If your family gets drinking water from a private well, do you know if your water is safe to drink? What health risks could you and your family face? Where can you go for help or advice? EPA regulates public water systems; it does not have the authority to regulate private drinking water wells. Approximately 15 percent of Americans rely on their own private drinking water supplies, and these supplies are not subject to EPA standards, although some state and local governments do set rules to protect users of these wells. Unlike public drinking water systems serving many people, they do not have experts regularly checking the water's source and its quality before it is sent to the tap. These households must take special precautions to ensure the protection and maintenance of their drinking water supplies. Basic Information There are three types of private drinking water wells: dug, driven, and drilled. Proper well construction and continued maintenance are keys to the safety of your water supply. Your state water-well contractor licensing agency, local health department, or local water system professional can provide information on well construction.The well should be located so rainwater flows away from it. Rainwater can pick up harmful bacteria and chemicals on the land's surface. If this water pools near your well, it can seep into it, potentially causing health problems.Water-well drillers and pump-well installers are listed in your local phone directory. The contractor should be bonded and insured. Make certain your ground water contractor is registered or licensed in your state, if required. If your state does not have a licensing/registration program contact the National Ground Water Association. They have a voluntary certification program for contractors. (In fact, some states use the Association's exams as their test for licensing.) For a list of certified contractors in your state contact the Association at (614) 898-7791 or (800) 551-7379. There is no cost for mailing or faxing the list to you. To keep your well safe, you must be sure possible sources of contamination are not close by. Experts suggest the following distances as a minimum for protection - farther is better (see graphic on the right): Septic Tanks, 50 feet Livestock yards, Silos, Septic Leach Fields, 50 feet Patroleum Tanks, Liquid-Tight Manure Storage and Fertilizer Storage and Handling, 100 feet Manure Stacks, 250 feet Many homeowners tend to forget the value of good maintenance until problems reach crisis levels. That can be expensive. It's better to maintain your well, find problems early, and correct them to protect your well's performance. Keep up-to-date records of well installation and repairs plus pumping and water tests. Such records can help spot changes and possible problems with your water system. If you have problems, ask a local expert to check your well construction and maintenance records. He or she can see if your system is okay or needs work. Protect your own well area. Be careful about storage and disposal of household and lawn care chemicals and wastes. Good farmers and gardeners minimize the use of fertilizers and pesticides. Take steps to reduce erosion and prevent surface water runoff. Regularly check underground storage tanks that hold home heating oil, diesel, or gasoline. Make sure your well is protected from the wastes of livestock, pets, and wildlife. Dug wells are holes in the ground dug by shovel or backhoe. Historically, a dug well was excavated below the groundwater table until incoming water exceeded the digger's bailing rate. The well was then lined (cased) with stones, brick, tile, or other material to prevent collapse. It was covered with a cap of wood, stone, or concrete. Since it is so difficult to dig beneath the ground water table, dug wells are not very deep. Typically, they are only 10 to 30 feet deep. Being so shallow, dug wells have the highest risk of becoming contaminated.To minimize the likelihood of contamination, your dug well should have certain features. These features help to prevent contaminants from traveling along the outside of the casing or through the casing and into the well. Dug Well Construction Features The well should be cased with a watertight material (for example, tongue-and-groove precast concrete) and a cement grout or bentoniteclay sealant poured along the outside of the casing to the top of the well. The well should be covered by a concrete curband cap that stands about a foot above the ground. The land surface around the well should be mounded so that surface water runs away from the well and is not allowed to pond around the outside of the wellhead. Ideally, the pump for your well should be inside your home or in a separate pump house, rather than in a pit next to the well. Land activities around a dug well can also contaminate it. While dug wells have been used as a household water supply source for many years, most are "relics" of older homes, dug before drilling equipment was readily available or when drilling was considered too