QUT Darcy Lecture: Characterization of a Complex, Sole-Source Aquifer System in Benin, West Africa

Uploaded by TheQUTube on 07.11.2011

Good evening. We'd better just get started.
I'd like to welcome you all to this Darcy lecture tonight in Brisbane with Professor
Steve Silliman.
He's been doing the tour through Australia. You've had what, 6 talks?
This is the 7th one.
This is the 7th one. Often the speakers go to New Zealand but he has to go back to the
The Darcy lecture is funded primarily by the U.S. National Groundwater Association, so
there are some requirements that they want.
But as you know there is no such thing as a free lunch so...
Darcy lecture sponsors...
National Groundwater Association of the U.S. so they have a research and education foundation
which essentially covers the fares for the speaker.
QUT has put funds into this.
IAH, our Queensland branch has also supported the funding.
As most of you know I am president of the Queensland IAH branch.
Waste Solutions have also put in some funding.
The other one is AGE, which has also contributed which we appreciate very much.
And the National Centre for Groundwater Research and Training is also contributing.
We're QUT, one of the members of program number 1 which is based at University of New South
Wales, so I think that's that bit.
I think that's most of the things Stephen, so... Steve is from Notre Dame University
which is a very traditional university in the sort of North-Easternie bit of the U.S..
It’s in Indiana, a couple of hours drive from East of Chicago.
Near the lake. A private university.
But what's very interesting, he's a professor of civil engineering and geological sciences,
which is a match made in heaven, I think (laughter).
I don't know and we're looking at some sort of amalgamation here at QUT between the faculty
of Built Environment and Science and Technology so it's sort of interesting in that context,
but I think we'll pass it over to you Steve, thank you very much.
Whether that's a match made in heaven or the other place depends on the weekend, and who's
talking at the time.
Thanks for coming out tonight.
I know you're all giving up time at home or with family or whatever to come to this talk,
I appreciate it very much.
One of the first things that I do want to make very clear is I'm going to be talking
about work that I'm doing in West Africa, a country called Benin, but it's not the person
standing in front of you who is the centre of that project.
We really have a number of folks involved, many of which are from Benin.
Moussa Boukari is a hydrogeology faculty member in their department of Earth Sciences.
Nicaise Yalo is a geophysicist in that same department.
Both of them are absolutely instrumental in this project to move things forward, get their
students involved, get us out into the field.
Felix Azonsi, up until about 3 months ago when he retired, was the lead hydro geologist
in government water agency.
Felix, I got 26 years of experience past schooling, Felix is probably the most astounding hydro
geologist I have ever met.
He knows his country, he knows the people there, he knows how to get wells in but more
importantly he knows how to maintain those wells, get people to do that, maintenance,
just a fascinating guy and has been critical to this project.
We've had a number of graduate students from Notre Dame involved, a number undergraduate
students at Notre Dame, as well as a number of graduate students at their university (Universite
And so there's quite a team that's behind this.
Please don't look at this as - when I talk about "we" or "I" etc., it's really these
guys that I'm talking about being involved in this.
As Malcolm said this is sponsor in part by the National Groundwater Association.
The idea is to get some people out into the world to talk about groundwater, get people
excited about groundwater.
This is the 25th year of it, and obviously it honours Henry Darcy.
When this series started we went to about 25 locations.
They have nudged us up just a touch since then, so this year between January and December,
I will hit approximately 60 locations in the U.S., Europe, China, Africa, and of course
Let me jump in and very rapidly, so we can get your responses at the end, and by the
way at the end of this I would love criticism and suggestions as much as I would love any
kind of questions you might have, so if you've got suggestions, you got critiques etc., please
jot them down and bring them up at the end.
Benin is a country in West Africa. Countries you might recognise immediately.
Nigeria is merely to the East of it. Ghana, two countries to the West of it.
Ivory Coast or Cote d'Ivoire is over a couple of countries to the West.
A very very interesting country, yes we will go forward... there we go... it’s very interesting
country and I began working in this country way back in 1998 so we've been there for quite
a while.
I will freely admit and one of the things that I hope you get out of this talk, well
actually two things kind of combined.
Number 1: the guy in front of you is not all that intelligent.
I think you'll pick up on that as we go forward if you haven't already but number 2: Getting
a successful group of projects going really requires collaboration with folks in that
country, closely working with them, listening to them very carefully.
And my favourite story on that: these 2 "X"s.
We did go out and drill some wells but I was, and maybe I am still the obnoxious American,
but I was definitely the obnoxious American back then, showed up at the drill rig and
said "I'm here to work with some small villages and drill wells".
We drilled one in '98, one in '2000. Both are decent wells but shallow, one has never
been used because of political purposes, and I spent about $25,000 U.S. to drill those
2 wells.
I got back there in 2003, started some work with Professor Boukari.
I got back there in 2003 without a lot going on, but I had another $25,000 in my pocket,
and Felix Azonsi, that was the gentleman who works for the national water agency.
He has this wonderful, very tight office, you kinda shimmy past each other.
You sit down in his chair and the arms always fall off and clatter across the floor.
So the first week I was in the country in 2003 I went into his office and sat down.
