Home » Air (Page 3)
Category Archives: Air
In February, 2013, I published a post on the Toxic Release Inventory with data through 2011. This post updates the information for 2012, the most recent year available.
Many industrial processes use or produce toxic substances. These substances must be properly handled to prevent harm to people, land, and water. After a series of disasters in the 1970s and 1980s, Congress passed the Emergency Planning and Community Right-to-Know Act in 1986, and the Pollution Prevention Act in 1990. These laws require facilities to report releases, transfers, and waste management of toxic material.
The Toxics Release Inventory (TRI) of the EPA gathers this information and makes it available to the public on their website. In addition, they publish annual fact sheets and analyses. The TRI data does not cover all toxic materials or facilities, but it does cover an important set of them. New for 2012, releases of hydrogen sulfide, a poisonous gas that smells like rotten eggs, have been included in the data.
Toxic substances are recycled, burned to generate electricity, injected into wells, stored, landfilled, emitted into the air, discharged into surface water, and spread over the land. They can be handled either on-site or off-site. Determining whether any of these activities represents a potential hazard to people, land, or water is complex. One cannot simply assume that on-site means safe, or that emission or discharge means that there is toxic exposure. The statistics in the TRI are only a starting point.
In the graph at right, the blue line shows total releases and disposal of toxic materials in Missouri from 2002-2012. In 2007, the EPA changed the minimum reporting threshold from 2,000 to 5,000 pounds, suggesting that comparisons between years may not be valid. Over the entire time period, toxic releases and disposal decreased by 4%. While there does not appear to be a significant discontinuity in the data following 2007, the data shows a rapid increase until 2005, followed by a rapid decrease. The percentage differences are very large: between 2002 and 2004 the increase was 56% From 2004-2012, the decrease was 39%. Such large , rapid changes may, indeed, reflect changes in data gathering rather than actual changes in the amount of toxic materials. A report tracking the changes and making an “apples-to-apples” comparison across years would be most welcome. If you know of one, let me know about it by posting a comment.
Interestingly, Missouri releases declined 2% between 2011 and 2012, despite the addition of sulfur dioxide to the list. Some other data suggests that Missouri may lag the nation by a couple of years in implementing changes, so it will be interesting to follow this trend in upcoming years.
My next post will explore the industries and the chemicals that account for the most toxic releases.
State Fact Sheet, TRI Explorer, 2012. http://iaspub.epa.gov/triexplorer/tri_broker_statefs.broker?p_view=STCO&SFS=YES&trilib=TRIQ1&state=MO&year=2012.
Manufacturing Employment in Missouri, FRED Economic Data, Federal Reserve Bank of St Louis, http://research.stlouisfed.org/fred2/graph/?s[id]=MOMFGN.
The preceding 4 posts have explored background air pollution. It is one thing to measure the quality of air in and around our major metropolitan areas and major pollution sources. That kind of air quality has clearly improved markedly over the decades. It is another thing to ask what the background level of pollution is. That you must measure in isolated areas, far away from large cities and pollution sources.
In Background Sulfur Dioxide and Nitric Acid Pollution Improves I reported that the background level of those two pollutants has markedly improved in the eastern part of the United States.
In The Worst Is Better, but the Best Is Worse I reported that at a monitoring site in Adair County, Oklahoma, the ozone level has increased slightly on the best air quality days, and it has decreased slightly on the worst air quality days.
In The Best is Better, but the Worst is Worse, I reported that the level of particulate matter at Bryce Canyon National Park has decreased slightly on the best air days, but it has increased significantly on the worst air days.
In Ozone on the Rise in the Mid-Troposphere, I reported that the background level of ozone at 10-16,000 ft. above sea level was rising in Hawaii and in the western United States, as a result from ozone precursors emitted in Asia.
Combined, these posts suggest that there is no single, simple answer about the background quality of air. In some ways it seems to have improved nationally, in others it seems to have deteriorated. In Missouri, the overall level of sulfur dioxide and nitric acid seems to have decreased. However, some of the measurements that show increased background pollution have not been made in Missouri, and we don’t know what they would show here. The data also does not address trends in peak and low values. The paper by Jack Fishman that I mentioned in a previous post may shed some light. I’ll report on it when it gets published.
