Malawi

December 19, 2006: Email to Hayden

See the photos associated with this entry.

Dear Hayden,

      Thanks for emailing, and I apologize for not writing sooner.  You said you were curious what I am doing for the Clinton Foundation, and I wanted to make sure I spent the time to give you a good answer. Over time though I realized I might not really be so sure.  Sometimes I understand, and it makes such good sense that I am surprised that it wasn’t already done, or that I’m not already finished.  Other times it seems so cumbersome, with so many caveats and unanswered questions, that I’m not sure what I’m doing, or at least if what I am doing is useful. 

For the last few weeks I have been trying to write a short report about how to use what I have been doing to plan a water project for one “rural growth center” where the Clinton Foundation is working, and instead I am sitting staring at the monkey jumping around in the tree outside my window because I can’t figure out how to make it useful.

So, I decided if I can explain to you what I’m doing, and why, I might have better luck figuring out what it is that will be useful.

Malawi is one of the developing countries committed to the "Millennium Development Goals (MDG)", which are a number of development goals agreed upon as critical to providing fundamental rights to people all over the world.  The goals include things like access to health care, and education, and safe drinking water. 

Under the MDGs, each country has agreed to targets for improving the lives of people throughout the country.  For example, with regards to water, the MDG is that during the 25 year period between 1990 and 2015 Malawi will cut by half the proportion of its population who do not have access to safe drinking water.  For rural areas, access to safe drinking water is defined as having a source of clean water, shared by 250 people or less, within 500 meters of each household.

The project that I am working on is building an inventory of all the existing safe drinking water sources in the rural areas of Malawi.  When the number and location of all the safe water sources are known, it should be possible to determine how well Malawi is doing toward meeting the MDGs for water in the rural areas.  Ideally, if the right information is collected for each water source, the inventory should also be used to know what types of water sources work best for each area of the country, and planning for providing new water sources could be based on solid information, particularly for determining where new water sources are needed.

As far as the MDGs for clean water in rural areas are concerned, rural areas include all of Malawi except the 3 major cities, the capital Lilongwe, largest city Blantyre/Limbe, and Mzuzu in the north part of the country, and the 20 or so small towns scattered around the country that have municipal piped water systems providing chlorine treated water either directly to houses or, more commonly, to shared communal tap stands, where a number of households use one or more taps to fill water containers that are carried back and used for drinking, cooking, and washing. 

In total, the population in the rural areas of Malawi is somewhere around 9 million, or 80-85% of the population of the country.

So far, the inventory contains data for more than 43 thousand water sources, or water points, collected from 23 of 28 districts in Malawi.  The rural population in the 23 districts with data is 6.8 million. The water sources in the inventory are called “improved community water points,” or “ICWPs”  They are “improved”, meaning the source of the water is protected from pollution by a structure such as a concrete cover, or the water is filtered and treated in the case of piped water systems, and they are available to a community, not private water points used by only a business or a single household.

In Malawi, most improved community water points, are either hand operated pumps mounted on brick and concrete slabs drawing groundwater from drilled or hand-dug wells, or taps, delivering water taken from streams and rivers, usually partially treated by settling or filtration, and (supposedly) chlorinated, and moved through a network of pipes and holding tanks by gravity. 

When people don’t have access to a functioning ICWP they use any water available.  Mostly, they use rivers and streams, or shallow wells with a bucket to draw water.  Along Lake Malawi they draw water directly from the lake.  A lot of times they use scoop holes, which are just shallow holes dug into hillsides or beside a road, where water collects and they scoop the water into buckets.  The biggest problem with these water sources is that they are often contaminated.  People are using the same water for cleaning clothes and washing themselves, or they are contaminated by animals, trash, and even industrial waste or untreated sewerage waste.

