Can Renewable Energy Actually Power the World?

On a flat, stony plain in the Thar Desert of Rajasthan, about 200 kilometers from the Pakistan border, roughly ten million photovoltaic panels sit bolted to steel frames and wired into combiner boxes that stretch across some 56 square kilometers of land. This is the Bhadla Solar Park, commissioned in stages between 2018 and 2020, and at full output it reaches a nameplate capacity of about 2,250 megawatts. By a comfortable margin it is one of the largest single solar installations on Earth. What makes it remarkable is not just its size but its ordinariness: the world is now building the equivalent of one entire Bhadla every few days.

That pace forces a question that sounds simple but turns out to have a precise, calculable answer. Can renewable energy actually power the world? Not in the sense of a slogan, but in the sense a geographer would mean it: with real installed capacity, real costs per unit of electricity, real land, and the awkward physics of getting power out of the sun and wind only when the sun shines and the wind blows. The honest answer requires a few tools, and once you have them, almost any headline about clean energy becomes readable.

The Scale Renewables Have Already Reached

It helps to start with how much has already been built, because the numbers are larger than most people assume. At the end of 2024, global installed renewable capacity crossed 4.1 terawatts. Solar photovoltaics led the way at about 1,560 gigawatts, hydropower sat at roughly 1,410 gigawatts, and wind at about 1,020 gigawatts, of which offshore wind contributed around 75 gigawatts. The annual flow of new construction is even more telling than the stock. In 2024 alone, new additions were dominated by solar at roughly 450 gigawatts, with wind a distant second at about 115 gigawatts.

To put that in perspective, the 450 gigawatts of solar added in a single year is two hundred times the capacity of Bhadla. The technology that barely registered on the global grid two decades ago is now the largest source of new generating capacity being installed anywhere, in any form, in any year. This is the backdrop against which the rest of the discussion has to happen: renewables are no longer a niche or a pilot. They are the default thing the world builds when it builds new power.

Why the Build-Out Lands Where It Does

A geographer's first instinct is to ask not how much, but where, and the locations of the world's giant solar parks are anything but random. Utility-scale solar concentrates in places that share four features: flat land, high insolation (the amount of solar energy arriving per square meter), low population density so that land is cheap and uncontested, and a developable transmission corridor to carry the power to where people actually live. Bhadla has all four, which is exactly why it sits in the Thar Desert rather than somewhere more convenient to a city.

The same logic explains the global map of large solar at a glance. Rajasthan, the Mojave Desert of the American Southwest, the Atacama in Chile, and the Tabuk plateau in northwestern Saudi Arabia host the planet's biggest solar parks, while overcast Hamburg and rainforest-shrouded Manaus do not, no matter how much those cities might want clean power. Geography is not destiny for renewables, but it sets the terms. A place with strong, reliable sun and empty, buildable land has a structural advantage that no policy can fully manufacture elsewhere. This is why so much of the energy transition is, underneath the technology, a story about specific landscapes.

The Four Pillars and the Cost That Changed Everything

Renewable electricity rests on four established technology pillars, each with its own physics, geography, scale, and cost. Solar photovoltaics convert photons directly into electric current with no moving parts. Wind turbines extract the kinetic energy of moving air. Hydropower converts the potential energy of water held behind a dam, and it has been the foundation of renewable electricity since the large dams of the 1930s. Geothermal taps the steady heat flux rising from Earth's interior, anchoring a small but durable share of generation in places like Iceland and the western United States. Solar leads the new build, wind sits just behind it, hydro is the old foundation, and storage is the missing piece the rest of this decade has to construct.

What turned these technologies from worthy experiments into the cheapest electricity in history is captured by a single metric: the levelized cost of energy, or LCOE. LCOE is the per-megawatt-hour cost of generation averaged over the entire lifetime of the asset, including the capital to build it, the cost to operate and maintain it, and any fuel it burns. The investment bank Lazard publishes the industry-standard annual estimate, and its 2024 update is striking. Unsubsidized utility-scale solar comes in at $29 to $92 per megawatt-hour, and onshore wind at $27 to $73 per megawatt-hour. Both sit well below new combined-cycle natural gas, which Lazard puts at $45 to $108 per megawatt-hour. The decisive fact about renewables in the 2020s is that, in good locations and without subsidy, they are simply the cheapest new electricity available.

Why a Megawatt Is Not a Megawatt-Hour

Here is where most casual reading of energy news goes wrong, and where a geographer earns the right to be skeptical. The most durable misconception about the renewable build-out is that a megawatt of installed solar capacity delivers a megawatt of power continuously, and therefore produces 8,760 megawatt-hours over a year (since a year contains 8,760 hours). It does not, and the gap matters enormously.

