthropogenic sources contribute at present between 50 and 65% of the total methane sources.”

The concentrations of CO2 and CH4 in the atmosphere are important for the earth's climate because it is well understood how they, as greenhouse gases, absorb heat radiation emitted from the earth's surface that would otherwise go to space. Energy from the sun arrives at wavelengths in the visible spectrum. Some is reflected back to space by clouds, but much of it is absorbed at the earth's surface, causing heating. The earth continually radiates energy at wavelengths in the infrared part of the spectrum. This energy is effectively absorbed by clouds and reradiated downward. (That's why nighttime clouds tend to elevate the minimum overnight temperature.) But, even in the absence of clouds, water vapor, carbon dioxide, and methane also absorb outgoing infrared radiation and reradiate it in all directions, including downward. Other greenhouse gases do the same, but H2O, CO2, and CH4 are the most important three. Greenhouse gases strongly regulate the average temperature of the earth-atmosphere system. With little doubt, the increasing concentration of these gases is raising the system's temperature.

Molecule for molecule, methane is about 10 times more effective than carbon dioxide in causing global warming, but it is important to remember that the latter is more than 200 times as abundant as the former. There is concern that thawing permafrost and methane hydrates on the ocean floor will be a major future source of methane on a warming planet, but carbon dioxide is currently a much greater concern, in part because of its much higher concentration, but also because of its residence time in the atmosphere. The IPCC report estimates that 15-40% of CO2, emitted between now and 2100, will remain in the atmosphere longer than 1,000 years. On the other hand, the average residence time of CH4 in the atmosphere is only about nine years.

I thank Pieter Tans of NOAA's Earth System Research Laboratory for help in answering this question.

Weatherwise Contributing Editor THOMAS SCHLATTER is a retired meteorologist and volunteer at NOAA's Earth System Research Laboratory in Boulder, Colorado. Submit queries to the author at, or by mail in care of Weatherwise, Taylor & Francis, 530 Walnut St., Suite 850, Philadelphia, PA 19106.

In this Issue

On this Topic

© 2017 Taylor & Francis Group · 530 Walnut Street, Suite 850, Philadelphia, PA · 19106