The data for each ICWP has been collected during the last 3 years.  Most was collected by government extension workers traveling around by motorcycle or on bicycles, collecting information using a hand-held GPS and by interviewing people and filling out a questionnaire. The money for the field work was provided by a number of big organizations such as UNICEF, JICA (Japanese International Cooperation Agency), DIFD (United Kingdom Department for International Development), Irish Aid, CIDA (Canadian International Development Agency), and others. And the efforts to do the data collection was organized mostly by non-governmental organizations (NGOs) including Water Aid Malawi, the Water Supply and Sanitation Collaborative Council (WSSCC), Concern Universal, Oxfam, Emmanuel International, GITEC, COMWASH, CCAP, and others.  The majority of the data was collected during specific water point mapping projects, but in some places the data was collected to support other ongoing projects, for example water point repair and rehabilitation projects.

So, data is available for lots of ICWPs around the country. And it should be possible to use this data to determine where people are getting their water, figure out how many are able to access clean drinking water, and map out where additional ICWP are needed and what types are most likely to work.

And I should be able to make practical and reasonable arguments using all this data.  But I am stuck in the details.

For all of Malawi the MDGs sound so clear.  “halve, by 2015, the proportion of people without sustainable access to safe drinking water” by providing at least one water point for every 250 people within 500 meters of their household.

Two things I will quickly do away with, since they are the most intractable. Since no one has an inventory of water points in 1990, all we can do is guess what the proportion with access was then, with estimates mostly between 40% and 50%.  I don’t think the “halve” part is really workable, so I have to act as if the goal is really to provide the safe drinking water to everyone, eventually. Will we be half way there by 2015?  Who knows. 

The other is “sustainable” access.  It is used to mean “if someone builds a water point can the community using it keep it functioning?”  There are many dissertations to be written to discuss this, so I will skip it by parroting the (unsubstantiated claim) that if the community feels ownership they will figure a way to keep it functioning.  (I feel ownership of my car, but if I don’t have money and a mechanic and spare parts, I can’t keep it functioning.)

The parts that might be clear are that there should be at least one water point for every 250 people within 500 meters of their household. Of course, households are distributed across the landscape.  There are few restrictions on where houses are built.  Sometimes new houses are built between existing houses, but equally they are built farther away to take advantage of newly available land for farming or cutting trees, or just to reduce crowding. So, having a source of clean water within 500 meters of each household, really would require that the location of each household is known, and that a source is provided to each household that does not have one. 

This isn't likely for a number of reasons.  From a financial perspective, the cost of a machine drilled borehole, drilled 20 or 30 meters into the ground and fitted with a standard pump, averages US$5,000 finished.  Shallow, hand-dug wells fitted with the same type of pump might cost US$500-US$1,000.  As importantly, even if the money was available, reliable sources of clean water are not available everywhere. There are some of areas in Malawi where you can put boreholes and get water almost every time, but there are also lots of areas where there is little or no groundwater available.  Unfortunately, these are often the areas where people are spread out, struggling to get by, using land that is only marginally suitable for subsistence farming, growing only enough crops to eat with little or none left over to sell for income.  In these areas, where people are spread out, even if it was financially possible to try to develop water sources by digging wells or boreholes, the water is just not there.  And gravity-fed tap systems require a good water source, uphill from where people are living.  A lot of people live on ridges or on the top of hills because that is where the flattest land is for farming.

The goal of no more than 250 people sharing a water source is the other part that sounds relatively clear.  From a practical view, the reason that this target is important is that most clean water sources in Malawi are either hand-pumps providing ground water, or water taps providing surface water through a network of pipes.  Most people, which really means women or girls who are responsible for fetching water, get water by walking to one of these water sources carrying a 20 liter galvanized or plastic bucket, filling the bucket, and then hoisting the bucket on their head and walking home with it. 