Felix wasn't in his office at that time, but he walked in and he shimmies behind me and
he slaps me, quite hard, up the backside of my head.
"Well gee thanks Felix."
And back here my French was limited to what I learnt in high school, so it was very very
I learnt a little bit of French by here, but Felix speaks with a lisp so he's really hard
to understand, but the word "stupid" actually comes across pretty well from French to English.
Well, Felix what are we talking about?
He said "Well you can use another 20 - 25 thousand dollars to try and drill 2 more wells,
probably won't have any success on either of them, or you can work with us instead of
against us, and we've got this large drilling project, and you could use your money to help
us get out to some of these remote villages, get them ready for drilling."
"I actually have the money - ", said Felix,
" - to drill the wells once the villages are prepared.
Could you get together with your NGO and actually go prepare these villages.
With that you could probably drill about 20 wells instead of 2 wells."
I don't know about you folks - I jumped for that latter one, and I'm actually pretty proud
of the fact.
We've gone back a couple of times with them now, and we are completing somewhere around
our 32nd well in that country.
That's not what I'm proud of, what I'm proud of is the fact that the guy you see standing
in front of you has not been onsite for drilling any of those wells.
We've built such a trust-base that I can actually work with these folks, trust them completely,
and they're doing it better than I could ever do it, and the white person showing up in
this part of Africa would actually change the project to being a white person's project,
and would take the ownership away from the village.
Now what happened because of that? Why tell that story?
Well again, notice there's very little going on up until 2003.
Once Felix Azonsi felt that he was a close enough friend and colleague that he could
tell me I was stupid, he also felt he could tell me what projects he thought would be
beneficial in Benin.
And so what you see is a very rapid increase in the number of projects that are being pursued
in Benin, and I will tell you, out of those projects, every single one of them except
this K-12 education project, every single one of these was the idea of either Moussa
Boukari or Felix Azonsi.
None of those ideas came from the U.S. side.
And so it’s a lot of fun, and I'm gonna talk about a coastal project today.
If folks want to know about some of these other projects, feel free to contact me or
talk to me afterwards.
So let me bring you into Southern Benin. We're down along the coast.
So here's the Atlantic Ocean.
This is going to come into it very clearly in just a moment.
This is Lake Nokoue. It is predominantly a saltwater lake through most of the year.
It is connected to the ocean by a man-made channel.
The white area down here is Cotonou and Abomey Calavi, Benin.
Cotonou is the largest population centre of Benin.
Estimates on the population is between 1.5 and 2 million people, so that's what we got
going on down there.
Their primary water supply in the city is a well field that's up to the west of the
lake and they pump that water down into the city.
There's also a lot of small hand-dug wells and things along that line.
A very very good water supply by the way - it’s quite healthy.
If we look in the middle of the city what are some of the hydrologic challenges that
we run into?
I'll show you the geology in just a moment but the first things that we note is that
they've got pavement almost everywhere.
And it turns out that this is beach sands, almost everywhere, so we have dramatically
changed the recharge and infiltration throughout this region, in most areas.
But there's going to be some areas, and you'll see some areas in a few moments, that are
wetlands, and they have not built onto the wetlands because the material simply doesn't
support that type of structure.
Unfortunately those wetlands then become very much a waste disposal area.
There is a little bit of centralised solid waste disposal in Benin, but not a lot. Here's
my students and Benin students together.
We are doing some sampling and coring in here and we hit everything under the sun out there.
And so we've got potential for tremendous anthropogenic contamination all through this
area of development.
And then finally along this channel and along the southern shore of Lake Nokoue. That becomes
a garbage dump.
So this is water in the channel, this picture was taken about right there. What you see
is some sheep actually eating off the top of the garbage pile.
That clearly leads to substantial contamination, but of the lake and into the groundwater system.
So in the city we have those challenges. Let's move a little bit further to the west. And
we change dramatically within about 15 kilometres.
We get away from the high development, we get into much more rural areas, a lot of lagoonal/wetland
type of property out there.
Some of these areas if you get far enough West, these will actually dry out during the
dry season, and they will harvest salt from it.
The people are no longer on centralised water supplies. Most of them are on shallow hand-dug
well or other roof-top collection-type systems.
And one of the interesting things - I'm not a surface water hydrologist - but I do play
with it a little bit.
One of the interesting things is this picture: these lagoonal deposits are parallel to the
shoreline, and there's quite an active flow in those, except for, you'll see it later,
there a road down south and a road about right here, and the people here want to talk to
the folks up there.
To do that, they want roads between those two.
To do those roads, what they do is bring in large trucks with sand, and they dump the
sand in and they compact that so they can drive north-south.
Effectively what they've done is put in a series of earthen dams across these lagoonal
The third area that's going to be quite interesting to us, is going to be out on the lake.
First, I love the birds out there, so I had to give them a little bit of credit for keeping
me coming back.
But the lake is a very very interesting lake and I'm gonna start with this picture which
seems to show nothing.
What it really shows is the intelligence of the person standing in front of you, so yes
it does show nothing.
We were out there, we were actually trying to measure vertical temperature gradients
in the sediments.