expensive. If you have a dug well on your property and are using it for drinking water, check to make sure it is properly covered and sealed. Another problem relating to the shallowness of a dug well is that it may go dry during a drought when the ground water table drops. Like dug wells, driven wells pull water from the water-saturated zone above the bedrock. Driven wells can be deeper than dug wells. They are typically 30 to 50 feet deep and are usually located in areas with thick sand and gravel deposits where the ground water table is within 15 feet of the ground's surface. In the proper geologic setting, driven wells can be easy and relatively inexpensive to install. Although deeper than dug wells, driven wells are still relatively shallow and have a moderate-to-high risk of contamination from nearby land activities. Driven Well Construction Features Assembled lengths of two inches to three inches diameter metal pipes are driven into the ground. Ascreened "well point" located at the end of the pipe helps drive the pipe through the sand and gravel. The screen allows water to enter the well and filters out sediment. The pump for the well is in one of two places: on top ofthe well or in the house. An access pit is usually dug around the well down to the frost line and a water dis-charge pipe to the house is joined to the well pipe with a fitting. The well and pit are capped with the same kind of large-diameter concrete tile used for a dug well. The access pit may be cased with pre-cast concrete. To minimize this risk, the well cover should be a tight-fitting concrete curb and cap with no cracks and should sit about a foot above the ground. Slope the ground away from the well so that surface water will not pond around the well. If there's a pit above the well, either to hold the pump or to access the fitting, you may also be able to pour a grout sealant along the outside of the well pipe. Protecting the water quality requires that you maintain proper well construction and monitor your activities around the well. It is also important to follow the same land use precautions around the driven well as described under dug wells. Drilled Wells Drilled wells penetrate about 100-400 feet into the bedrock. Where you find bedrock at the surface, it is commonly called ledge. To serve as a water supply, a drilled well must intersect bedrock fractures containing ground water. Drilled Well Construction Features The casing is usually metal or plastic pipe, six inches in diameter that extends into the bedrock to prevent shallow ground water from entering the well. By law, the casing has to extend at least 18 feet into the ground, with at least five feet extending into the bedrock. The casing should also extend a foot or two above the ground's surface. A sealant, such as cement grout or bentonite clay, should be poured along the outside of the casing to the top of the well. The well is capped to prevent surface water from entering the well. Submersible pumps, located near the bottom of the well, are most commonly used in drilled wells. Wells with a shallow water table may feature a jet pump located inside the home. Pumps require special wiring and electrical service. Well pumps should be installed and serviced by a qualified professional registered with your state. Most modern drilled wells incorporate a pitless adapter designed to provide a sanitary seal at the point where the discharge water line leaves the well to enter your home. The device attaches directly to the casing below the frost line and provides a watertight subsurface connection, protecting the well from frost and contamination. Older drilled wells may lack some of these sanitary features. The well pipe used was oftene ight-, 10- or 12- inches in diameter, and covered with a concrete well cap either at or below the ground's surface. This outmoded type of construction does not provide the same degree of protection from surface contamination. Also, older wells may not have a pitless adapter to provide a seal at the point of discharge from the well. Hydrofracting A Drilled Well Hydrofracting is a process that applies water or air under pressure into your well to open up existing fractures near your well and can even create new ones. Often this can increase the yield of your well. This process can be applied to new wells with insufficient yield and to improve the quantity of older wells. How can I test the quality of my private drinking water supply? Consider testing your well for pesticides, organic chemicals, and heavy metals before you use it for the first time. Test private water supplies annually for nitrate and coliform bacteria to detect contamination problems early. Test them more frequently if you suspect a problem. Be aware of activities in your watershed that may affect the water quality of your well, especially if you live in an unsewered area. Back to Top Human Health The first step to protect your health and the health of your family is learning about what may pollute your source of drinking water. Potential contamination may