For a household with 5 or 6 people, they are going to use 4 or 5 or more buckets each day for drinking and cooking.  So 4 or 5 trips to the pump or the tap.  Say on average each household has 5 people, that means that there are going to be 50 households using the water source, and they are going to make 4 trips per day, or 200 trips.  That is 200 20-liter buckets.  Unless the pump or the tap is really productive, filling each bucket takes a long time.  Maybe 5 minutes.  Sometimes less, sometimes more.   Even if it only takes 2 minutes to fill each bucket, the water source is going to be in use filling buckets for more than 6 hours each day.  If the wait for water is too long, people will go and get water from unprotected sources.

Unfortunately, I don’t think anyone has collected any systematic data on how many people are using any specific water point.  It might be possible in some places, where the community is organized enough to record who is using the water points, but I have never seen any data like that.  So, absent this information, the number of people using any water point has to be an average for the area. But what area should be used? 

On the biggest scale, if you take the rural population of the districts where data is available, which is about 6.8 million people, and divide it by 43,000, the total number of water points in the database, you get 158 people per-water point. The MDG for access to safe water in the country is already met, and then some.  But, not all the water points are functioning, or were functioning when they were surveyed.  About 35% of the surveyed water points were not functioning.  So, do the same calculation with functioning water points, which is around 28,000, and you get 242 people per-functioning water point.  Still above the MDG. 

Considering the fact that most people get their water on foot, and carry their water back to their household in a bucket, it is obvious that dividing all the rural population by all the water points in Malawi really should not be used.  It is not really useful for figuring out if individual people have access to safe drinking water.  People can only get to a water point within walking distance of their household, really within carrying distance of a 20 liter bucket.  Areas where there are more than 1 water point per 250 people don’t really provide any benefit to people living in areas where there are less than 1 water point per 250 people.  The extra water points don’t benefit anyone other than the people living in those areas.

The best available data to describe the distribution of people in rural areas of Malawi was collected in 1998 by the National Statistical Office.  In that year, the National Statistical Office carried out a census of all households in the country.  Individual surveyors were each given a census track, or “enumeration area”, and they collected population and other data through interviews at each household.  The boundaries of each enumeration areas are mostly roads, rivers, forest boundaries, or other natural or man-made features.  On average, each enumeration area has a population of around 1000 people, although the range is from zero in national parks and other reserve areas, to as many as 5,000 thousand in urban areas. 

Since the field surveys of the ICWPs included getting the east and north coordinates using a hand-held GPS unit, the location of each water point is available in the ICWP database.  Using the a map for the enumeration areas, and a map of the ICWPs, it is possible to put them together and calculate the average number of people per water point in each enumeration area. 

Excluding the 5 districts where no data is in the inventory, and urban areas, there are 6,434 enumeration area so the problem of averaging areas is reduced.  Assuming that the people can access water points anywhere within their enumeration areas, the number of water points needed to provide safe water in each area can be calculated. 

Considering just the set of 470 enumeration areas with 0 ICWPs, the total population in all of these areas is 471,346 people. To supply 1 water point per 250 people would require 1,885 water points.  But that is an average for all 470 enumeration areas but really they can not share water points.  Calculating for all 470 areas separately, it would require either 1,657 or 2,126 water points. 1,657 if you round down to an integer the number of water points needed for each area, and 2,126 to achieve full coverage of at least 1 water point per 250 people. Rounding down means not providing a water point for “extra” people, and full coverage mean installing an “extra” water point for any number of people from 1 to 249.  Considering all the enumeration areas with less than 1 point per 250 people, 3,920 areas out of 6,434, the total additional water points needed is 7,200 or 10,644 depending on how the calculation is made.
 
Now, no matter how it is calculated, there is a need for additional water points.  Lots of them.

These could be provided by either fixing existing ICWPs, or upgrading unprotected sources like shallow wells or springs, or installing new water points.  But the inventory doesn’t have the information to say what is wrong with each ICWP.  Some just need a minor repair, for example a washer or some other quick wearing part.  Others have been abandoned and dismantled for some reason, or they might never have really provided water.  Some boreholes only have water during the rainy season. A large number of non-functioning water points are taps where the whole scheme, from intake to tap, has broken or been abandoned and the only way to fix anything is to repair the whole scheme. Without the information about what is wrong with each non-functioning water point it is impossible to figure out what is the best way to provide the necessary number of points. 