Trying to look at recharge.
And we had pounded our device down into the ground.
It had a nice little drill-point on it.
Tried to pull it back and we had popped into a clay layer that had swelled in on us.
And so we couldn't get it back and we were in a very narrow boat, looking something like
this, and so we can't really heave on one side, without someone sort of laying out the
other side to make sure we don't tip over.
So we're scratching our heads, trying to figure out what on earth we're going to do when this
gentleman comes up in a boat like that and says "Do you need help?".
"Yeah we got equipment stuck."
He said, "have you got a few coins I think I could probably solve this for you?"
I said "yeah I'd be glad to pay you if you can get our equipment back".
He says "no problem". He jumps out of the boat, and the water level is only to here.
It wasn't real hard to get it back once you'd figured that one out.
As you'll see the Bathymetry just a little bit later, this is a very shallow lake.
Even in, we were there last month, which was about a month into the rainy season.
Even in that season, the deepest spot in most of the lake was 2.2 metres.
A lot of it's less than a 1/2 metre. So very shallow lake. Very salty during most of the
Very close to ocean water salinity. An interesting part of that is an area called Ganvie.
Right up here and it actually has a little river that wonders off the top of here.
Ganvie in the 1600s had a population move out onto the lake as a way to protect them
from tribal warfare and there's a whole religious tradition behind it.
But they moved out there back then and a lot of those people still live there.
They live on stilts, you know.
They don't have cyclones, but this lake does go up and down, but they live on stilts.
No real centralised facilities.
To get their water, the government has come in on the few plots of land.
It has built some deep wells, either between 90 and 120 metres down.
Those are freshwater wells.
So these folks will come up in their boats and you notice the water jugs all over the
place here.
This gentleman will control the hose and he'll actually fill up all those water-jugs.
That's where they get their fresh water from. Now why is this important to us, well it's
gonna be important to us for a few reasons, cause we're gonna worry about this area down
here being a recharge zone.
And I still wouldn't worry about these folks too much if there were maybe 300 of them living
out there.
However, the population estimates that we have right now for Ganvie and the villages
north of it are somewhere between 35,000 and 70,000 people living on that lake.
So we have a substantial city living on that lake, with substantial nutrient inputs.
Geology, I'm a civil engineer so I'm going to go through this very quickly, before you
guys can tear me apart.
Again we got a strong-short current going from West to East so we have a lot of sand
deposits in these beach-rich barriers.
Between those we have those lagoonal deposits. That goes up to a river called the Dunia River.
North of that you've got some very very old geology.
Essentially you got the African shield north of you here, and you got sedimentary sequences
in this central area.
You have very old clastic sediments; so the gravels, the sands, the clays, etc..
Very well inter-bedded, very dis-continuous, relatively complex geology.
Alright. Why are we worried about the water supply down here?
Well we were invited to come down and take a look.
Why did they invite us to take a look?
I apologise for some of the small numbers here but, this is year vs pumping-rate.
This is in million gallons per day, so, multiply that by about 4 if you want to get it in your
million litres per day.
1950, in 1960 things were pretty stable.
They were pumping less than one million gallons per day.
The migration to the coast started in the late 60s and 70s, and notice what's happening
to the pumping rate.
It went from less than 1 million gallon per day, (but) til 2008, is the last point on
this slide, we're up to about 13.5.
As of 2011 we crossed 14 million gallons per day. So we're pumping this groundwater system
pretty hard.
Does that cause any problems? Well this plot on the right is chloride vs year.
1992 through to 2008. And there's a number of wells where we saw this pattern.
Relatively high quality water, less than 50 kilograms per litre, early days, But as this
pumping rate went up, we started seeing some dramatic increases.
Notice this is only over 4 years. Some dramatic increases in chloride.
Well you might say that "well that only goes up to around 280 grams per litre, that's not
high enough to really destroy a water supply".
The reason it stops at about 280 grams per litre, once they get into this region they
actually turn off those wells.
So they're allowing the salinity to move farther into the system, and they're not working with
these wells anymore.
We actually just went back last month, I don't have the numbers yet, and resampled those
wells to see where they're at now.
So we got some concerns.
We're concerned that salinity is getting into this system, we don't know exactly where it's
coming from.
Professor Boukari actually created an initial model with some friends up in Europe.
This is a finite difference grid.
Here's the Atlantic Coast.
Here is Lake Nokoue.
The grid is slanted because north is actually straight up and down on this grid.
On this slide the grid is off of north just a bit.
And what you're looking at is a series of wells here.
There's also a couple of wells up there.
He actually calculated the path lines to the wells.
This outer yellow line is effectively the capture zone of his wells.
That's the area contributing recharge to his wells.
Couple things you notice on this.
That picture of Cotonou with that wetland was over about here, so that's not included
in its capture zone.
It's not getting very close to the coastline down here.
It's also, if you actually follow the lines, see this yellow line going across Lake Nokoue
- these flow lines actually do not come up underneath Lake Nokoue.
So there was not a strong indication that Lake Nokoue was getting recharge.
We came back in and did a second model which I'll describe in just a second.
More recently Professor Boukari has come back in and done a transient model.