The bridge between installed capacity and delivered energy is a quantity called capacity factor, the ratio of the energy a plant actually delivers over a year to the maximum it could theoretically deliver if it ran flat out the whole time. A one-megawatt plant producing the full 8,760 megawatt-hours would have a capacity factor of 100 percent, but nothing real reaches that. Solar runs at about 22 percent because the sun sets every night and clouds pass during the day. Onshore wind manages roughly 35 percent, hydropower about 40 percent, and nuclear, which is designed to run continuously, hits about 92 percent. So a typical one-megawatt utility-scale solar plant does not produce 8,760 megawatt-hours a year. It produces closer to 1,927 megawatt-hours, because the sun is simply not always shining on it.

This single correction reshapes every comparison. When you read that a country installed a gigawatt of solar and a gigawatt of nuclear, you have not learned that they added equal amounts of electricity. The nuclear plant, running near 92 percent, will deliver roughly four times the annual energy of the solar farm running near 22 percent. None of this makes solar a bad investment, since the cost numbers already account for it, but it does mean that headlines counting installed megawatts are counting the wrong thing if what you care about is electricity delivered. Capacity factor is the missing piece that turns a misleading number into an honest one.

The Storage Problem That the Decade Has to Solve

The capacity factors expose the deepest challenge renewables face, which is not cost and not geography but timing. Renewable electricity is intermittent. Solar produces only when the sun is up, and wind produces only when the air is moving, yet an electrical grid must balance supply and demand second by second, because electricity is consumed at the instant it is generated. A grid running mostly on renewables therefore needs somewhere to put energy when the sun is high and the wind is strong, and somewhere to draw it from when they are not. That somewhere is storage, and it is the hinge on which a fully renewable grid turns.

Two technologies carry most of this weight. Pumped hydro, which pushes water uphill into a reservoir when power is abundant and releases it through turbines when power is scarce, remains the largest stored-energy resource on the planet by a wide margin. The fast-growing newcomer is the lithium-ion battery, the same chemistry as in a laptop or an electric car, scaled up to grid size. Lazard's 2024 figures put the levelized cost of a four-hour utility-scale battery system at $170 to $296 per megawatt-hour, well above the cost of the solar or wind that charges it. That premium is the real price of intermittency, and it is why storage is the piece the rest of the 2020s has to build. The generation problem is largely solved; the question of when that generation is available is not.

Land, Minerals, and the Concentration of the Build

Building at terawatt scale runs into the physical world in two more ways worth naming honestly. The first is land. A gigawatt of solar photovoltaics needs roughly 5 to 10 square kilometers of ground, a real footprint that has to come from somewhere, which is part of why deserts are so attractive. A gigawatt of onshore wind needs even more total area, but with a crucial difference: the turbines occupy only small footprints, so the land underneath them can stay in agricultural use, with crops or livestock around the bases. The second constraint is minerals. Both technologies depend on critical-mineral supply chains, especially lithium for the batteries that store their output and the rare-earth magnets that sit inside wind turbines. The energy transition is, in part, a mining story, and the geography of those mines is its own subject.

These constraints help explain why the build-out is so geographically concentrated. China alone held about 887 gigawatts of solar and 520 gigawatts of wind at the end of 2024, dominating both manufacturing and installation to a degree that shapes prices and politics everywhere else. The energy transition is global in its ambition but lopsided in its execution, anchored heavily in a single country's industrial base. That concentration is a genuine vulnerability and a genuine source of efficiency at the same time, and reasonable people disagree about which matters more.

Key Takeaways

So can renewables power the world? The honest answer is that the pieces are real and mostly affordable, but the arithmetic has to be done carefully. By the end of 2024, solar photovoltaics had reached about 1,560 gigawatts of installed capacity worldwide, wind about 1,020 gigawatts, and hydropower about 1,410 gigawatts, with solar adding roughly 450 gigawatts in a single year and the whole renewable fleet crossing 4.1 terawatts. Lazard's 2024 levelized cost figures put unsubsidized utility solar at $29 to $92 per megawatt-hour and onshore wind at $27 to $73, both comfortably below new combined-cycle gas, which is why renewables are now the default new build rather than a subsidized luxury. The catch is timing, not cost: capacity factor (about 22 percent for solar, 35 percent for wind, 40 percent for hydro, and 92 percent for nuclear) is the essential bridge between installed megawatts and delivered megawatt-hours, so a one-megawatt solar plant yields around 1,927 megawatt-hours a year rather than the naive 8,760, and intermittency forces an expensive build-out of storage (utility batteries at $170 to $296 per megawatt-hour, alongside the world's pumped hydro). Add the land footprint, the dependence on lithium and rare earths, and China's commanding 887 gigawatts of solar and 520 of wind, and you have everything you need to read a renewable headline the way a geographer does: as a question about specific places, specific technologies, specific delivered electricity, and specific integration costs, where the trade-offs are real but, crucially, calculable.