occur naturally, or as a result of human activity. What are Some Naturally Occurring Sources of Pollution? Microorganisms: Bacteria, viruses, parasites and other microorganisms are sometimes found in water. Shallow wells - those with water close to ground level - are at most risk. Runoff, or water flowing over the land surface, may pick up these pollutants from wildlife and soils. This is often the case after flooding. Some of these organisms can cause a variety of illnesses. Symptoms include nausea and diarrhea. These can occur shortly after drinking contaminated water. The effects could be short-term yet severe (similar to food poisoning) or might recur frequently or develop slowly over a long time. Radionuclides: Radionuclides are radioactive elements such as uranium and radium. They may be present in underlying rock and ground water Radon: Radon isa gas that is a natural product of the breakdown of uranium in the soil - can also pose a threat. Radon is most dangerous when inhaled and contributes to lung cancer. Although soil is the primary source, using household water containing Radon contributes to elevated indoor Radon levels. Radon is less dangerous when consumed in water, but remains a risk to health. Nitrates and Nitrites: Although high nitrate levels are usually due to human activities (see below), they may be found naturally in ground water. They come from the breakdown of nitrogen compounds in the soil. Flowing ground water picks them up from the soil. Drinking large amounts of nitrates and nitrites is particularly threatening to infants (for example, when mixed in formula). Heavy Metals: Underground rocks and soils may contain arsenic, cadmium, chromium, lead, and selenium. However, these contaminants are not often found in household wells at dangerous levels from natural sources. Fluoride: Fluoride is helpful in dental health, so many water systems add small amounts to drinking water. However, excessive consumption of naturally occurring fluoride can damage bone tissue. High levels of fluoride occur naturally in some areas. It may discolor teeth, but this is not a health risk. What Human Activities Can Pollute Ground Water? Septic tanks are designed to have a "leach field" around them an area where wastewater flows out of the tank. This wastewater can also move into the ground water. Bacteria and Nitrates: These pollutants are found in human and animal wastes. Septic tanks can cause bacterial and nitrate pollution. So can large numbers of farm animals. Both septic systems and animal manures must be carefully managed to prevent pollution. Sanitary landfills and garbage dumps are also sources. Children and some adults are at extra risk when exposed to water-born bacteria. These include the elderly and people whose immune systems are weak due to AIDS or treatments for cancer. Fertilizers can add to nitrate problems. Nitrates cause a health threat in very young infants called "blue baby" syndrome. This condition disrupts oxygen flow in the blood. Concentrated Animal Feeding Operations (CAFOs): The number of CAFOs, often called "factory farms," is growing. On these farms thousands of animals are raised in a small space. The large amounts of animal wastes/manures from these farms can threaten water supplies. Strict and careful manure management is needed to prevent pathogen and nutrient problems. Salts from high levels of manures can also pollute ground water. Heavy Metals: Activities such as mining and construction can release large amounts of heavy metals into nearby ground water sources. Some older fruit orchards may contain high levels of arsenic, once used as a pesticide. At high levels, these metals pose a health risk. Fertilizers and Pesticides: Farmers use fertilizers and pesticides to promote growth and reduce insect damage. These products are also used on golf courses and suburban lawns and gardens. The chemicals in these products may end up in ground water. Such pollution depends on the types and amounts of chemicals used and how they are applied. Local environmental conditions (soil types, seasonal snow and rainfall) also affect this pollution. Many fertilizers contain forms of nitrogen that can break down into harmful nitrates. This could add to other sources of nitrates mentioned above. Some underground agricultural drainage systems collect fertilizers and pesticides. This polluted water can pose problems to ground water and local streams and rivers. In addition, chemicals used to treat buildings and homes for termites or other pests may also pose a threat. Again, the possibility of problems depends on the amount and kind of chemicals. The types of soil and the amount of water moving through the soil also play a role. Industrial Products and Wastes: Many harmful chemicals are used widely in local business and industry. These can become drinking water pollutants if not well managed. The most common sources of such problems are: Local Businesses: These include nearby factories, industrial plants, and even small businesses such as gas stations and dry cleaners. All