Sometimes the simplest concepts are the most powerful. In a more volatile world, one brought on by the converging challenges of energy, environment, and society, one such concept might be of a lot of use: buffers.

To buffer is “to lessen or moderate the impact” of something. That’s just what we need—volatility itself can be a problem. Consider climate volatility. It’s not just that the planet is gradually warming, but that weather patterns are becoming more volatile as a result. That volatility manifests as extremes—floods followed by droughts, extreme heat one year and extreme cold the next—and it’s hard for any plant or animal (humans included) to adapt to such wild swings. Consider financial volatility. In the last four years we’ve seen the price of oil explore a range of a factor of five in price from top to bottom, and that variability has made it hard at times to (short-term-profitably) invest in alternatives among other things. The question then is how to respond to such volatility.

In a classic article, Donella Meadows identified twelve general points of leverage in systems (if you haven’t read it, it’s a brilliant article and well worth the time). She begins with this:

Folks who do systems analysis have a great belief in “leverage points.” These are places within a complex system (a corporation, an economy, a living body, a city, an ecosystem) where a small shift in one thing can produce big changes in everything.

The idea is not unique to systems analysis—it’s embedded in legend… We not only want to believe that there are leverage points, we want to know where they are and how to get our hands on them. Leverage points are points of power.

The leverage points she identifies are in fact quite mundane, but their simplicity belies their effectiveness. Buffering is one of them. While buffers increase stability, they also decrease responsiveness; while they increase resilience, they decrease efficiency. They aren’t a panacea, but as it turns out the concept of buffering shows up in all sorts of places as a valuable general approach to mitigating the challenges we’re facing. Buffers in and of themselves can’t solve anything—Meadows ranks changing the size of buffers as a less powerful leverage point, and introducing delays (an effect of many buffers) as only slightly more powerful. But they can make the intractable slightly more tractable.

Here I’d like to consider some useful applications of the concept. Some are obvious cases of buffering while some are more subtle. I’d be very interested in hearing about more examples of where buffers are of value.

Swales. While some climates receive a steady trickle of rain throughout the year, they are increasingly the exception. When rain does fall in quantities that the ground and hillsides cannot absorb it immediately, the flow runs off, taking nutrients and of course much-needed water with it. Swales serve as buffers, capturing water in the soil by impeding its flow downhill, and in doing so can dampen the water cycle and mitigate the effects of increasing extremes of precipitation.

Strategic Petroleum Reserve. The SPR is an simple buffer: a stored reserve of oil that is released when prices are high or access to oil imports is limited. There have been many arguments for and against keeping the SPR in its current form. I made the (somewhat unrealistic but I hope plausible) argument that instead of eliminating the SPR we should increase it dramatically and use it as a buffer to dampen oil price swings. Dampening price swings would provide predictability and as a result would enable longer-term planning for investment in alternatives.

Energy Storage. Among the most obvious buffers are devices and systems for energy storage—batteries, flywheels, pumped hydro, molten salt. Each enables the smoothing out of energy availability—taking what’s available at one time (or place) and making it available at or over another.

Delay-Tolerant Networks. All communication networks use buffers: almost every computer, switch, and router in the Internet has a buffer to smooth out and cope with potentially-bursty packet arrivals. As those buffers are increased, the ability to cope with intermittent network availability in increased (and thus the network’s resilience is increased), though at the potential cost of its ability to support real-time communication. Networks of this sort are thought of as store-and-forward: each hop along the way buffers some data before passing it along. Delay-tolerant networks are ones that can operate at the more extreme end of network disconnectedness: their buffers are large enough to be able to store up large amounts of data for later delivery.

Thermal Mass. One of the key ideas in passive solar architecture is the use of thermal masses. Such thermal masses can be as simple as a South-facing dense house wall (or North-facing, for those in the Southern Hemisphere) designed to buffer the sun’s heat: absorb it slowly through the day and release it slowly at night. A stone wall or stone ground cover can provide similar benefit to crops by dampening day/night temperature fluctuations. Within homes, thermal masses can slowly release heat from a heat source—a rocket mass heater is a good example of this.