I do not trust the first or third models and Professor Boukari and I are in agreement on
Why? Because he went in and he very much did layer-cake geology.
Undefined aquifer, under that a continuous defined layer, under that a continuous defined
aquifer, etc..
He assumed recharge values along the coast.
He assumed constant recharge, in fact a constant head in Lake Nokoue.
Within his grid that Lake Nokoue also went down 17 metres, so it went much deeper in
the system than it actually goes down.
For his trans-positivities and restorativities he did not have pump-tests so he went to a
Fetter's book (groundwater text book) and picked off values from his book.
And so we've got a model that is not precise, let's just leave it at that. And most of those
mistakes were repeated in the third one.
What did we do?
Well with Professor Boukari and his students, we went back starting in around 2006, we went
in and there's actually about 1300 bore-holes in this region.
Unfortunately there's only about 10 down here.
We went in and did the geologic interpretation.
Try to get a much better model of this.
We incorporated that geological interpretation into the mod-flow model and then we said “well,
what happens if we adjust the pumping rate?
He'd assumed the 2002 pumping rate.
Let’s move it on to 2010. Let's play with that assumed recharge.
Let's play with the boundary conditions up top and down bottom.
Let's go ahead and play with the lake sediment properties and let's play a bit with the aquifer
And when we do that with very very reasonable numbers, we can change the capture zone from
what was yellow before is now blue to this outer pink and a couple of things you'll notice
with that.
We have moved under Cotonou, Benin.
We are moving closer to the coast.
We are moving out into Lake Nokoue.
Now, am I going to claim that the pink is better than the blue?
Absolutely not.
We don't have the field information that we require for that.
But what it shows us is that this model is extremely sensitive to a combination of parameters
out there and we don't have the correct parameters yet, so we can't really guide the government
yet in how to protect those wells.
So what are we gonna do?
We'd like to go out and characterise that system.
We'd like to be able to hand the government a beautiful model that is heavily constrained
and gives them a good idea of how to manage this system.
Our problem is that we are trying to do that under a number of constraints.
Huge variability of infrastructure: think about the picture you saw in downtown Cotonou
- everything paved, and the open area to the west.
So if I'm in Cotonou I can drive anywhere I want to.
I can't always do that out west. But if I go into Cotonou and I want to put out a geophysical
line that's 300 metres long - it's going to be really tough to do it, whereas out-west
I can.
So we've got this high variability in our site access and our ability to get to places
and do our measurements in the field.
Very limited in-country capacity to run water-quality samples for us, and so we've had to bring
most of our work back to the United States.
Very limited budget and field season, particularly for the U.S. participants.
The vast majority of this work is funded by a U.S. foundation, and I cannot allow my students
over in that country for more than about 8 weeks at a time.
It's just not possible for me.
When we got over there, there was very little in the way of in-country expertise in the
field methods we were using, and substantial equipment limitations, both in terms of importing
it and in terms of size and cost.
So we're gonna try and go out and characterise this system as well as we can.
Over the last 5 years extending into the next 3 years we just got a little bit more money.
We're gonna try and use appropriate field methods. We're gonna have to realise that
it is not the United States.
It's not Australia. We're working in a very different type of environment that has some
physical and cultural limitations.
We're gonna try and guide that by our numerical analysis and what the model tells us as to
where the sensitivities are.
And finally we're gonna identify those 3 zones we talked about earlier: the region right
underneath the well-field, the region along the coast including the city and Lake Nokoue.
So let's jump in and talk about what we're doing in those areas.
First area right underneath the well field - we're really doing basics at this point
in time.
We don't have good pump-tests.
We need to get some good pump-tests to get trans-positivities, restorativities, specific
yields etc.
So we're just going out with pressure transducers, this is one from about 3 years ago.
We're now using some of the Somerset devices, so a lot smaller and easier to work with.
Put these down the abandoned wells or piezometers and then there's a well back there - that's
one of their active production wells.
Use our data loggers.
Our very very sophisticated field sealing devices.
Those are buckets from down the street.
Lots of duct tape and electrical tape, and try to run a pump-test.
Now I think that there are quite a few consultants in this room.
A lot of you probably play with groundwater quite a bit of time.
If I told you that a marvellous full-term pump test here was 3 hours long, I suspect
that you'd laugh me out of the room.
The problem is in Benin all of their production wells are going 24/7.
So when you go up and you tell them that you want a 24, 48, or 72 hour pump test, they
look at you like you are from some other planet.
They cannot do that.
Even in this case, this was a brand new production well, they were just doing the hydraulic testing
on it.
We had an abandoned well over here, and this well was right over there.
Even in that case they said, "why don't we do it for like 20 or 30 minutes?”
So we had to argue with them. We started at 72 hours, and we wound up at, you barter for
everything over there, we bartered in the end for 3 hours.
That's the maximum we could get out of them.
So we take what we can get, so we ran our 3 hours, we run that through and you get some
initial parameters.
Yes we can run it, but we're not seeing whether we have any boundaries or leakage etc., that's
going to be a real problem.
This year we tried something a little different.
We put a pressure transducer in one of the wells, and then we had them turn on and off
various wells around the region.