handle a variety of hazardous chemicals that need careful management. Spills and improper disposal of these chemicals or of industrial wastes can threaten ground water supplies. Leaking Underground Tanks & Piping: Petroleum products, chemicals, and wastes stored in underground storage tanks and pipes may end up in the ground water. Tanks and piping leak if they are constructed or installed improperly. Steel tanks and piping corrode with age. Tanks are often found on farms. The possibility of leaking tanks is great on old, abandoned farm sites. Farm tanks are exempt from the EPA rules for petroleum and chemical tanks. Landfills and Waste Dumps: Modern landfills are designed to contain any leaking liquids. But floods can carry them over the barriers. Older dumpsites may have a wide variety of pollutants that can seep into ground water. Household Wastes: Improper disposal of many common products can pollute ground water. These include cleaning solvents, used motor oil, paints, and paint thinners. Even soaps and detergents can harm drinking water. These are often a problem from faulty septic tanks and septic leaching fields. Lead & Copper: Household plumbing materials are the most common source of lead and copper in home drinking water. Corrosive water may cause metals in pipes or soldered joints to leach into your tap water. Your water's acidity or alkalinity (often measured as pH) greatly affects corrosion. Temperature and mineral content also affect how corrosive it is. They are often used in pipes, solder, or plumbing fixtures. Lead can cause serious damage to the brain, kidneys, nervous system, and red blood cells. The age of plumbing materials - in particular, copper pipes soldered with lead - is also important. Even in relatively low amounts these metals can be harmful. EPA rules under the Safe Drinking Water Act limit lead in drinking water to 15 parts per billion. Since 1988 the Act only allows "lead free" pipe, solder, and flux in drinking water systems. The law covers both new installations and repairs of plumbing. Back to Top What You Can Do... Private, individual wells are the responsibility of the homeowner. To help protect your well, here are some steps you can take: Have your water tested periodically. It is recommended that water be tested every year for total coliform bacteria, nitrates, total dissolved solids, and pH levels. If you suspect other contaminants, test for those. Always use a state certified laboratory that conducts drinking water tests. Since these can be expensive, spend some time identifying potential problems. Testing more than once a year may be warranted in special situations: someone in your household is pregnant or nursing there are unexplained illnesses in the family your neighbors find a dangerous contaminant in their water you note a change in water taste, odor, color or clarity there is a spill of chemicals or fuels into or near your well when you replace or repair any part of your well system Identify potential problems as the first step to safeguarding your drinking water. The best way to start is to consult a local expert, someone that knows your area, such as the local health department, agricultural extension agent, a nearby public water system, or a geologist at a local university. Be aware of your surroundings. As you drive around your community, take note of new construction. Check the local newspaper for articles about new construction in your area. Check the paper or call your local planning or zoning commission for announcements about hearings or zoning appeals on development or industrial projects that could possibly affect your water. Attend these hearings, ask questions about how your water source is being protected, and don't be satisfied with general answers. Make statements like "If you build this landfill, (just an example) what will you do to ensure that my water will be protected." See how quickly they answer and provide specifics about what plans have been made to specifically address that issue. Back to Top Identify Potential Problem Sources To start your search for potential problems, begin close to home. Do a survey around your well: is there livestock nearby? are pesticides being used on nearby agricultural crops or nurseries? do you use lawn fertilizers near the well? is your well "downstream" from your own or a neighbor's septic system? is your well located near a road that is frequently salted or sprayed with de-icers during winter months? do you or your neighbors dispose of household wastes or used motor oil in the backyard, even in small amounts? If any of these items apply, it may be best to have your water tested and talk to your local public health department or agricultural extension agent to find way to change some of the practices which can affect your private well. In addition to the immediate area around your well, you should be aware of other possible sources of contamination that may already be part of your community or may be moving into your area. Attend any local planning or appeal hearings to find out