Colombian oil producers in the Llanos Basin pump far more water than oil. The percentage of water they produce is going to get ever-more dramatic in coming years, as the oil runs lower and what is called “water cut” gets higher. State oil company Ecopetrol, which has some of the biggest llanos oilfields, is now trying all sorts of ways to turn that water into something useful, such as irrigating forage for cattle and irrigating water-hungry lumber trees. They are improving treatment for the water they dump into rivers and they recently built their first injection well.
They are pumping the water into a wet landscape, and the aquifer they are pumping from is so full of water that in decades of oil production in the area, the water pressure has dropped about 1%, according to engineers who briefed me and another reporter in August (yes, took me a while to post this). So this is not one of those cases where the environmental impacts are obvious and horrifying.
Rather, what jumped out at me from the presentations we saw there was the epic scale of water production that Ecopetrol has planned. Here are charts of oil and water output projections for four fields in Ecopetrol’s Central Region. These are all in the llanos department of Meta, and almost all of these plans are in the Cubarral block, which includes the company’s biggest field, Castilla, as well as Chichimene, Akacias and a new exploration area called CPO-9.
The top chart shows how current wells are declining. Without massive new investment, the field will taper off, as in purple. With massive new investment, it can increase output before tapering off a bit more slowly, as shown in green.
The second chart shows water output. The skinny dark blue strip at the bottom is water output in the ield as it stands today. The light blue mountain on top of that is water output if the investment plan is fulfilled.
To put this in real terms, water output in this one region for this one company would peak in 2018 at 4.12 million barrels a day – 98 million gallons a day. At current Colombian urban consumption (131 liters a day per capita), that would be enough for a city of 5 million people, or 10 percent of the country’s population.
And all this doesn’t even include any new water production in the many non-Ecopetrol projects in the region. Note that Pacific Rubiales, which operates the very successful Rubiales field, had to hold back oil output in the first quarter of 2012 as it waited for permits to increase its water disposal rates.
What is perhaps most surprising about this is how the water production stays high over the years, while oil production tapers off. Ecopetrol is going to have to deal with more and more water for every barrel of oil.
I have no idea what will be the long-term impact of taking that much water out of the ground. Anyone who happens to know, feel free to write in.
Setty's notebook 
If the aquifer below has only dropped 1% then to me e most obvious solution would be to reinject it as cleaning it up i imagine would be rather expensive. Water from oil reservoirs is not generally of drinking waternquality, besides having a number of minerals in it, the salt content would be above human consumption tolerances requiring a reverse osmosis unit to treat the water. Any idea on what the water quality is like?
Reinjecting the water would allow you to maintain reservoir pressure above the bubble point which is key to stop the gas production to shoot p at the expence of oil production.
Yup, they are starting some reinjection. However, remember that in the llanos, the oil is quite deep. And unlike in, say, East Texas, these are new oilfields, so every well that’s drilled is generally the first one in its area. To reinject, the companies need to invest in injection wells built just for that. They seem to be planning to mostly use it to water palm trees and to use bioremediation wetlands to clean up the hydrocarbons.
I have no idea about the water quality. I would love to find out.
Have a look at what , Santos, Origin, are having to build in queensland australia for their CBM to LNG projects. They are each looking at producing at their peak 1 mmb/d of water which is relatively clean, not to far off drinking water spec. But they are still having to build multiple reverse osmosis units to clean up the water. They looked at using crops and trees to deal with the water but they reckoned they would only be able to get rid of 100 mb/d that way at best.
The cleaned water is either then being put into the river systems or re injected into aquifers below. Like you said there will be no getting away from having to drill specific injection wells.