Welcome to the second week of this course. Today, we will talk about drinking water safety. Well first, what does water safety mean? In fact, what does safety mean at all? Is it safe to swim in a lake, to walk from your home to school? Is it safe to eat a plate of chicken curry? Is it safe to fly in an airplane? Well, none of these things, actually, in fact, nothing in life at all is completely safe. There's always some risk, however small, that something can go wrong, and that while doing any of these activities, your health can be harmed. However, once we acknowledge that all activity has some risk, the critical question becomes, what level of risk is acceptable? And, how can problems be identified and controlled before they occur to manage risks and keep them at an acceptable level? If you go on an airplane flight, you're assuming that the plane won't crash. And why do you think that? Because there are guidelines and regulations in place to keep risks to acceptable levels. The airplane pilots have to go through a rigorous training program long before they can fly a real plane. The aircraft itself has to be regularly inspected and preventative maintenance checks must be performed routinely. There are auditing systems and inspections to make sure that these safety measures are applied correctly. And when the safety measures are all in place, traveling by plane is very safe. It's not ever 100% safe, but on a major airline today, the risk of being involved in a fatal accident is much, much less than one in a million, and most people would consider that to be a safe level of risk. While the same concept applies to drinking water. The World Health Organization has a definition of safe drinking water. Safe drinking-water is that which does not represent any significant risk to health over a lifetime of consumption, including different sensitivities that may occur between life stages. Now notice the definition doesn't say, does not represent any risk to health. It says, doesn't represent any significant risk to health. The WHO guidelines for drinking water quality include recommended maximum levels or guideline values for 90 different chemicals, and these guidelines values are normally set to represent a tolerable level of risk, which in the guidelines, is taken as one DALY per million person years. Remember DALY's? Disability-Adjusted Life Years? Well this risk level means that if a million people drink water for a year, with the chemical at the guideline value, then the chemical will cause health effects with a disease burden of one DALY over that population. And chemicals that cause cancer are treated in a similar way. A lifetime of drinking water at the guideline value should cause no more than one excess cancer case per 100,000 people. Now this is not zero risk, but it is considered a tolerable level. Put another way, risks smaller than this are considered not to be significant. WHO guideline values, of course, are not national standards, and it is up to each state to determine its own acceptable level of risk. Now health outcomes, or water quality guidelines link to health outcomes, are two examples of health based targets, which define an acceptable level of risk. There are other kinds of health based targets, such as performance targets, which specify a certain level of removal for a contaminant or a pathogen, or specified technologies, for example, saying that surface water must always be filtered before further treatment. Let's look a bit more at performance targets, since those will be important in the context of household water treatment. The main example of a performance target is a log reduction value, or LRV. This is the number of log units by which a chemical or other contaminate is reduced through treatment. Now you can easily calculate the LRV by taking the ratio of the contaminant before and after treatment, and then taking the base ten logarithm. Now, if the concentration is unchanged by the treatment, that means there's no removal, then the LRV is zero. It's also possible for a contaminate to increase during treatment, in which case the LRV is negative. Hopefully that doesn't happen too much. Now, if a hypothetical contaminant was reduced from a million to half a million counts, and here it doesn't matter what units are used, as long as they're consistent, that would be an LRV of 0.3, since 0.3 is the logarithm of two, roughly. And if it were reduced to 100,000, that's 90% reduction, that would be a one log reduction. So you can see that if we talk about reducing something 100 fold by 99%, that would be a two log reduction. And to go from one million of something to a single unit would be a six log reduction. So performance targets can specify the amount of log reduction values that should be achieved for a certain class of pathogens. For example, a three log removal of bacteria would mean removing 99.9% of bacteria. And a target of five log removal would be 100 times more strict. >> To find the total effectiveness of a multi-stage treatment system, you would calculate the LRV for each stage using the formula on the last slide, and then simply add the LRV's together. >> So, that's a quick introduction to health based targets. Health based targets are one part of the water safety framework described in the WHO guidelines for drinking water quality. In brief, the other main components are system assessment, operational monitoring, management and communication, and these are then confirmed by water quality testing for verification. Ideally, the verification step is done by an independent authority and not by the service provider. Information from the verification step can then feed back into the setting of health-based targets and acceptable levels of risks. The three steps in the middle are the actions that can be done by the water service provider. And taken together, they can constitute a water safety plan for a particular water supply system. First, in order to prevent risks, you have to understand all of the components of a water supply system from catchment to consumer. By examining each step in the water supply chain, potential hazards can be identified as places where something could go wrong, or where microbial contamination could occur. For each hazard, a control measure can then be put in place, which defines acceptable performance. Once control measures are established, system operators should routinely monitor that the control measures are being applied, and that if anything goes wrong, it is noticed quickly. This might include testing for E coli. But it's more likely to include very simple actions, like keeping pumping logs up to date, checking turbidity levels after filtration, or measuring residual chlorine as finished water leaves the plant. Think of the airplane example. You don't want your measure of air flight safety to be, hey, the plane didn't crash, it must have been safe. You want the airline to be routinely monitoring all of those safety measures that they have put in place. Management plans mainly describe the routine operations that are done under normal conditions. But it's also important to know what to do when a problem or incident occurs. Incident response plans should be developed and should be well known by operators so that if something bad happens, the operators know exactly what to do, and unsafe water is not delivered to consumers. Good records should be kept of both routine operations and any incidents that occur. And