more about the construction of facilities that may pollute your drinking water. Ask to see the environmental impact statement on the project. See if underground drinking water sources has been addressed. If not, ask why. Common Sources of Potiental Ground Water Contamination Category Contaminant Source Agricultural Animal burial areas Drainage fields/wells Animal feedlots Irrigation sites Fertilizer storage/use Manure spreading areas/pits, lagoons Pesticide storage/use Commercial Airports Jewelry/metal plating Auto repair shops Laundromats Boatyards Medical institutions Car washes Paint shops Construction areas Photography establishments Cemeteries Process waste water drainage Dry cleaners fields/wells Gas stations Railroad tracks and yards Gulf courses Research laboratories Scrap and junkyards Storage tanks Industrial Asphalt plants Petroleum production/storage Chemical manufacture/storage Pipelines Electronic manufacture Process waste water drainage Electroplaters fields/wells Foundries/metal fabricators Septage lagoons and sludge Machine/metalworking shops Storage tanks Mining and mine drainage Toxic and hazardous spills Wood preserving facilities Residential Fuel Oil Septic systems, cesspools Furniture stripping/refinishing Sewer lines Household hazardous products Swimming pools (chemicals) Household lawns Other Hazardous waste landfills Recycling/reduction facilities Municipal incinerators Road deicing operations Municipal landfills Road maintenance depots Municipal sewer lines Storm water drains/basins/wells Open burning sites Transfer stations

Douglas Go Nose, George W. Mushrush and Stephen Kline Center of Applied Science, George Mason University, Fairfax, VA 22030 ABSTRACT The ability of activated charcoal and alpha-track radon monitors to estimate annual concentrations is a consequence of the measurement interval over which either type of monitor is used. In a case study of several hundred homes in Virginia and Maryland, a k 90% uncertainty must be applied to single charcoal measurements to estimate the annual radon concentrations, and a * 50% uncertainty must be applied to three-month alpha-track measurements (90% confidence levels). INTRODUCTION The realization that Radon0222 and its progeny constitute a health risk was brought to the attention of Virginia and Maryland homeowners in 1986 and 1987. Although the scientific community had recently reported on radon (1, 2, 3), homeowners were alerted by a series of news media reports that noted the recent discoveries of homes with radon problems in Pennsylvania and New Jersey. This quickly was followed by reports about radon activities of federal, state and county agencies concerned with public health. ~t the same time, news media and civic association attention was directed toward radon testing companies, and toward the Center of Applied Science at George Mason University. In the summer of 1986, only shortly before the regional news media started its coverage on radon, the Center had started to develop a pilot study in area homes. From late in 1986 to the present, the Center has received about 5000 letters and phone calls in which homeowners request information about the study. About 1800 homes are now being examined. Almost all of the study homes are in Fairfax County in northern Virginia, and the contiguous Montgomery County in south central Maryland. A few homes are from surrounding counties. The geological rock units under Fairfax County extend through Montgomery County and to adjacent counties in central Virginia and northern Maryland. The radon "signaturen of the rock units vary considerably from one rock unit to another. Aeroradioactivity maps (total gamma signal, measured from an airplane, due to radon, thorium, and potassium) show that each rock unit tends to vary in total radioactivity. Aeroradioactivity naps appear to be a good device with which to predict indoor radon, at least at the community level. No attempt has yet been made to examine the comparison between single radon measurements and the annual radon measurements in terms.of the geological units under the home. This may be a fruitful line of study since season-toseason indoor radon variations are not the same in homes over different rock units. It may be that short term fluctuations of indoor radon are related to soil radon concentration as well as to soil permeability. This paper concentrates on comparisons using single charcoal and alpha-track measurements, without any consideration of the geology, home construction and home use factors. This is not an unreasonable approach, since the typical homeowner takes none of these factors into consideration. Also, studies in progress suggest that these variables are important but not significant enough to alter the conclusions discussed below. PRECISION AND ACCURACY At the present time, only two types of inexpensive radon moitors are widely used to measure indoor radon. The less expensive (about $10-$20 per monitor to the homeowner) type uses activated charcoal, packaged in a porous bag or in a small metal