The phosphorus crisis − 3

More thoughts on the P crisis. Each one of us excretes about a gram of P per day in our urine (according to The Main Features of Ecological Sanitation, Ecosanres Factsheet #2, 2005), so roughly 365 g P per person per year. There are some 2.6 billion people without sanitation and about half of these live in rural areas. If all these 1.3 billion rural people used EcoSan toilets, then they would urinate a total of 1.3 billion × 365 g P per year which they'd use as fertilizer. That’s around 475,000 tonnes P per year and, with DAP [chemical formula (NH4)2HPO4] costing, say, USD 1000 per tonne − which is equivalent to P costing USD 4260 per tonne, the phosphorus in all this urine is worth just over USD 2 billion per year. And then there’s the value of the nitrogen and potassium in it as well …

So, is this a lot of money? Well, the estimated spending required in developing countries on new coverage to meet the MDG sanitation target (see Global costs of attaining the Millennium Development Goal for water supply and sanitation by Guy Hutton and Jamie Bartram, Bulletin of the World Health Organization, January 2008) is USD 142 billions or some USD 14 billions per year during 2005−2014, so the value of the P in the urine of 1.3 billion people is approx. 14% of this.

So is it all worth it? Very probably − but this is not a reason to be fanatical about sanitation and the impending P crisis. As I keep saying, if poor rural farmers want to use the nutrients in their excreta to increase their incomes, then fine − and we should help them to do so at low cost and at minimal risk to their health and the health of those who eat their crops (and this is where Dr Peter Morgan’s book Toilets That Make Compost is so very useful). And if this nutrient use helps in any way at all to postpone the P crisis by a few years, even a decade or two, then that's all to the good.

The phosphorus crisis − 2

A quick update on the price of DAP (diammonium phosphate). Latest prices are given in The Market. The issue of 13 March 2008 quoted a DAP price of USD 1000−1010 per tonne (Tampa, fob). Even higher was Moroccan DAP, sold at USD 1,110 per tonne fob, an increase of USD 140 per tonne in a single week. The value of urine is clearly on the up and up!

The phosphorus crisis

I received an e-mail the other day from someone in the Water and Sanitation Management unit of IFAD (the International Fund for Agricultural Development) taking me to task for the entry on Phosphorus which appears right at the bottom of my webpage on EcoSan. The entry reads as follows: «‘EcoSanologists’ make much of the impending phosphorus crisis (e.g., Closing the Loop on Phosphorus, EcoSanRes Fact Sheet #4, April 2005), but this has little to do with the periurban and rural poor in developing countries and it’s not really much of a reason why they should have EcoSan toilets − better for industrialized countries to stop using phosphorus in detergents (see, for example, Phosphates and Alternative Detergent Builders, European Commission, 2002).»

The e-mail said this was wrong for three reasons: (i) poor rural farmers don’t have access to chemical or organic fertilizers (in sub-Saharan Africa they used only 8 kg of fertilizer per ha in 2002/03, compared to 80 kg in South America, 98 kg in North America and 202 kg in East Asia); (ii) the price of DAP (diammonium phosphate) is rocketing − now USD 750 per tonne, up from USD 240 in 2007; and (iii) in Mauritania (as an example) urine is now worth about €0.25 per 20 litres. I was also given a link to “current IFAD activities on sustainable sanitation” (here) but this turned out to be a record of a seminar held at IFAD Headquarters in Rome on 29 January 2008 with participants from IFAD and the Stockholm Environment Institute on “Safe and sustainable livelihoods in agricultural communities: Optimizing the recycling of human waste” − the usual EcoSan “gospel”!

Well, as I said in the blog of 16 January, «I may not think much of EcoSan in periurban areas but it’s certainly more than OK in rural areas if that what the users want − if you’re really interested in rural EcoSan, then read the excellent Toilets that make compost: Low-cost, sanitary toilets that produce valuable compost for crops in an African context by Dr Peter Morgan.» The key words here, to me at least, are “if that’s what the users want”.