Hugelkultur. One of the more interesting and off-the-beaten-path ideas in the world of permaculture is hugelkultur: burying wood. In burying wood, the soil stores not only more water but more carbon and more nutrients: it buffers these for the sake of the garden. There are plenty of reports of large hugelkultur beds obviating irrigation even in places that have dry summers, and of doing so while extending the planting season, making harvesting easier, and building up the soil.

Fat. Fat—both human and non-human—is such an obvious buffer that it’s almost weird to include. Nevertheless it’s worth remembering that it’s evolutionarily good for something—it smooths out the times of plenty and the times of want—and has energy density comparable to jet fuel to boot.

Phosphorescence. Phosphors are used in all sorts of applications, from fluorescent bulbs to CRT displays. The persistence of the phosphor can be thought of as its buffering ability: how long does the phosphor maintain light output upon stimulation? A long-persistence phosphor can help decrease visible flickering—the pulses of stimulation are buffered from the viewer—at the cost of display lag—responding slowly to changes. My limited understanding is that modern CFLs have less flicker in part because of improved phosphors.

Wetlands. Just as swales store water in the soil, wetlands store it above ground. In doing so, they buffer the effects of runoff: they help prevent nutrient loss (and potential downstream nutrient excess), help process toxins and waste products, and release them slowly to the downstream environment.

Cover crops. Many of the approaches to preventing nutrient loss are complementary. Cover crops help store the soil’s nutrients in the bodies of plants and increase the soil’s stability, decreasing what’s available to run off. (This can be combined with an artificial wetland downhill/downstream and swales uphill/upstream to significantly decrease nutrient, soil, and water loss.)

Glycemic Index. The ubiquity of refined sugar in the modern Western diet (itself a consequence of bad agriculture policy and a side effect of fossil fuel availability) has led to numerous health problems. One of the primary issues has to do with the rate of absorption, something which is measured by the glycemic index of the food. Since high glycemic index foods break down quickly, they cause a rapid spike in blood sugar levels (and consequently, rapid crashes as well); such volatility can contribute to problems such as type 2 diabetes. Of all the examples I’ve given here, this is perhaps the least obvious case of buffering, as the buffer in this context is the complex carbohydrates that release their energy slowly—low glycemic index foods such as vegetables and whole grains—as opposed to foods made of simple sugars that release their energy immediately.

Where else can buffers be useful?

Leave a Reply


Responses to “Buffers”

  1. Buffers are also possible in realtime communications networks via the simple expedient of queuing.
    Example: GPO Telephones in the UK (the British equivalent of the Bell System) before Subscriber Trunk Dialling, had operators placing all long distance calls. If all trunks were busy, they would go into “delay working,” whereby callers were queued and called back in order, as trunks became available.
    This recognizes that the key benefit of realtime voice communication is not that the connection is made immediately, but that the conversation itself is interactive in a manner that most closely resembles face-to-face conversation.
    In point of fact, a delayed connection into a channel that provides realtime analog communication, is more effective for conversations that require a high degree of cognitive engagement, than an immediate connection to a communications channel with random and variable delays, such as occurs in packet-switched voice networks (cellular and VOIP).
    Each of these variations has its more effective and less effective uses, and the key to designing and using communications networks is to ascertain what is needed for each purpose.

  2. g2geek – Interesting observation. Basically that it’s sometimes (maybe often) better to have queueing delay on the control channel rather than the data channel.

  3. g2geek,
    VoIP these days is well managed in terms of delay because of QoS enforcement. VoIP and video packets get higher priority over the network compared to data packets [non real-time traffic]. This combined with packet loss concealment techniques delivers toll quality speech and wideband codecs give even better speech quality. Newer cellular standards like LTE have a requirement of call setup time of 100 ms and VoIP delay of 10 ms, which is quite good. For conversational speech, end to end delay needs to be within 200 ms or so to sound normal. In the old days international calls were very annoying [lot of talk overlap] because of going through two satellite hops of geostationary satellites. Now it is VoIP on fiber optic cables on the surface of earth, so it is much less delay. VoIP has finally enabled global voice communications at very low cost.