can. The somewhat more expensive (about $20-30 to the homeowner) type of monitor uses a radon sensitive film packaged in a plastic cup. Other methods, most of which involve the use of "grab samplesw of indoor air, will not be discussed. The most important problem faced by any measuring device for the homeowner is the natural variation in radon concentration. Radon concentrations are known to fluctuate rapidly, with low point to high point changes of more than 100%. These changes are due to changes in weather and home use. The question that this paper will address concerns the length of time required to obtain a meaningful estimate of the annual radon concentration. This requires deciding how precisely one wishes to estimate the annual radon concentration. In -short, to obtain any given level of precision in estimating the annual radon level, how long must the radon concentration be measured? The other important question that might be considered is accuracy. For discussions that follow, it will be assumed that the commercially available charcoal and alpha-track monitors are equally accurate. That is, over the interval of radon measurement, both are considered to be equally accurate. Although there is some debate on this point, it is generally recognized that both types of monitors carry a measurement uncertainty of about i 25% at the 90% confidence level. For the following discussions, it will also be assumed that neither the charcoal monitor nor the alpha-track monitor yield results that are biased toward too-high or too-low measurements. The commonly used laboratories all routinely pass the EPA Proficiency Program, particularly over the early 1987 to present interval of this study. Consequently, the pattern of deviation of single measurements from annual measurements is considered to be related to natural variations in indoor radon concentrations. Finally, it will be assumed that the annual radon concentration can be adequately estimated by averaging radon concentrations from a series of four alpha-track measurements, each over three months. With these assumptions, single charcoal and single alpha-track measurements can be compared to annual radon concentrations. CHARCOAL RADON MONITORS The charcoal monitors used in this study include monitors from The Radon Project in Pittsburgh, Pennsylvania, Enrad Corporation in Gaithersburg, Maryland, and AirChek Corporation in Penrose, North Carolina. Charcoal monitors from The Radon Project have a vapor barrier to retard the adsorption of water which interferes with the adsorption of radon. These monitors are therefore, according to the manufacturer, to be left exposed to indoor air for about 7 days. The Enrad monitors do not have a vapor barrier, but are carefully weighed to determine the amount of water collection so as to appropriately adjust the measured radon concentration. The Enrad monitors are to be left exposed to indoor air for about 2-3 days. Monitors from The Radon Project and from Enrad are metal cans, while radon monitors from AirChek are porous paper bags. The AirChek monitors are to be exposed to indoor air for only 1-2 days, and water adsorption is not considered to be important. The major benefit from the charcoal monitor is its low cost and rapidity in which it yields results for the homeowner. The major disadvantages are its sensitivity to water and to temperature (adsorption characteristics change measurably if indoor air changes to wuncomfortablett temperature conditions), and the fact that the radiation record is lost over a few days, providing no legal record other than the analyst's report. The limitation on the measurement interval to only a few days means that a useful estimate of the annual radon concentration requires a series of many charcoal monitors, or it requires that an estimate of uncertainty be applied to a single measurement. Some estimates on uncertainty obtained by using charcoal monitors over several weeks have been proposed (4), but no study of variations using a series of charcoal monitors over an entire year has yet been reported. In the following discussions, we report on homes in which we have one charcoal measurement and an estimate of annual radon concentration, plus we report on additional homes for which we have a single charcoal measurement and an alpha-track measurement over the season of charcoal measurement. ALPHA-TRACK RADON MONITORS The alpha-track monitors used in this study are from Tech/Ops Landauer Corporation in Illinois. With an adequate soil shield, they have been used for hydrocarbon exploration ( 5 ) , earthquake prediction ( 6 ) , and in the search for uranium and gold. In the past few years, they have been used without the soil shield to measure indoor radon. The indoor radon monitors also do not require the use of a permeable membrane required on soil monitors to keep out radon-219 and radon-220, the other radon isotopes that fortunately have very short half-lives. The indoor radon monitor does have a dust filter, through which the radon can pass. A fraction of the radon produces