Ozone is on the rise in the mid-troposphere (10-16,000 ft.), and the source traces back to China and India. This contrasts with the air quality index, of which ozone is a principal component, which has improved around most major metropolitan areas.
Ozone is a complex subject. High in the stratosphere it is good because it absorbs ultraviolet radiation. A decline in stratospheric ozone became a concern several years ago (the ozone hole). But at ground level ozone is harmful. It is a powerful oxidizer that is harmful to most living matter.
Ozone in the lower atmosphere is thought to be quickly decomposed. Thus, any that exists is generally thought to be created freshly from a mixture of sunlight and ozone precursor gases (nitrogen oxides and others). Thus, the ozone level is higher during the summer and during the day (more sunshine), lower during the winter and during the night. Cars and power plants emit lots of ozone precursors, which is why ozone is of concern around major metropolitan areas.
The graphic at right shows the level of ozone measured at Mauna Loa and Hilo, Hawaii. Mauna Loa is 11,145 ft. above sea level, Hilo is at sea level.
(Click on graph for larger view.)
First, notice that there is a yearly peak and a yearly trough, representing the difference between summer and winter. Second, notice that the ozone level on Mauna Loa increased over the time period, but the level at Hilo did not.
This finding parallels the results from another study that used an airplane to repeatedly measure the ozone level over the western United States. The second chart at right shows the results. On this complicated chart, the yellow marks represent the median yearly measurement, and the yellow line represents their trend. The green marks show the mean of the yearly measurements, and the red and blue lines show the 67% and 95% confidence intervals.
(Click on graph for larger view.)
Let’s stay with just the median of the yearly measurements (yellow). It has climbed significantly, from below 50 ppb. to over 60 ppb.
These measurements were taken 10-16,000 feet above the earth, similar to the height of Mauna Loa. Interestingly, the study failed to find a similar increase in ozone at lower levels, mirroring the results in Hilo.
The authors concluded that the ozone over Hawaii and the western United States is being formed from a mixture of local sunlight and precursor gases that originate in India and China. In the third graphic at right, the maps on the left shows a plume of increased ozone readings that originates in Asia, and which extends completely across the Pacific Ocean.
(Click on map for larger view.)
Thanks to Dr. Jack Fishman of St. Louis University who got me started on background air pollution. He has a paper in review that compares the ozone level of air entering and exiting the St. Louis region, and the impact of ozone on plants in this area. I hope to report on it once it has been accepted for publication.
The graphics above come from a PowerPoint presentation: Cooper, Owen. (2013). Global Surface Ozone Trends, a Synthesis of Recently Published Findings. Given at the NOAA GMD Global Monitoring Annual Conference, May 21-22, Boulder, CO. http://www.researchgate.net/publication/41103085_Increasing_springtime_ozone_mixing_ratios_in_the_free_troposphere_over_western_North_America/file/32bfe50e5aeebd8bce.pdf.
The article detailing the research over the wester United States is: Cooper, O.R., D.D. Parrish, A. Stohl, M. Trainer, P. Nedelec, V. Thouret, J.P. Cammas, S.J. Oltmans, B.J. Johnson, D. Tarasick, T. Leblanc, I.S. McDermid, D. Jaffe, R. Jaffe, R. Gao, J. Stith, T. Ryerson, K. Aikin, T. Campos, A. Weinheimer, & M.A. Avery. (2010). Increasing Springtime Ozone Mixing Ratios in the Free Troposphere Over Western North America. Nature, 463,(21), 334-348. Nature, charges a fee for access to its articles. I was, however, able to download a free copy of the article from Research Gate: http://www.researchgate.net/publication/41103085_Increasing_springtime_ozone_mixing_ratios_in_the_free_troposphere_over_western_North_America/file/32bfe50e5aeebd8bce.pdf.
I began this series of posts with photographs taken in Bryce Canyon National Park on a clear day and a hazy day, shown again at right. Given that the data I had reviewed previously suggested that air quality had markedly improved, I asked what could account for such haze.
(Click on photo for larger view.)