Poor rural farmers can be responsibly encouraged to have EcoSan systems, but the reason for this should be for them to use the nutrients in human wastes and the water in greywater, so that they grow more food and save money by not buying so much fertilizer − rather than as a contribution to solving the impending global P crisis (which isn’t of their making).

So I think I’m right, but I should update my EcoSan webpage make it all clearer.

Hardware and software

I’m often struck these days by something rather curious: that in the mid-1970s, when John Kalbermatten, then Senior Water and Wastes Advisor at the World Bank, started the two-year Bank-funded research project on low-cost sanitation, we found out quite a lot on sanitation technologies, the so-called ‘hardware’ (and this was the ‘easy’ part), but we struggled with the ‘software’ − how to interact with communities to take their views into account − but by 1978 the Kalbermatten model was developed (and published here). Eventually, of course, community participation developed into an ‘industry’ of its own and became accepted as part and parcel of sanitation planning. So today there’s a lot of knowledge and experience about this: PHAST, for example, and Community-driven Development. [Probably the best document on urban sanitation planning is GHK’s 2002 report Effective Strategic Planning for Urban Sanitation Services: Fundamentals of Good Practice − if you want a really good short introduction to urban sanitation planning, then read Section 3 of this report A Strategic Approach towards Sanitation Planning.]

But, in contrast, so it seems to me, the hardware is less well understood today than the software − most engineers and planners just don’t know, never mind fully understand, the whole range of sanitation technologies available. This situation is particularly acute at the local level, but often not much better at national level. There’s No substitute for knowledge!

So how did we get into this situation, the reverse of the 1970s, that the hardware has fallen behind the software? Graham Alabaster and I talked about this at AfricaSan2008, but we couldn’t really explain this ‘failure of engineering’ − or, perhaps, this ‘failure of engineering education’. But it’s something that has to be rectified and IYS2008 is a good time to start − and a good place to start is here.

Alternating twin-pit pour-flush toilets vs. simplified sewerage

India used to have (maybe still has) a policy of no sewerage in cities with a population less than one lakh (100,000); instead alternating twin-pit pour-flush (PF) toilets were to be used (details here). Is this now a policy to stay with? The huge success of ‘Slum Networking’ in India, developed by Himanshu Parikh, would suggest not. Slum networking (details here and here) is slum upgrading focussed on WatSan improvements: ‘The present cost of providing complete physical infrastructure is about INR 16000 per family (i.e., GBP 200). This includes house-to-house sewerage, water supply, roads, stormwater drainage, receptacles for solid waste, and landscaping. Out of this the sewerage component is INR 6000 (INR 4000 for the network and INR 2000 for wastewater treatment per family of 5 people)’ (email from Himanshu Parikh, September 2006); if a ‘pucca’ PF toilet has to be provided as well, this adds a further INR 4000 (email from HP, April 2007). The sewerage is essentially Brazilian-style simplified sewerage (but HP uses the Colebrook-White equation, rather than Manning’s). The Indian Sewerage and Sewage Treatment Manual (2nd ed., Ministry of Urban Development, New Delhi, 1993) does at least mention simplified sewerage (which it calls ‘shallow’ sewerage), but clearly simplified or ‘slum-networking’ sewerage needs to be much more widely understood in India (and elsewhere) than it is at present. The late A. K. Roy, the doyen of low-cost sanitation in India, once said to me that he doubted Brazilian-style simplified sewerage would work in India as “our diet is very different from Brazilians’” − true, but not so different from Sri Lankans’ and there are about 40 simplified sewerage schemes working well in Sri Lanka, with the earliest dating from the mid-1980s.

It wouldn’t take an Indian student long to work out comparative costs of alternating twin-pit PF toilets and simplified sewerage − it’s basically the construction cost of a flow-diversion box plus two 1.5-m deep brick-lined leach pits vs. the cost per household of installing 100-mm diameter sewers (one simple junction box and something like 3−4 m of 100-mm dia. sewer laid at a depth-to-invert of ~350 mm and a gradient of 1 in 200). You could vary the housing density and you’d then get a graph rather like this one to show when sewers become cheaper than on-site sanitation. This sort of work needs to be done in many parts of the developing world, especially in Africa and Asia − and the simplified sewerage costs would be useful in any comparison with periurban EcoSan systems!