alpha particles that penetrate the small square of plastic film inside the monitor, producing alpha-tracks that are enhanced by chemical etching. The nuclear tracks recorded on the small square of plastic film inside these monitors are not affected by normal variation in home humidity and temperature, and the dislocation sites (more commonly called alpha-tracks) are permanently recorded on the film. The humidity and temperature insensitivity and the permanent record keeping are probably the major advantages of the alpha-track monitors. The inexpensive alpha-track monitors with their small fragment of film require at least one month for enough tracks to accumulate in a typical home to generate a useful measurement. This relatively long measurement interval is a disadvantage when a homeowner or more often home buyer wants a measurement quickly. Estimates of analytical uncertainty for the alpha-track monitors are dependent on the measurement interval ( 7 ) , so intervals of three months to a full year are often utilized. ANNUAL RADON AND CHARCOAL MONITORS The radon study of the Center of Applied Science has gathered charcoal and alpha-track measurements during 1987 and 1988. Because homeowners paid for their radon monitors, most returned their monitors and the questionnaires related to the monitors. There are now 152 homes which have a single charcoal measurement and an estimate of the annual radon concentration using the average of four seasonal alpha-track monitors. There are 329 homes which have a charcoal measurement and an alpha-track measurement over the season of charcoal measurement. In this paper, winter is Novenber- January, spring is February-April, summer is May-July, and fall is August-October, The charcoal measurements are from each month of the year, though a few months provided most of the comparisons (Figure 1). An overview shows that deviations of single charcoal measurements commonly occur that are more than 50% higher or lower than the seasonal alpha-track measurement. Only rarely, but sometimes, the deviation is on the order of 100%. Of course, these deviations are to some extent due to inaccuracies in the charcoal measurement and inaccuracies in the alpha-track measurement. However, since these analytical variations are thought to be random, the situation shown in Figure 1 is thought to represent a realistic picture of deviations caused by natural variations in radon. Table 1 compares the single charcoal measurements to the annual radon concentrations. Charcoal-to-annual ratios of less than 1.0 represent cases where the charcoal measurement was less than the annual measurement; ratios of more than 1.0 represent cases where the charcoal was greater than the annual measurement. Most of the available charcoal measurements are from the summer interval (Figure 2), and even though the homeowners were instructed to use closed-home conditions, the majority of the summer measurements are less than the annual measurement. We suspect that closed-home conditions may somewhat simulate winter conditions, but not very completely. The lack of indoor-to-outdoor thermal convection in the summer, plus variable weather conditions probably prevent closed home conditions from reliably simulating a winter condition, or even a condition that might be considered average in terms of radon concentration. The deviation of charcoal monitor radon measurements from annual radon concentrations does not appear to be a function of the indoor radon concentration (Figure 3), at least over the 1-20 pCi/1 annual radon range of the study set. This observation supports the idea that the deviations are real, and not a consequence of measuring low radon concentrations. Table 2 provides an estimate of the uncertainly that should be applied to a single charcoal measurement in order to estimate the possible annual radon concentration. For example, the data show that 67% of the homes yield a charcoal-to-annual deviation of up to  40%. his could be rephrased to say that at the 67% confidence level, the uncertainty that would be applied to a single charcoal measurement is i 40% of the charcoal measurement. Similarly, one would apply a  90% uncertainty to the charcoal measurement if one wanted to be very sure (e.g., 90% confidence level) of the possible annual radon concentration. Obviously these uncertainties are considerably larger than the $ 25% uncertainty noted earlier that is applied to a single measurement, but only over the measurement interval. The much larger uncertainty is a consequence of the need to estimate the annual radon concentration, compared to the much less useful radon concentration during only the measurement interval. ANNUAL RADON AND ALPHA-TRACK MONITORS The number of alpha-track measurements in this study greatly exceed the number of charcoal measurements, mainly because the homeowners were alerted to the probably inherent uncertainty associated with the charcoal monitors. The charcoal monitors were probably selected by homeowners who wanted to experiment with the technology, or