So we could get various radii, from less than 10 metres out to 700 metres, but once again
they are unwilling to turn them off for more than 3 hours.
So you see that we have extraordinarily minimal data when we get about more than 200metres
away from our wells.
This is something we're working on, but probably the thing that was most frustrating.
I return to this picture, this is where we put our pressure transducer, and we talked
to the water company.
He said that he will turn off the well back there for 3 hours. So we came back, 3 days
By the way, we don't know when they start and stop, so we set these at 5 second intervals,
so we can get close to catching the beginning of the pumping period.
We came back 3 days later and his assistant came out and he said "I'm very sorry. His
daughter passed away immediately after you left.
He had to go up village, he never got a chance to turn off the well.
And so I was assuming that I would get something like this, with a little bit of noise in it
What did we get out of this well where the pumping well right next to it never turned
on or off?
We got that.
Well that well did turn on and turn off several different times during that 72 hour period.
He argued with us.
He said it absolutely did not.
For those who have worked in more rural areas, you can probably guess where I'm gonna go
with this one. It took us about 2 1/2 days to find this - we wound up in the electric
They're doing rolling brown-outs and grey-outs. They're not telling the water company.
The water company never knew that they were doing it. And so this is a real challenge
for us.
We ran into it this year as well.
They were gonna turn of the well for us and he said "yes I did turn it off", and I looked
at the data and it was absolutely flat through that period, but it was flat at the maximum
water level.
All the wells were off when he actually turned the switch off on this well, but there was
no power going to the grid at the time, so he actually didn't turn off anything.
So standard pump tests are going to be very valuable to us, because the historical pump
tests are just horrid.
And there's minimal current known, so we're gonna get somewhere around here - I'll talk
about this in just a second , that fourth bullet, but we got several real challenges.
We can't really do long-term pump tests. They're not going to allow us to turn things on and
off for than a few hours at a time. We've got this random variation in pumping.
We've got multiple wells. There's about 30 wells out there that are pumping.
We don't know the vertical connections between the aquifers, we don't know what kind of link,
what kind of connectivities we're gonna get.
We did go out this year and do some water-quality testing to see if we could pick up on that
- hopefully that will give us a little bit more information.
But there's an interesting idea - in the U.S. there's a lot of folks working with something
called hydraulic tomography.
You pump from different directions and look at the response.
What we're looking at is whether we could put pressure transducers down the abandoned
wells, and actually monitor the current from the other wells, and actually do hydraulic
tomography by accident here, by knowing when wells are going on and off.
So it’s something we're looking into because the power grid actually goes off at times
for 10, 12, 14 hours, so we get better pump-tests from the power company turning off the wells
than we do from the water agency.
Ok, let me switch down to the coast, and I'm gonna go through this one relatively quickly,
so I can get to the third one which is a little bit more interesting, but the coast is a very
complicated area.
Here's the south-west corner of Lake Nokoue.
You can see all the white is development, so we are paving an awful lot, except in the
wetlands. And then move south of there. Here's the coast of the Atlantic.
Here's the airport. I'm gonna refer to a couple of points here in just a moment, but we're
gonna be talking about somewhere less than a kilometre off the coast.
You move a little bit west, development goes down, you start getting the open surface water
in those lagoonal areas.
Get further west than that, you get these open lagoons, except where they've cut it
with roads, and so again, there's that southern road, and here's that northern road we've
been talking about a couple of times.
So we're gonna jump out there, and we'd like to know the distribution of salinity in the
southern coastline.
Now remember Professor Boukari assumed nice uniform recharge along the southern coast,
assuming that we had some kind of nice saline intrusion, a nice saline wedge coming in.
Well we can't get out here with a big drilling rig, so we decided to look for other approaches.
What we wound up with is a direct-push method that is manual.
That's about a 25kg hammer, and you simply just drive this, sounds like a posthole digger.
Once you get it in, you can get your cores.
You pull it out with a truck-jack, and you can go in and you can sample. It's got a screen
down the bottom of it.
And we did that, and fortunately for us the folks in Benin are very generous people.
These two young guys were watching Mary my student try to work one of these by herself
out in this particular part of the lagoon.
They jumped out and helped her. They've actually been working with us now for about 5 years.
Every time we show up, those 2 young men show up.
Kind of a neat thing. We were doing that all along the coast. What kinds of things do we
Well top left. You know how I mentioned that airport site?
This is the airport site, you're looking at depth down to about 5 metres, and chloride:
0 - 8000.
At the airport, everything down to 5 metres was fresh.
No salinity what-so-ever. You move 2 kilometres to the west and you get this profile, and
so at the same depth we're getting about 8000 mg per litre.
And so we know at least at the kilometre scale, we have quite a bit of variability.
How 'bout at a more local scale? This is where those 2 young men were working.
This is their profile that they got with Mary. This is the profile that Mary and I got earlier
that day.
These 2 curves are 20 metres apart.
This is a log scale in chloride, so 20 metres apart at 3 1/2 metres down, we got 3 1/2 orders
of magnitude variation in our chloride concentration.
We're beginning to suggest that this uniform model down south is perhaps not the most precise.