The previous several posts provide part of the answer. Fortunately, air quality data is also collected at Bryce Canyon, so we can also look at pollutant levels right there. Bryce Canyon is 80 miles from the nearest city (St. George, UT), so pollution there truly represent background levels.
The site at Bryce Canyon focuses on small particulate matter – tiny particles that float freely in the air. They are too small to be seen individually with the naked eye, but collectively they cause haze. They also get into your lungs when you breathe, where they cause lung disease and other problems. The smallest ones (PM2.5) get most deeply into your lungs and are the greatest health hazard.
I downloaded PM2.5 data from the Bryce Canyon IMPROVE Site. For each year, I selected the 10 highest readings and I averaged them. Then I selected the 10 lowest readings and I averaged them. The results are shown in the graph at right. The blue line represents the high readings, the red line the low readings. The black lines show the trends.
(Click on chart for larger view.)
First, notice that in 1983 the bad days already had 5 times as much particulate matter in the air as the good days.
In addition, however, the level of particulates on good days seems to be trending slightly down. But on bad days, it is trending up. That is the opposite of what we saw with ozone at the Cherokee Nation Site (previous post). The chart covers a longer period (25 years vs. 10), but the slope of the trend line is significantly more steep. In 2009, the level of particulate matter on bad days was approximately 20 times the level on low days.
This would seem to answer the question posed by the two pictures. On bad days, particulate pollution at Bryce Canyon is much worse, and it is causing the haze.
IMPROVE Aerosol RHR (New Equation) Dataset, Database Query Wizard, Federal Land Manager Database, Interagency Monitoring of Protected Visual Environments (IMPROVE). http://views.cira.colostate.edu/web/DataWizard.
Last April I reported data that suggested the air quality in many Missouri counties had markedly improved since the 1930s. In my previous post, I reported that the background level of two important pollutants (sulfur dioxide and nitric acid) had significantly improved over the eastern part of the United States, including Missouri.
The data covered in all of these posts used summary data covering a whole year. Summary data like that smooths out day-to-day variation. If, for instance, air pollution on the worst days was getting worse, but on the best days it was getting better, summary data would show no change.
This post looks at just this question: what are the background pollution trends for selected pollutants on the worst days, and on the best days? As discussed previously, to study the background level of air pollution, you can use sensing stations located in rural areas, far from big cities and far from major sources of pollution. The CASTNET system does just that. There are no CASTNET sensors in Missouri, but the Cherokee Nation Site is in Adair County, Oklahoma, not too far from the Missouri border. (If you like to play with Google Earth, you can find it at 35.7507N -94.6700W.)
I downloaded the ozone data from the Cherokee Nation Site. Ozone peaks during the day, so the data showed the daily maximum ozone concentration. The data covered the period of time from 1/1/2003 to 12/31/2012. For each year, I first calculated an average of all the readings. Then I selected the 10 highest readings and averaged them. Then I selected the 10 lowest readings and averaged them. The first chart at right shows the results. The blue line is the average of the 10 highest readings for each year, the red line is the average of all readings for each year, and the green line is the average of the 10 lowest readings for each year. The dotted lines show the trend for each.
(Click on chart for larger view.)
It is clear that the background level of ozone is not changing much, at least not at the Cherokee Nation Site. The trend for the highest days seems to be declining slightly. The trend for the overall average is also declining, though even less. And the trend for the lowest days seems to show slightly increasing ozone pollution.
I didn’t perform a statistical analysis on this data. I’m not sure the changes would be statistically significant, they are very small. Further, an article supplied to me by Dr. Jack Fishman at St. Louis University makes it clear that many factors influence ozone levels. However, there is a very slight trend: the worst days were less polluted, but the best days were more polluted.
The next 2 posts will look at more background pollution data, and one of them will draw very different conclusions.
“Daily Ozone 8-Hour Maximum Reading, Cherokee Nation Site, Oklahoma.” EPA Home Page » AirData » Download Data » Download Daily Data. http://www.epa.gov/airdata/ad_data_daily.html.
Fishman, Jack, John Creilson, Peter Parker, Elizabeth Ainsworth, Geoffrey Vining, John Szarka, Fitzgerald Booker, Xiaojing Xu. (2010). And Investigation of Widespread Ozone Damage to the Soybean Crop in the Upper Midwest Determined from Ground-Based and Satellite Measurements. Atmospheric Environment, 44, 2248-2256.
