Why Pastoralists Don’t Bet on One Best Animal

To someone accustomed to livestock operations organized around one kind of animal, a pastoral herd can contain a surprise.

Cattle may be moving alongside goats, sheep, and camels. They overlap in their broad purpose: each can turn the landscape into food, income, transport, or stored wealth. Yet keeping several species also means managing animals with different needs, behaviors, and productive strengths.

If the goal were to choose the animal that performs best in an average year, the mixture could look inefficient.

Pastoralists do not live only in average years.

Dryland environments can change sharply. Grasses may decline while woody browse remains. Water may become harder to reach. Heat, disease, and forage conditions do not affect every species in the same way. Cattle, goats, sheep, and camels can therefore overlap in what they provide while differing in what threatens them.

The herd is not simply diversified because having more kinds of things is generally good. Its components respond differently to the same bad year.

Ecologists have a name for that property: response diversity. Different species can contribute to a similar function while reacting differently to disturbance or environmental change.

That changes what “best” means. The animal with the highest productivity under ordinary conditions is not necessarily the herd composition most likely to continue providing milk, meat, transport, or tradable wealth across droughts and other disruptions.

The same distinction appears in engineered systems, although every field gives it different language.

Safety engineers worry about common-cause failure. NASA guidance warns that redundancy can be invalidated when supposedly independent systems share a cause of failure. Reliability discussions sometimes contrast identical redundancy with dissimilar or diverse redundancy.

The labels matter because two systems can be separate at the functional level and effectively identical at the failure level.

Two generators may both depend on the same fuel line. Two network connections may reach the building through the same buried conduit. Two software services may contain the same defect because they were built from the same library. Two suppliers may rely on the same upstream factory.

The equipment list shows two. The failure map shows one.

The pipeline does not count only generators

The Trans-Alaska Pipeline System makes this distinction unusually concrete.

Pump stations along the 800-mile pipeline need electricity for pumps, heat, lighting, control systems, and other essential operations. Alyeska’s published description of those stations does not present resilience as a simple count of duplicate machines.

At Pump Station 1, one turbine generator normally operates and another serves as a backup. The station can also import power from the Prudhoe-area grid if necessary. Those alternatives do not merely add another generator; they provide a path to power from outside the local generating arrangement.

At Pump Stations 3 and 4, one turbine generator runs only on natural gas while another can run on natural gas or liquid diesel. If the natural-gas line is unavailable, the second machine has another fuel path.

Alyeska also describes warm-standby stations and facilities maintained for cold restart. These are not all interchangeable backups. Local generation, imported power, alternate fuel, warm standby, and cold-restart capability protect against different obstacles to keeping or restoring the pipeline’s operation.

Counting the generators would miss much of the design. The revealing question is what each alternative still depends on.

The military calls one version PACE

Military communications planning uses another name: PACE, for Primary, Alternate, Contingency, and Emergency.

The acronym can sound like a four-item inventory. But four radios are not four meaningful paths if they all depend on the same carrier, tower, spectrum, power source, or upstream network.

A useful PACE plan may cross physical infrastructure and transmission methods: terrestrial voice, cellular or data networks, satellite communication, and systems using different waveforms or frequencies. The value is not in reaching the number four. It is in preserving communication after different kinds of disruption.

Here the military’s planning language and NASA’s safety language point toward the same question: what could disable both?

A power grid needs a way to start without itself

Electrical-grid restoration makes the dependency problem harder to ignore.

Large generating plants often need electricity to start pumps, controls, cooling systems, and other equipment. After a widespread blackout, they cannot rely on the grid they are supposed to restore.

Utilities therefore maintain blackstart resources: generation capable of beginning the restoration process without outside grid power. Those resources energize other equipment and allow larger sections of the system to return gradually.

But a generator’s ability to start is only part of its failure path. Its fuel must also be available. Fuel delivery may depend on pipelines, compressors, pumps, transportation, communications, or electricity elsewhere. Modern blackstart studies consequently examine fuel diversity, alternate fuels, single points of failure, and non-fuel options rather than assuming that several generators are several independent protections.

The vocabulary changes from one domain to another: response diversity, common-cause failure, dissimilar redundancy, PACE, blackstart, fuel assurance, warm standby.

They are not identical doctrines. They developed around different work and different hazards. What connects them is a refusal to treat every additional unit as an additional way to survive.

The pastoral herd makes that distinction visible before any technical vocabulary is needed. Several animals may serve the same larger purpose without sharing the same relationship to drought, heat, water, forage, or disease.

Once that is visible, a row of identical backups begins to look less reassuring.

The inventory may show many units. The failure map may show only one.

Sources and further reading

• Food and Agriculture Organization — The management and mitigation of vulnerability
• Leslie and McCabe — Response Diversity and Resilience in Social-Ecological Systems
• NASA — General Safety Program Requirements, Chapter 1
• NASA Technical Reports Server — Common Cause Failures Dominate and Defeat Redundancy
• Alyeska Pipeline Service Company — Pump Stations
• U.S. Army — Industry Insight: Communications Skills
• National Laboratory of the Rockies — Black Start
• Federal Energy Regulatory Commission — Blackstart Study Recommends Collaboration, Planning for Resilience

Featured image: MelissaW via Wikimedia Commons, CC BY-SA 4.0.