Went out there with resistivity.
Very quickly, an interesting aside, this gentleman is driving a moped down this path.
We thought it was a great place to go ahead and do geophysics.
So there's Dan, my grad student's laying out the geophysics line.
That looks like this, and we're gonna drive a little post into the ground and put electrical
current in there.
I put the post in the ground, which is about 30cm long.
Put it down about 2cm, turn around to get the hammer, turn around again and it was gone.
Where is it? This gentleman was standing behind me, and I turned to him and said "where did
it go?", and he just started to laugh.
These folks are absolute geniuses in making do with what they've got.
This path is actually sitting on 2 metres of water.
They've taken these reeds and they have woven a beautiful mat on-top of the water, and so
there's houses down here, houses there, and 2 metres of water under there.
So we knew right where my stake was, we just couldn't get to it. What did we get out of
that? Well between the resistivity and the direct sampling we see the following.
These are spots where we saw salinity at less than 10 metres. So we see it somewhere along
the coast. Interestingly we see some of it go up here towards Lake Nokoue.
The second set of points is where we did not see any salinity indicated by the geophysics
for 60 metres.
So, in here we've got spots that are within a couple of hundred metres of each other where
we're seeing salinity right at the surface.
Those are usually in those lagoonal deposits.
Right next to them, no indication of salinity down to a great depth.
We got some points that are somewhere between 30 and 60.
We've got some other points that were a little bit more complicated to interpret, but again
indicated at least 30 metres.
If you know what a salt water wedge is, and what it's supposed to look like, that is not
So what we've got is a very complicated coastal environment.
The good news is this doesn't irritate us too much because there was actually a resistivity
study done is 1993 that showed almost the exact same penetration, at least in the centre
Almost the exact same penetration of salinity in '93 as we saw last year. So it's not moving
rapidly, at least from the south.
So southern coast, complex geology, complex hydrogeology, nothing down there is uniform,
nothing down there is going to be simple.
Our direct-push method was pretty effective down to 10 or 15 metres.
We've gotten a lot of suggestions this year about how to get deeper than that.
Our resistivity got us down to about 60 metres.
I am a little bit frustrated.
I'd like to do that resistivity at various times during the year to see the Spring rains
etc. effect it.
I cannot leave my resistivity equipment in the country because of customs and duty problems.
We're trying to correct that in the new project, but there's some pretty strong evidence that
there is some shallow saltwater intrusion going on, and there's also quite a bit of
anthropogenic contamination down there - nitrates and hydrocarbons.
Ok, let me take you to the third part of the talk. Now I'm gonna jump on a little soap
box here for just a moment.
This gentleman is a fisherman, and he's out in the middle of the lake, and note these
structures right there, if you want to call them structures.
Now just a quick show of hands: how many people look at that and see incredibly well designed
engineering structures out there?
Well, not many hands went up on that one. Let's take a look at that from space.
We're talking about a kilometre by a little bit less than a kilometre here.
The dark areas are vegetation growing up through the lake.
That gentleman's boat was about right here - this is open lake water.
There's the structures you were looking at from space.
These guys are building absolutely beautiful spirals. What are these things? They're fish-traps.
I can walk you through it later if somebody's interested, but they have figured out the
direction and rate of flow in this lake, and they've been doing it for hundreds of years.
And so we tend to go in and say "We need to bring heavy science and engineering in".
Always talk to the locals and figure out what they actually know about the system before
you embarrass yourself and make a fool of yourself.
Ok, let’s go out onto the lake.
We would like to know if Lake Nokoue is recharging the groundwater system with saltwater.
Why might Lake Nokoue do that?
Well the historical data indicates this very shallow lake is saline, again very close to
the salinity of ocean water, about 9 - 10 months out of the year, and you'll see some
data on that in just a moment.
So we've decided that we want to go out and do some vertical permeability measurements,
we're going to do that with falling-head parameters.
Measure the bathymetry because there was not a very good bathymetry map before this.
Put in some piezometer nests so we can see what's going on the lake and the underlying
Do some salinity profiles in the lake, and really set up some long-term monitoring on
that lake.
What's that gonna require?
That's gonna require this gentleman, probably the most valuable member of my research team
at this point in time.
This gentleman is called a pirogue, he drives these boats.
This was that narrow boat I was talking about, where Andy, my rugby player, can't really
put full strength into pulling up a pipe on one side unless 4 of us are leaning off the
other side.
Otherwise this thing flips.
This guy knows his lake, he knows his people, he knows how to get around.
This is his boat. We're gonna do all of our work out there.
We're gonna do relatively simple types of measurements out there: falling-head permeameters,
bathymetry etc..
This is actually down on the channel and that will come into play in just a moment.
So here's the bathymetry. These data are probably about 40 days old about now, so we haven't
patch-word this yet.
But I think there's a pretty clear pattern here. The yellows are very shallow areas,
less than 0.4 metres.
It's not until you get to the magentas and the red that you get down to around 2 metres.
And so really, except for the centre part and maybe up in here in this lake, we're not
seeing any depth.
We are seeing a very deep part here in the mouth of the channel that is carried out,
but that point is only 15 metres wide.