The program to reduce the air pollution that causes acid rain has been one of the most successful in our nation’s history.
As described in the previous post, CASTNET is a network of 81 sensing stations located in rural areas, far from cities and major sources of air pollution. They are meant to capture the background level of pollution, unaffected by any local source.
Two of the principal causes of acid rain are sulfur dioxide and nitrogen dioxide. When these gases are emitted by power plants and vehicles into the atmosphere, they mix with water vapor already present in the air to form sulfuric acid and nitric acid. Even in this diluted form, these powerful acids fall with the rain, killing plants and dissolving metals, stonework, and concrete. Forests are affected, of course, but in addition, billions of dollars of damage has been done to buildings, bridges, and roads.
The first set of maps on the right shows the background concentration of sulfur dioxide over 3 periods: 1989-1991, 1999-2001, and 2009-2011. The second set of maps shows the background nitric acid concentration over the same time periods. The scale runs from green, which represents less of the pollutant, to dark red, which represents the most.
(Click on maps for larger view.)
First, notice that the white space on the maps disappears over time, This represents the development of the CASTNET system to cover the whole country.
Second, notice that in 1989-1991, the area of high pollution concentration extended from roughly Missouri to the east and northeast portions of the country. The prevailing wind blows west-to-east, blowing pollution from the Midwest towards the East.
Third, notice that over time the amount of red, orange, and yellow has decreased. In fact, by 2009-2011 areas of red and orange had disappeared entirely from the maps.
The maps are sufficiently complete in all 3 time periods to infer change from the Missouri-Kansas border eastward. The background level of sulfur dioxide and nitric acid appear to have improved significantly in Missouri. However, west of the Missouri-Kansas border, the maps are not sufficiently complete in all time periods to infer change. I included a photo of haze at Bryce Canyon National Park in the previous post. These maps don’t provide much information about the air quality there over the whole time period.
A couple of cautions need to be mentioned here. First, these data only cover two pollutants, sulfur dioxide and nitric acid. The two pollutants of most concern in Missouri are currently ozone and small particulate matter (PM2.5).
Second, these maps show average concentrations over three years. What if pollution was decreasing at some times of years, but increasing at others? These maps would not show it, a more detailed analysis would be required. The next post will look into that issue.
Ambient Air Concentrations, Clean Air Status and Trends Network (CASTNET), EPA, http://epa.gov/castnet/javaweb/airconc.html.
The two photos at right show Bryce Canyon National Park on a clear day and a hazy day. Bryce Canyon is one of the remotest locations in the continental United States. It is close to no cities and no major sources of air pollution. It is also dry, so the haze in the photo is not from humidity. Previous posts in this blog suggested that air quality was improving. What’s going on?
The air pollution monitoring program in Missouri focuses on large metropolitan areas or potentially large sources of pollution. Monitoring sites are often located next to pollution sources such as busy highways, industrial areas, or large lead smelters. The sites monitor pollution where it is most likely to be directly harmful. But they don’t tell us much about the background level of pollution that has dispersed into the atmosphere.
There are two ways to measure the background level of pollution. One method involves a network of rural monitoring sites far from cities and significant sources of pollution. Only there can you measure the degree to which pollutants have dispersed into the ambient air.
Spurred by the problem of acid rain, the federal government established just such a network in 1990, the Clean Air Status and Trends Network (CASTNET). Smaller at first, and focused on the eastern half of the country, it has grown into a national network of 81 monitoring sites. Each site is located in a rural area, far away from cities and significant sources of air pollution. CASTNET focuses on only a few pollutants most relevant for acid rain: sulfur dioxide and sulfates, nitric acid and nitrates, and ozone.
There are no CASTNET monitoring sites in Missouri; the nearest sites are in Alhambra, IL, Riley County, KS, Adair County, OK, Clark County, AR, and Trigg County, KY. I don’t know why Missouri is absent from the network.
The second way to measure the background level of pollution is to measure it from space. Several pollutant gases can be measured from space,and they include some important ones: ozone, carbon monoxide, carbon dioxide, methane, sulfur dioxide, nitric oxide, and formaldehyde.