New paradigm for periurban WatSan

The paper A new paradigm for low-cost urban water supplies and sanitation in developing countries (Water Policy 10 (2), 119−129), by myself and Dr Graham Alabaster of UN-Habitat, is now available on-line − pdf here, and more info. here. The New Paradigm is very simply stated: Supply water and sanitation to groups of households, not individual households. Why? Because it’s much cheaper − and likely to be one of the main ways the MDG sanitation target can be met.

VIPs or UD-VIVs?

I think this is now an important question. Should we go for VIP latrines (single-pit or alternating twin-pit version) or for UD-VIV latrines? [Just in case anyone doesn’t know these acronyms: VIP = ventilated improved pit, UD-VIV = urine-diverting ventilated improved vault − these are always alternating twin-vault systems, see blog of 16 January; info. on VIPs here and very well illustrated info. on UD-VIVs here.]

The problem with VIP latrines is that desludging is in practice rather difficult − not technically (there’s the excellent Vacutug, for instance), not organizationally (eThekwini Water, for example, has an on-going VIP latrine emptying programme − but eThekwini Water may not be typical: it’s a highly organized, well-staffed, and financially strong institution), but rather because local governments generally don’t offer a pit-emptying service and private operators can be expensive − so pits are often not emptied or at best only partially emptied. It shouldn’t really be a problem, but in practice it’s a major one.

It’s even a problem in some rural areas where there’s space to dig a new pit after 10 years or so. In South Africa, for example, tens of thousands of rural households were given a VIP latrine by the government (as part of the post-1992 social reconstruction of the country), and many householders think it’s therefore the government’s responsibility to desludge them, but the government doesn’t agree! (South Africa has very few eThekwini Waters.)

With the UD-VIV latrine things are very different. The small above-ground vaults can easily be emptied by the householder using a long-handled shovel, and the material removed simply buried on-site. This process takes no more than 30−40 minutes a year and costs nothing.

Of course, the eThekwini-style UD-VIV is not perfect − well, it’s almost perfect in eThekwini. But in other parts of the developing world (even in South Africa, outside of the four big cities) it’s less than perfect. For a start the cost of the eThekwini UD-VIV in 2003 was ZAR 2900 (then equivalent to USD 370 / EUR 230) + VAT (details here − but why is VAT charged on a latrine? Surely it should be either exempt or zero-rated). Of course, in eThekwini cost is irrelevant to the householders as they get their UD-VIVs free as they’re paid for by eThekwini Water and the government. But this is not a model that can be used elsewhere as in general governments can’t afford this level of subsidy.

The technical specification of the eThekwini UD-VIVs is pretty high (details, including a drawing, here; note that in this document UD-VIVs are called VIDPs with urine diversion − VIDP = ventilated improved double pit) − possibly too high. There would certainly seem to be the opportunity to reduce costs by adopting a lower spec. for, for example, the superstructure (which could be built by the householders − after all, in most rural areas of the developing world people build their own houses).

Is a separate urinal necessary? Could be that men and boys don’t aim properly and their urine enters the faeces-only compartment ? But males do have a high degree of ‘directionality’ and they should be encouraged to use it − a role for hygiene education here!

So, while I think on balance that UD-VIVs are ‘better’ than VIPs, there is some developmental work to be done on them to make them more widely applicable in the developing world − but this really shouldn’t be difficult at all.

Then there’s the added bonus of UD-VIVs: they can easily be converted to ‘proper’ EcoSan systems. Personally I think it’s better for most communities to operate UD-VIVs just as latrines for something like 2−3 years, after which they can be shown how they can be operated in EcoSan mode and, if they like that, then fine. It’s much better, at least in my opinion, to get communities to use latrines properly before exhorting them to “close the loop on sanitation” and delay the impending global phosphorus crisis by recycling their excreted nutrients.