So it's a very narrow channel.
There's a couple of rivers, a major one coming in there, and another one coming in up here,
which are a bit deeper, so they get down to 3 or 4 metres.
So that's the bathymetry, and I'll point out right now, notice we're not seeing any obvious
channel right here, and you're gonna understand why I just said that in just a moment.
Went out and tried to figure out how to sample this lake - I took an American viewpoint into
that, and said "well we're gonna go out there, and with the help of the pirogue we're gonna
talk to land owners.
Well what are land owners?
Those are those guys on the boats, they have those fish hatcheries out there.
I'm gonna put my equipment into their fish hatcheries and get their permission.
Everything is therefore gonna be safe, everything is gonna work.
We went out and put them in 5 locations, the multi-level piezometers, came back one week
later - absolutely every single one of them was gone, except the one we drove too deep
and it was below water level.
Where are they? Well we asked the fishermen where are they and they said "we took 'em
out". "Why did you take out my equipment? We talked to you about protecting it." "Oh
its protected, I know right where the pipe is, I just had other uses for the pipe."
"It wasn't for you fish."
Ok, so we had to step back, rethink, and talk to the pirogue, talk to the folks in-country,
and they said "That's not how you do it here."
"The way you do it here is you talk to the politicians, and if you can get the politician
to buy into the fact that you got a good project, he's gonna sell the people on it, and he's
gonna choose the most public place that he can possibly find to put your installation."
Where did we put it?
Well here is one of the villages just north of Ganvie, again very boat oriented community.
We're gonna go right across here, there's another boat landing right there.
And we're gonna put in multi-level piezometers.
What are those things gonna look like in the end?
You'll see 'em in just a moment, but let me tell ya first, we put these things in, put
them at 1, 2 and 3 metre depths.
PVC outer pipe to protect them. Put them in in 2010, and they are still there, not one
of them has disappeared.
Well, did we hide these things? Here's the edge of the boat dock. Here's our piezometer
Now why is it so high? The lake water level varies by that much during the season.
Not only have they respected this, but they've had to put wood around here because this has
become part of the dock, and they actually tie their boats up to our piezometer pipes.
And so everything's great, but we gotta learn how to do it. What do we get outta that? Distribution
of sediment hydraulic-conductivity.
Very very high around the edges where we have active flow with the rivers.
Gets quite low in the middle of the lake, where the fine sediments settle down in, very
much as we would expect.
How ‘bout the direction of the hydraulic gradient?
Now, I'm not showing the actual numbers here because I don't trust them completely at this
point in time, but to date, with the exception of one measurement one day out of about 150
measurements, everything indicates that we got higher hydraulic heads in the surface
than we have in the subsurface, so we have down-flow.
We have recharge going on.
How ‘bout water quality?
Well, this is where we tend to take a left and get really confused.
As I suggested that this lake is saltwater about 10 months out of the year, and turned
fresh for maybe a month or 2 during the rainy season.
What are you looking at? Here's months from February through to November.
This is from 2009 and 2010.
Here's chloride on a log scale, so here's 1 all the way up - there's 10,000 and there's
The black line, if you see red stars, the red stars are from 2009, but the black line's
from 2010. And what do you see.
The lake at each of these points tends to be very high salinity for a significant part
of the year.
The rains start up, I wish I had another point in here to prove this, but the rains start
up and very very rapidly flush out the lake, or at least that’s what it appears.
Well why does this confuse us? Well, here's our piezometer data for those same locations
one metre down.
With the exception of one little blip here, we're not seeing any response in our piezometer
We're not seeing them change.
Remember we've got recharge at all these points, so we supposed to, in theory, have down-flow.
We've got the lake going to very fresh water, in some cases down, well less than 10 milligrams
per litre and yet we're seeing nothing in our piezometers.
I should tell you that these lake samples were collected at about 15 centimetres down,
but with this shallow lake, with the heavy winds going across it, we never imagined that
we had salinity stratification in this lake.
But we decided that we better check that, so we went back this year to see if we could
find that out.
And again, data that is about 40 days old, so again forgive me if I can't walk you through
it in great detail, but again the yellows now are going to be the very freshwater.
And if you see a circle, that's where we sampled with a monitoring device about 15 centimetres
below the water surface.
If you see a dot inside that circle, that's the location where we dropped our device down,
just above the sediments.
So we're looking for stratification.
What do we see? Well over here in the eastern part of the lake there is very little fishing
over there.
It is a very wide-open area.
Lots of wind, lots of mixing, but also a major river coming into it over here.
We see a lot of fresh water, but almost no evidence of stratification until you get way
out towards the centre of the lake.
How bout in the west?
The north-west, our model predicts that we have groundwater up-flow there.
We're seeing a lot of freshwater there.
And while it doesn't show any dots in here, that's because the lake was very shallow here.
We could not get a 2nd measurement.
No indication of stratification up there.
Go down in the south-west, this is interaction with the Dunia River.
We do have some fresh water coming in here.
We've got the light and dark blue with the magenta inside.
Here we've got some substantial stratification going on, unfortunately right in the middle
of our capture zone.