My next few posts will focus on the ambient background levels of certain pollutants, as measured by the CASTNET program and from space. We’ve made progress in reducing the air pollution in our cities. Has the background level of pollution decreased as well? Or as the photos above might suggest, has it increased?
Site Information, Clean Air Status and Trends Network, EPA, http://java.epa.gov/castnet/epa_jsp/sites.jsp.
Remote Sensing of Tropospheric Pollution from Space. (2008). Fishman, J., K.W. Bowman, J.P. Burrows, A. Richter, K.V. Chance, D.P. Edwards, R.V. Martin, G.A. Morris, R. B. Pierce, J.R. Ziemke, J.A. Al-Saadi, J. K. Creilson, T.K. Schaack, and A.M. Thompson. Bulletin of the American Meteorological Society, 89,(6), 805-821.
The previous two posts have discussed the occurrence of tornadoes in Missouri and surrounding states, as well as national and Missouri trends in tornado frequency. This post concerns the most violent and destructive tornadoes, the F 3-5 tornadoes.
The amount of destruction a tornado causes depends not only on the strength of the tornado, but also on where it occurs. If it occurs in unpopulated, undeveloped land, the destruction may be less. If it occurs in a heavily populated area, and it if catches people unprepared, the destruction and loss of life may be greater. The last 50+ years have seen a significant increase in population, and a significant expansion of the amount of developed land. Thus, tornadoes are more likely to strike populated and developed areas.
The graph at right, copied from the U.S. Tornado Climatology web page of the National Climatic Data Center shows the annual number of strong to violent tornadoes in the United States. As has been the case with all of the annual tornado data we have looked at so far, this data varies greatly from year-to-year. However, it seems pretty clear that there is no overall trend toward increasing frequency. In fact, if anything, the trend has been towards decreased frequency – there are fewer highly destructive tornadoes now than there were during the mid-1960s. Who’da thunk it!
Some may take this as a disconfirmation of the theory of global warming. It turns out, however, that an increase in the number of tornadoes was not forecast by the IPCC Fourth Assessment Report or in the most recent two assessments of Global Climate Change Impacts in the United States. These documents all conclude that it is “…difficult to know if and how such events have changed as climate has warmed, and how they might change in the future.” (Global Climate Change Impacts in the United States, Thomas R. Karl, Jerry M. Melillo, and Thomas C. Peterson, (eds.). Cambridge University Press, 2009, p. 11.)
The confusion may arise because these documents did forecast an increase in some kinds of extreme weather events (the intensity of hurricanes, for instance). While all kinds of extreme weather may seem similar, a tornado is a very unique and specific weather phenomenon that may or may not be related to other kinds of extreme weather.
Thus, in summary, the total number of tornadoes reported annually has increased, but almost entirely from an increase in the number of small, weak tornadoes reported. If you eliminate the F-0 tornadoes from the count, the trend is basically flat. And the trend in the most powerful tornadoes seems to be towards decreased frequency.
U.S. Tornado Climatology, National Climate Data Center, http://www.ncdc.noaa.gov/oa/climate/severeweather/tornadoes.html#history.
Global Climate Change Impacts in the United States, Thomas R. Karl, Jerry M. Melillo, and Thomas C. Peterson, (eds.). Cambridge University Press, 2009.
IPCC, 2007: Summary for Policymakers. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA..
In my previous post, I discussed some statistics regarding tornado frequency in Missouri and surrounding states. In this post, I look at national and Missouri trends over time. Are tornadoes increasing in frequency, or are we just better at spotting them? Are they stronger and more destructive, or are we simply saturated with dramatic video from storm chasers?
At first, it might appear that tornadoes have increased in frequency. The first graph at right shows the total number of tornadoes reported since 1950. The blue line shows the number reported nationally, and should be read on the left vertical axis. The red line shows the number reported in Missouri, and should be read on the right vertical axis. There is a lot of variation from year-to-year, but the trends increase dramatically.
