dave is this not just putting a bigger straw in the milk,
Not really. It's more like squeezing the bottle, but that's a poor analogy unless you're willing to accept a pretty abstract generalization of "bottle". A better analogy is not sealing the bottle around the straw, so air flows in to replace the milk you suck out. Even better is pushing one fluid into a sponge to displace whatever was there before, and/or to keep it full while you drain fluid out.
"Putting a bigger straw in" is more analogous to drilling more producing wells, which will still only recover whatever oil has the energy and mobility to flow through the rock to get to them. However many wells you've got, you're done when the pressure differential from the reservoir to the bottom of the well is too small or the "oil" saturation is too disconnected to produce at paying rates. (I quote "oil" because the same concept applies to both reservoir liquid and vapor, either or both of which contribute to the surface separator's output where we call the liquid "oil" and the vapor "gas").
Injecting water can reduce the amount of oil left behind when that time comes. With enough injection (or influx from an aquifer, or expansion of a gascap or gas evolved below bubblepoint pressure, or compression/compaction of the reservoir rock itself), the pressure might never decline so far you can't produce
something at reasonable rates. However, when the local hydrocarbon phase ("oil" [reservoir liquid HC] or "gas" [reservoir vapor HC]) saturation drops, so does the permeability of that phase.
Injecting water (or gas, or aquifer influx) can displace oil toward wells to keep it flowing longer. Even if the trapped residual saturation is the same number of barrels in the reservoir, trapping reservoir oil at near-bubblepoint pressure leaves what would be fewer "stock tank" barrels behind. (smart guys will notice that trapping reservoir
gas at high pressure leaves more hydrocarbon behind than would be trapped at low pressure)
how much extra oil do you recover?
Like my grandfather's haberdasher advised: Depends. How long is a piece of string?
The answer depends strongly on many details of each particular field. How permeable is the reservoir rock? Is it well-connected like one big slab of sand, or is it shattered into discontinuous faultblocks, separated into non-communicating layers, or a collection of more-or-less disconnected blobs? Is it a nice uniform porous medium through which water will flow uniformly, or does the permeability vary manyfold from place to place? Is the porosity more or less evenly distributed, or is most of the volume tied up in tombstone-tight blocks around which most of the flow capacity exists in thin fractures? Is the reservoir more or less level, or is it tipped at severe angles? Is the oil thick and viscous, or light and fluid? Is it oil in the reservoir? Gas? Neither and both (e.g. supercritical fluid)? How does that change when pressure declines? Is there a lot of dissolved gas to slow pressure decline as it bubbles out of the oil, or will it take its compression energy with it by squirting quickly out of the wells, or is the oil just too "dead" for gas to matter much? How squishy is the rock? Will it fail mechanically and pinch off the wells if we let pressure fall too far? Is there a large aquifer that will contribute water as we remove oil? A big gas cap that will expand as pressure declines?
Considering that you've only "seen" the reservoir through a few goop-filled boreholes inches wide and miles long, or by the functional equivalent of farting in the ocean and listening to the bubbles echo, how confident are you that you're right about these things? As one UT professor said about core analyses: "Congratulations! You now know all there is to know about a part of the reservoir
that isn't part of it anymore."
And a big question... How much more money are you willing to invest *today* to have another barrel to produce in the distant future? You won't see those "extra" barrels for a long time, typically decades. Your contractors and suppliers cling to that old-fashioned business model and expect to be paid promptly after drilling your well or delivering your pipe. Wimpy customers seem reluctant to pay today for oil you'll be giving them in April 2019.
Working that stuff out is what's been buying my beans for a long time. Evicting old ladies and drowning kittens are just hobbies
From what i underdate this just helps increase the extraction rate!
Sure, maintaining pressure might allow for more reservoir->wellbore pressure differential to drive higher production rates, and can also keep reservoir liquid viscosities low by keeping "gas" in solution with the "oil". The displacement effects can greatly increase ultimate recovery, too.
If you are in the business, are we close to peak oil?? or we there now?
Alas, I'm not sufficiently competent and informed to pick a date (and I suspect the same is true of most people claiming otherwise). It's hard enough to estimate the ultimate recovery and producing lifetime of a specific field with specific data; guessing at recovery from other fields, including those that haven't yet been discovered, is a really dirty business.
How close we are depends strongly on how much you're willing to pay. Seriously -- nobody's going to spend several billiion dollars to develop an oilfield under a mile and a half of ocean a hundred miles from land unless they're fairly confident the oil it will produce can be sold for enough money to make it worth doing. Don't forget that all that income will be spread out over decades, but nearly all the investment has to be made up front -- including a fair chunk of change that has to be spent before you can really decide whether it's worth doing.
Besides, does it really matter whether production truly hits a peak or simply gets outrun by demand? Either way, somebody who wants grease ain't gonna get enough.