At US$5,000 per completed borehole, even fixing a few broken pumps rather than building new ones can save a lot of money. 

It is obvious that improving the data collected for each water point would really improve the ability to use the inventory for planning. 

And that is where I come in.

I’m working with people in the Ministry of Irrigation and Water Development, and a guy from the WSSCC who has been involved with water point mapping in Malawi for 3 or 4 years, and some other NGO people.  Together we rewrote the field survey questionnaire.  We added questions trying to find which water points have problems that can be corrected, and which should be considered non-repairable, and what kinds of problems communities are having with both functioning and non-functioning water points, and what kind of training they have for maintaining them, and who they turn to when problems arise.

The British Department for International Development (DFID), is funding an NGO to complete the water point mapping in the remaining 4 districts.  This project started in September of this year, and field work is going to finish in January.  The field surveyors are using the new questionnaire, and although the information from the new survey is only starting to be entered into the database, it is already proving to be much more interesting, and I hope more useful, than the existing data. 

For example, with only 750 or so sites entered, we already have 2 boreholes that were installed but never used because they were built too close to graveyards, lots of taps that were vandalized because nearby communities were jealous that these ones worked and theirs didn’t, and more than a few water points that are not working because the person who was supposed to come and repair it stole the expensive parts. (There are 3-meter long stainless steel rods that go down deep boreholes that cost somewhere around US$25 each and if one or two go missing no one is going to know about it until the water level drops below the end of the rods, where the plunger is mounted.)

I am spending most of my time working on the database that holds the inventory.  I had to make lots of changes to accept all the changes we made in the field data questionnaire.  Plus, I am trying to build in reports so each district can create planning documents right from the database. Hopefully I can figure out how to make it easy enough to install and use so that each district will be able to maintain their own copy and keep it up to date.  So, I expect I will be providing training after the beginning of next year.

As a proof-of-concept, I am trying right now to use the database, and the enumeration area map, to create a report for one of the areas where other people in the Clinton Foundation area are working.  In addition to projects like mine, working with the central government on infrastructure type projects, we are working in 3 “rural growth centers”, supporting the local government in development efforts, one sector of which is water and sanitation.  I am taking all the data in one of the rural growth centers, Neno, which is about one and a half hours from where I live in Blantyre, and trying to come up with a report and maps that we can use to plan work there. 

If I can get it right in one area, it should carry over to other areas well.

As to be expected though, it is more complicated than I thought. For one, Neno happens to be one of the districts where the water point survey has been done twice. Once in 2004 and again just this year.  So I made all sorts of reports and things, and then went to do them again to use the more recent data. But the two sets of data have inconsistencies that I now have to sort out.  Water points missing from the second survey that must have just been missed, ones that are supposed to be 5 or 6 years old that were not in the first survey so they were missed earlier.  And both sets just say a point is functioning or not functioning, so I have no information about repairable water points.

Plus, the Neno District Assembly already has a list of where they think water points are needed, supposedly assembled from requests from communities in the areas.  Some of the requested sites are in enumeration areas where the number of water points is way above 1 per 250, while there are no requests from other areas where there are no water points and lots of people.  And there are areas where there are lots of people and no roads, so drilling a borehole is not possible because you can’t get a drill rig in there.  And there are areas where there are lots of people and no groundwater. So even if you could drill a borehole there is no groundwater. Which is why there are already existing water points that are not functioning. And probably never did and never will. So an alternative is needed.

So I am again looking out the window.  I think what I have written explains what I’m doing, and covers some of the why.  It is interesting to me, and it seems to have a lot of potential.  I guess I will find out if it is useful.  I’ll let you know how things develop.  Take care, and be careful of asking a simple question that might get you a 5-page answer.

Jimmy