However, there are problems. Rain and temperature tend to occur over hundreds of square miles – they can be witnessed even in sparsely populated places, or by remote sensors. On the other hand, a tornado, and the damage it causes, are confined to the path of the tornado on the ground. To witness a tornado, it used to be that someone had to be right there – not likely in sparsely populated lands. Then they had to make a report, which might involve significant delay and travel over many miles. And finally, tornadoes can be hidden by darkness or rain. Thus, many tornadoes may have gone unwitnessed or unreported.
Our ability to witness and report has increased dramatically over the years, however. First, land that used to be empty is empty no longer – more of the land is populated with potential witnesses. Second, cell phones and the Internet greatly increase our ability to report tornadoes. And finally, the National Weather Service has deployed doppler radar, which now covers most of the country. Even tornadoes that escape human witness are seen by the radar. Consequently, many tornado experts believe that tornadoes have not increased dramatically in frequency, but rather, our ability to detect them has improved.
Scientists believe that the weakest of tornadoes are the ones most likely to have escaped detection in the past. So one way of trying to determine if the increase comes from more tornadoes or better detection would be to eliminate the weakest ones from the count, the F-0 tornadoes. The second graph at right shows the number of F-1 to F-5 tornadoes, again with the national numbers on the left vertical axis in blue, and the Missouri numbers on the right vertical axis in red. As before, there is tremendous variation from year-to-year. But this time, the overall trends do not appear to be dramatically increasing, they appears to be rather flat.
Thus, it appears that tornadoes are not really increasing in frequency. Rather, our improving ability to detect them seems to be making them seem more frequent.
Storm Prediction Center WCM Page, National Storm Prediction Center, NOAA, http://www.spc.noaa.gov/wcm/#enso.
In recent years, the news has been filled with dramatic reports of devastating tornadoes. From the Joplin tornado in 2011 to this year’s outbreak in Oklahoma, it seems like every year brings reports of terrible devastation from these powerful storms. Are they getting more frequent, or are we just better at detecting them? Are they getting more powerful, or are we just saturated with dramatic video from storm chasers? I thought it might be interesting to look at trends in the number of tornadoes nationally and in Missouri. In this post I will cover some tornado basics. In the next, I will look at trends in the number of tornadoes over time. In a third post, I will look at trends in the most severe tornadoes.
Tornadoes are most common during the spring and summer, but they can occur any time of the year. Texas is the state that experiences the largest number of tornadoes (half again as many as Kansas, which is second), but it is also the second largest state (after Alaska). If you divide the annual number of tornadoes by the land area of the state, the result is surprising. The table below shows the top states in tornadoes per 10,000 square miles of land area. Missouri is 12th.
|State||Tornadoes per 10,000 sq. mi.||State||Tornadoes per 10,000 sq. mi.|
The most important tornadoes, however, are the large, powerful ones. The National Oceanic and Atmospheric Administration ranks tornado strength on the Enhanced Fujita Scale from 0-5, with 5 being the most powerful. States with the most tornadoes ranked EF-3 to EF-5 are Kansas, Arkansas, and Texas. But again, if you divide by the state’s land area, you get some surprising results. The table below shows the stop states in EF-3 to EF-5 tornadoes per 10,000 square miles of land area. Missouri is tied with four other states for 8th.
|State||EF-3 to EF-5 Tornadoes per 10,000 sq. mi.||State||EF-3 to EF-5 Tornadoes per 10,000 sq. mi.|
Of the 10 most deadliest tornadoes, 3 have involved Missouri. The Joplin tornado, which occurred in 2011, is listed as the 7th most deadly in history. The 3rd most deadly tornado struck St. Louis in 1896. And the deadliest tornado in history was the Tri-State Tornado of 1925. It touched down near Ellington, in Reynolds County, MO. It travelled northeast, crossing the Mississippi River, traveling completely through Illinois, , finally dissipating about 3 miles southwest of Petersburg, Indiana. The total track was 219 miles in length with an average width of 3/4 of a mile. Some 695 people were killed, and 2,027 injured.
For national and state tornadoes: U.S. Tornado Climatology, National Climate Data Center, http://www.ncdc.noaa.gov/oa/climate/severeweather/tornadoes.html#history.
For the Tri-State Tornado: 1925 Tri-State Tornado: A Look Back, Paducah, KY National Weather Service Weather Forecast Office, http://www.crh.noaa.gov/pah/?n=1925tor.