Off-Grid · Remote · First Nations · Solar Microgrid · LED Efficacy · Policy

From Firesticks to Microgrids: Why a Watt of Lighting Costs More in Remote Australia

Most of the calculations on this site answer one question: how fast does an LED upgrade pay for itself out of your electricity bill? On a grid connection in Sydney or Adelaide, that is the right question. Power is there when you want it, you are billed in arrears, and efficiency shows up as a smaller number on an invoice.

Move a few hundred kilometres past the end of the network and the question changes completely. Off-grid, a watt of lighting load is not a line on a bill. It is a piece of infrastructure you had to buy, ship, install and eventually replace. You pay for it in panels, in battery, in freight, in a diesel genset that runs longer than it should. The bill never stops being the point, but it stops being the main point.

This is worth understanding right now, because remote Australia is in the middle of the biggest change to how it gets electricity in fifty years, and it is being led, increasingly, by the communities themselves.

Part One

Australia has done energy scarcity before

It is easy to talk about energy efficiency as a modern engineering discipline. In this country it is closer to the oldest one.

The archaeological and anthropological record of Aboriginal Australia describes something an engineer would recognise immediately: a set of strategies for getting the most useful work out of the least energy input, in an environment with very little slack.

Fire, released in small controlled doses. The practice Rhys Jones named "fire-stick farming" in 1969, and which Bill Gammage mapped across the continent in The Biggest Estate on Earth: was not simply burning country. It was frequent, low-intensity, mosaic burning that kept fuel loads down. The consequence is that fire released its energy in many small managed increments instead of one catastrophic pulse. That is load management. Charcoal records now show a continent-wide shift toward more frequent, less intense fire beginning around 11,000 years ago, associated with Indigenous management.

Transporting stored heat instead of remaking it. Geoffrey Blainey noted in Triumph of the Nomads that fire was routinely carried: a slow-smouldering firestick moved from camp to camp. Igniting a fire from nothing costs real effort. Carrying an ember costs almost nothing. Anyone who has sized a battery bank to avoid a cold start understands the logic.

Thermal mass, engineered from what was available. In parts of the country with no suitable stone, earth ovens were lined with fired clay balls or fragments of termite mound to act as heat retainers. Firewood was selected (in one documented case, consistently under about 10 cm diameter) for burn characteristics rather than gathered indiscriminately. That is fuel specification.

Accepting a metabolic cost to avoid a fuel cost. Physiological studies in Central Australia in the 1950s recorded people sleeping between small fires and allowing core body temperature to fall several degrees overnight, rather than burning more fuel or carrying more covering. Whatever else it is, that is an explicit trade-off between comfort and energy expenditure, made deliberately, and repeatedly.

Infrastructure that does work without an energy input. At Budj Bim in western Victoria, the Gunditjmara built channels, weirs and traps that used the landscape's own gradient to move water and harvest eels. Parts of the system are thousands of years old. It was inscribed on the World Heritage List in 2019. It is, functionally, passive infrastructure: it does useful work indefinitely because it was designed around a resource that was already there.

None of this is folklore, and none of it needs romanticising to be interesting. It is a long, well-documented record of a society treating energy as the scarce input it actually was, and designing around that constraint rather than pretending it away.

Which makes the current situation in remote Australia particularly hard to look at.

Part Two

The most energy-insecure households in the developed world

Research led by the Australian National University, using smart-meter data from around 3,300 households across 28 remote communities, found that 74% of households on prepaid electricity meters were disconnected more than ten times in a single year. In some communities, 91% of households experienced a disconnection.

That reframes efficiency entirely. On a prepaid meter, a watt saved is not cents saved. It is minutes of runtime before the power goes off.

Part Three

Communities building their own supply

The response has been a wave of First Nations-owned generation, and it has moved fast.

Policy has followed. The First Nations Clean Energy Strategy 2024–2030 was released in December 2024 with $70 million attached, and from 2026 includes a provision encouraging developers to enter agreements delivering at least 5% equity or equivalent revenue share to First Nations partners.

Every one of these projects has a lighting load inside it.

Part Four

The arithmetic of a watt, off-grid

Here is where it gets concrete.

On grid, one watt of lighting load running 12 hours a night costs you 4.38 kWh a year. At the Northern Territory commercial average of about 28c/kWh, that is $1.23 per watt per year. Straightforward, and exactly what the savings calculator on this site is built to model.

Off-grid, that same watt has to be built before it can be run.

What that means in efficacy terms

Take a single area or street light with a design requirement of 10,000 delivered lumens.

This is the thing that catches people out. On grid, cheap fittings cost you slowly. Off-grid, they cost you at the moment you write the cheque, because you are paying for the generation and storage to feed their inefficiency, up front, at remote-Australia prices.

The second lever: controls

Efficacy sets the ceiling. Controls decide how much of it you actually use.

Area and street lighting in a small community rarely needs full output at 3am. Dimming to 30% after midnight cuts roughly 40–50% of the night's lighting energy, with a corresponding reduction in storage requirement. Motion-triggered step-dimming does better again on paths, car parks and yard lighting. For an off-grid system, a good controls strategy is often cheaper per kWh avoided than adding another battery module, and it has the useful property of protecting the batteries from deep cycling on overcast days.

Specification points that only matter out there

• Driver temperature rating. Ambient temperatures inside a Pilbara or Barkly luminaire enclosure routinely sit far above the 25°C at which most efficacy and lifetime figures are quoted. Drivers, not diodes, are the usual failure point. Ask for performance at realistic ambient, not lab conditions.
• The real cost of a failure. A fitting replacement in a capital city is an hour of labour. In a remote community it can be a two-day trip for one person plus vehicle. Rated life and warranty terms should be weighted accordingly. The cheapest fitting is very often the most expensive one.
• DC options. Where a system is DC-coupled, native DC luminaires avoid a conversion stage and its losses. Worth investigating on standalone lighting rather than assuming the AC default.
• Dust and IP integrity. Ingress protection ratings degrade over years of thermal cycling in high-dust environments. Gasket quality and housing design matter more than the number on the box.
• Warmer colour temperatures. 3000K and below reduces skyglow, is easier on nocturnal wildlife, and is generally preferred where dark skies matter, a consideration across a great deal of remote and regional Australia.

Part Five

Who is actually buying

If you work in lighting supply or installation, the procurement pathway in this space is not the standard utility tender. Increasingly it runs through Indigenous corporations, land councils, community-controlled housing bodies and Aboriginal-owned developers, supported by ARENA programs, the First Nations Clean Energy Strategy funding, and state remote-power programs.

The practical implication is that the buyer's decision framework is different. They are usually not optimising a payback period against a tariff. They are minimising installed system cost, maximising reliability at distance, and, in the community-owned projects, trying to convert generation into a benefit that reaches households directly rather than disappearing into a retailer's margin.

The through-line

There is a reasonable temptation to draw a neat moral here about ancient wisdom and modern technology. It is not necessary, and the record is more interesting than the moral.

What the archaeology actually shows is a very long practical engagement with a hard constraint: energy is expensive, so use less of it, and design the system so it does more work per unit input. Carry the ember. Burn small and often. Line the oven with something that holds heat. Let the landscape do the pumping.

The remote communities now building their own solar and storage are working the same problem with different tools. The most valuable watt in those systems, exactly as it was then, is the one nobody has to generate.