such information should be shared with consumers and there should be channels for consumers to give feedback to the system operators. There's a lot more to be said about water safety plans. In fact, it would be a good topic for a full MOOC. But, for our purposes, water safety plans are important as an example of how water safety can be conceptualized. At a very simple level, the water safety framework consists of health-based targets to define an acceptable level of risk, some system for identifying and managing risks, that's the water safety plan part, and independent verification that the risk management system is in fact working. In this framework, there is a need for measuring water quality. But mainly, as a verification measure. Most of the work to ensure water safety is in the system assessment in the developing of control measures and the operational monitoring to make sure they're being applied. If you only do water quality measurements, you can't be sure that the drinking water is safe, even if you don't find any E coli. People often make the mistake that if they've checked drinking water for E coli and didn't find any, they assume that the water must be safe. In fact, water without E coli could be unsafe for several reasons. First, contamination can be very variable in space. If you collect your sample at the wrong place, you could miss the problem. Second, contamination can be very variable in time. Maybe on the day that you do your water quality testing, there is no contamination. But the day before or even an hour before, there was, and you weren't testing then. Finally, E coli is not a perfect fecal indicator, though it is the best one available. It's possible that treatment has killed all of the E coli, but some other pathogens, for instance, cryptosporidium or rotavirus, are still viable. So to say drinking water is safe, you need both things. You need to show that the water is free from contamination, but also that there is a system in place to proactively identify and prevent contamination from occurring. Water safety plans are a fairly new concept, but they're taking off rapidly. There are currently over 20 countries that require water safety plans in their regulations. And many other countries are in the process of including them in future regulations. Water safety plans are typically applied for conventional treatment and pipe supplies at the level of the water supply scheme. There are examples of rural water safety plans, including for point sources such as boreholes or protected springs. But most of experience today has been with larger, professionally managed pipe systems. An important concept from the system assessment's step of water safety planning is that of multiple barriers. Remember, the main source of pathogens is fecal matter. So, multiple barriers can be put in place to prevent fecal material from entering the drinking water supply and to remove the pathogens that do make it in, using a variety of different treatment processes. Different pathogens have different levels of resistant to different kinds of treatment. Also, any single control measure can fail at some point. So by including multiple barriers to remove pathogens, the risk of providing unsafe drinking water can be minimized. Conventional treatment of surface water typically follows this multiple-barrier approach. Before reaching the treatment plant, the water resource itself can be protected. For example, the use of fertilizers, pesticides, or other chemicals may be restricted within the catchment area of a river or an aquifer used for drinking water supply. Within the treatment plant itself, a series of processes are normally followed. Some kind of pretreatment step may be necessary to reduce high levels of suspended solids, which could cause problems for subsequent treatment steps, storage reservoirs allow some sedimentation to take place, and roughing filters or bank infiltration can remove many particles, as well as improving the microbiological quality to some degree. Coagulation and flocculation remove turbidity, or suspended solids, by addition of a coagulant salt, typically aluminum sulfate or ferric chloride, which forms flocks that grow, removing suspended solids, as well as some dissolved compounds. These flocks then settle out of solution and clarification tanks, or in some cases, can float to the top of a tank, where they can be removed. Filtration can then remove any remaining flocks or may be applied directly without a coagulation step. Depending on the filter material, filtration can be an effective control measure for some pathogens, especially larger ones like cryptosporidium. Primary disinfection is an essential step in most treatment systems to achieve the necessary levels of pathogen reduction and to provide a safety net, in case previous steps fail. Chlorine is the most widely applied disinfectant, but chlorine dioxide, ozone, and ultraviolet light are used in some cases. Finally, after leaving the treatment plant, water suppliers can prevent contamination in the distribution system by ensuring continuous pressure, so that contamination can't enter the pipes, and by maintaining a residual disinfectant if chlorine is used. The same processes that are applied at conventional treatment plants can also be adapted and applied at the household level. In fact, all of the household water treatment systems use some combination of sedimentation, filtration, and disinfection. And they should all be followed by safe storage to prevent recontamination. In the next several videos, we will look at examples of these processes that are used in the main household water treatment systems. The main difference between conventional treatment and HWTS is not in the processes, but in the point where the processes are applied. HWTS requires households to take charge of their own water safety by treating the water at the household level and preventing recontamination. This is quite a challenge, since most household residents are not trained engineers and don't necessarily even have a modern concept of germ theory. The concepts of the water safety framework can be applied for HWTS, but implementing something like a water safety plan for household water treatment and storage requires fundamental changes in behavior. HWTS should not be seen as an alternative or competitor to conventional water supply. And the long term goal should be to extend access to safe drinking water supplies at the household level to everyone everywhere. HWTS can represent a short term strategy to improve drinking water quality until that long term goal can be reached. So in summary, we've heard today about the water safety framework described in the WHO Guidelines for Drinking Water Quality. The framework consists of three main components, health-based targets, water safety plans, and verification monitoring, independent verification. There are a lot of resources available that describe water safety plans in much more detail, if you're interested in learning more. We also learned about one kind of health-based target, log reduction values, or LRV. We'll come back to those, as we look at the different treatment processes used in HWTS. And we learned that the treatment process is used in household water treatment are the same that are applied in conventional water treatment plants. The key differences are in the location of the treatment and in the motivation and skills of the person applying the treatment.