What going on up here?
Well this is a river called the River So, one of the more confusing areas because, this
is one of those points you saw in the previous plot where the sediments just showed salinity.
The question is where on earth did that salinity come from?
The subsurface has none. It's been studied.
There's no old halite deposits or anything, gypsum or anything along that line that can
give us salinity up here.
Well, now what we found is up there is the shallow, 15 centimetres down is extremely
fresh, because the rains have started and we got fresh water coming in.
But a full 7 kilometres up that river we can find high concentration saline waters, about
50% of ocean water, fully 7 kilometres up that river.
So that water has come not only up here, it’s come all the way from the ocean, all the way
through, and all the way up there.
So we have a very unusual surface water system now, and again, remember I mentioned the bathymetry
earlier that there were no obvious channels going up this way.
This is not just a deep channel carrying salinity up there, there's something very interesting
going on there.
What we think is going on is through the dry season the salinity can get up the stream
quite a ways.
Over here because it’s a bigger river it is being flushed out, but over here it simply
doesn't have the strength to flush it out as quickly, but we're not sure yet, that's
why we're beginning to study it.
So observations on the lake.
The majority of sediments we believe are losing so it is a recharge area, except for the north-west
The historical data suggested this wild variation in salinity is happening, we've seen it.
We are now pursuing chemical signatures.
We collected the samples this summer, both isotopic and chemical, to try and figure out
what's in the wells, what's in the lake, what's the distribution of chemistry in the lake.
What are we seeing in so far as sources of water?
Hydraulics were distributed in very much as we would expect from earlier studies on smaller
lakes in the U.S. by Winter and others.
And again, we can get a lot of information for very simple equipment and get started
in this area, because we don't have many other options.
With that I return to the question: "Can we get a decent hydrogeologic investigation going
under these constraints?".
And I'm going to argue that the answer to that is "yes". Is it perfect? No.
There is gonna be uncertainties. We're gonna have to monitor the system down the road.
But with creative and appropriate technologies, with the right partners in-country, doing
things that are consistent with the local culture, we really could get a very consistent
and good sampling design going here.
Got good interaction with the numerical models.
So they were guiding us as to where to sample, and we were bringing the data back to the
And recognising the limitations and uncertainties, so we're not gonna go out and tell the government
"Hey this is what's going on, we understand it, and this is what you should do."
We're using hydraulic tests to try and understand the complex flow of air.
We're also doing chemistry there now, using our piezometer complexes where we've got a
series of permanent multi-level sampling devices, and down here with the direct-push in geophysics,
we're starting to understand the coastline.
Are we done? We're nowhere close to being done.
We would like in the next three years to hand the government some kind of model that says
here's how to approach managing this system.
To do that we need to continue to better characterise the hydrology.
That means we got to characterise the source and the impact of uncertainties.
We need to be able to tell them those things.
Again, that'd be to provide the government with likely solutions, but when I say possible
solutions here, that means this is what you should try but as well this is you should
monitor and keep updating your models as you go forward with this.
Don't just take our word for it.
Why? We don't know the interconnections here, we don't know the parameters here, we don't
really know the recharging quality and quantity over here.
We don't know the contributions down there.
So we need to help them get an advantage of the system, but we need to do it in such a
way that they're very flexible to work with us.
So with that I'm gonna take about 30 seconds just to finish up with this.
It is my strong belief that if I had gone into Benin and continued to be the stupid
American I was back then.
Again you might argue that I still am in certain ways.
But if I had refused to work with their government agencies, we wouldn't be anywhere near where
we are now.
But this is Landry Loubegnon.
He works with our NGO over there, and he has gone out, and he has surveyed all the different
groups that we've worked with.
That's about 50 different groups now and Landry wanted me to give you about 5 slides about
how many people we've impacted in all that.
I don't think any of that's important, but one thing that is important is his third statement.
I haven't changed his wording except translating it from French.
But everybody that we're working with over there is noting that this is predominantly
a Benin project that happens to have a U.S. counterpart, and it is being run in such a
way that is consistent with national priorities and strategies.
Why do I believe that's absolutely critical?
Well think about what has to happen in the next 4 to 5 years?
We need to go to their government water agency folks and say "Please change how you run your
well field.
Please look at our models. Please look at our suggestions. Please look at our monitoring."
How many hands in this room think I could go over there as an American who was unknown
to them, and walk into their offices and get them to listen to me?
How many think however, if we got the head of their water agency, 2 of the major faculty
members in the country, a lot of the local folks who have been working with us, all going
in together, leaving the white person out, leaving the American out of the discussion
entirely - now we've got a chance to walk into their office and actually get some change.
And so I find this one to be very very reassuring that we've been doing this the right way.
With that, this is Professor Boukari.
As a matter of fact because of this project, they have started a new masters and PhD program,
so if any of your students want to find a very unique place to try graduate studies,
go on over.
But based on this collaboration, Professor Boukari wants me to invite anybody and everybody
who listens to this talk to come on over and help him work on the hydrogeology of West
Should be a lot of fun. With that, I will say thank you and open this to discussion,
critiques and any questions you might have.
Thanks a lot for your attention.