homeowners who wanted to make their own judgement as to the usefulness of the charcoal monitors. There are presently a total of 828 homes for which the entire sequence of four three-month alpha-track monitors are available (Table 3). Winter alpha-track measurements tend to be greater than the annual concentration, and summer measurements tend to be less than the annual concentration; spring and fall measurements are less biased toward higher or lower measurements (Figure 4). This situation is obviously related to natural seasonal variations. One could apply a correction factor to adjust a measurement, and one could then apply an uncertainty to the measurement to estimate the annual radon concentration (Table 4). A comparison of Tables 2 and 4 shows the dramatic difference in the uncertainty estimate between the charcoal and the alpha-track detectors. For example, one could say that at the 67% confidence level, one would apply a & 40% uncertainty to the charcoal measurement and a  25% uncertainty to the alpha-track measurement. To be very sure of the possible annual radon concentration, one would apply a  90% uncertainty to the charcoal measurement and a  50% uncertainty to the alpha-track measurement. As was noted for the charcoal monitors (see Figure 3) , the deviation of single alpha-track measurements from annual radon concentrations does not appear to be a function of indoor radon concentration (Figure 5). Deviations of about the same magnitude occur for both low and high radon concentrations, for all the seasonal intervals. The deviations are therefore concluded to be the result of natural variations in radon, and not measurement inaccuracies. CONCLUSION Almost all indoor radon measurements are currently obtained by homeowners using activated charcoal radon monitors or alpha-track monitors. Manufacturer estimates for the measurement interval (a few days for the charcoal and a few months for the alpha-track monitors) uncertainties are about  25% at the 90% confidence level. However much larger uncertainties must be applied to estimate the annual radon concentration. This uncertainty is about i 90% for the charcoal monitors and about  50% for the alpha-track monitors. One implication of these uncertainty estimates is that charcoal monitors should best be considered a "sampler" of indoor radon that is useful only for the measurement interval. Homeowners who wish to obtain a useful estimate of annual radon should be advised to use a series of charcoal monitors (perhaps 5 over 10 weeks), or a single alpha-track monitor exposed for perhaps three months. It may also be important to reconsider the validity of using 4 pCi/1 as an "action levelw to be applied to a single charcoal measurement. A single charcoal measurement of 3.9 pCi/l could in fact come from a home that has an annual radon concentration of more than 7 pCi/l. An "action levelw of 2 pCi/1 should perhaps be applied to charcoal measurements. Another important observation concerns the concept of klosed-homeM measurements. The available data show that the closed-home condition often yields measurements that are less than the annual radon concentrations, and very often less than the wworst-casew conditions thought to prevail in the winter. Variables such as geology-determined permeability behavior, weather and home construction may interact in ways that often prevent a closed-home condition from facilitating a short-term (charcoal monitor) worst-case measurement. The obvious implication is that homeowners, realtors and scientists should be cautious when using charcoal measurements to estimate annual radon concentrations. This caution, plus a realistic estimate of the measurement uncertainties, can generate radon estimates that have significance. The work described in this paper was not funded by the U.S. Environmental Protection Agency and therefore the contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred. REFERENCES 1. Moshandreas, D.L. , and Rector, HOE. Indoor radon concentrations: Envir. Inter. 8:77, 1982. 2. Nero, A.V., Schwehr, M.B., Nazaroff, W.W., and Revzan, K. L. Distribution of airbourne radon9222 concentrations in U.S. homes: Science, 234: 992,1986. 3. Alter, H. W., and Oswald, R.A. Nationwide distribution of indoor radon measurements, a preliminary data base: J. Air Pollution Control ASSOC. 37: 227, 1987. 4. Cohen, B. L., and Gromicko, N. Adequacy of time averaging with diffusion barrier charcoal adsorption collectors for radon0222 measurements in homes: Health Physics 54: 195, 1988. 5. Fleischer, R.L., and Turner, LOG. Correlations of radon and carbon measurements with petroleum and natural gas at Cement, Oklahoma: Geophysics 49:810, 1984. 6. ~leischer, R. L. , and Mogro-Campero, A. Association of subsurface radon changes in Alaska and the northeastern United States with earthquakes: Geochim. et Cosmochem. Ada 49:1061, 1985. 7. Cohen, B.L. Comparison of nuclear track and diffusion barrier charcoal adsorption methods for measurments for measurement of radon9222 levels in indoor air: Health Physics 50: 828, 1986.