If Only There Were a Crop You Could Grow Without Water
It's not technically a crop because it isn't a plant, but solar panels are the trendy new crop.
The United States has 645 solar farms in operation with a total capacity of 37 gigawatts, according to Department of Energy data. Assuming 10 acres per megawatt, this means solar farms occupy about 370,000 acres. This is about 50% more than the amount of farmland owned by Bill Gates, who owns more farmland than anyone else. It's a little less than the area planted to chickpeas and a little less than lentils.
Most states have at least one solar farm, but southern states have more.
Spurred by renewable generation mandates and rapidly declining costs, solar farming is set to grow. There are currently 340 planned solar farms, some under construction and others at an earlier stage. These new farms would add another 43 gigawatts of capacity and occupy another 430,000 acres.
Many of these solar farms are not on viable farmland, like this one in the Mojave desert east of Los Angeles. It is the largest operating solar farm in the state at 253 MW. It began operation in 2014 and supplies power to Pacific Gas and Electric on a 25-year contract. Contracts such as this, signed at a time when solar was much more expensive, are a big reason California consumers pay more for electricity than consumers in other states.
However, some solar farms do compete with agriculture, such as these ones in the Imperial Valley of California, an important region for early-season lettuce and other specialty crops.
Before digging in further, I should be clear what I mean by a solar farm. I define any solar generation facility with more than 18 MW of capacity as a solar farm. Such a farm occupies at least a quarter section. Most of the squares in the picture above are a full section (i.e., a square mile, or 640 acres), but farmers often plant crops on quarter sections. By setting the threshold at 18 MW, I avoid counting solar panels on the roof of a parking lot, or other small projects.
The decline in water availability due to the drought has raised the specter of solar farms in the Central Valley of California. This is particularly relevant for the Westlands region in the southern San Joaquin Valley. The satellite image below shows the different land in this region relative to the regions to their east. The Tulare Lake, which J.G. Boswell famously drained to grow cotton, also appears distinctive in these images.
In 2020, most of the Tulare Lake bed was still planted to cotton, although there were some pistachios around the edges. The biggest crops in Westlands are almonds, pistachios, and tomatoes. A third of fields in the Westlands Water District were fallow in 2020; there will likely be more this year. (Tomatoes and fallow/idle are similar tan colors in the figure below.)
There are currently 21 solar farms in Fresno County and 16 in Kings County, essentially all in the Westlands region. Together, these farms have 1.6 GW of capacity and occupy about 16,000 acres. This is about 3% of 2020 crop acreage in the Westlands Water District, substantially less than almonds (17%), pistachios (10%), and tomatoes (11%).
Most of the solar farms in Kern County are in the desert on the other side of the Tehachapi Mountains and therefore not competing for land with crops.
Another 13 solar farms are planned for the Westlands region, which would add 1.8 GW of capacity and occupy about 18,000 additional acres. Much of this new capacity will be in the Westlands Solar Park. The first phase of this park is slated for completion this fall, and includes contracted delivery of 50MW to Davis through Valley Clean Energy.
Contract prices for the electricity generated from these solar farms are not public. However, the Public Utilities Commission reports aggregate data on contract prices for renewable electricity. This graph from their 2021 report shows that new solar resources are now contracted at about 3c per kWh, which is a 75% decline in the last decade.
What does 3c/kWh mean on a per acre basis? Using a capacity factor of 30% and assuming 10 acres per MW of capacity, an acre of solar panels would generate revenue of (0.03 $/kWh)*(0.3 CF)*(24 hours)*(365 days)*(100 kW capacity) = $7,884 per acre per year, and supply enough electricity for about 20 homes. This number could be increased by packing the panels more densely than they currently appear to be in these projects.
Revenue from almonds has averaged about $6,000 per acre in recent years. Whether solar panels are more profitable than almonds depends on cost. In this region, water for almonds costs about $1,200 per year in a normal year. A drought-induced increase in water costs could be enough to tip the balance away from almonds.
How many more solar farms will we see in the coming years? In California, I wouldn't be surprised to see the growth rate slow down. The state already has so much solar generation that the wholesale price sometimes goes negative during the day. My colleagues Kevin Novan and Jim Bushnell show in a recent paper how the market value of new solar capacity declines as more solar comes online. If we have enough power to satisfy demand at close to zero marginal cost, what is the incentive to build more solar? The answer at present is policy incentives, and the renewable portfolio standard will continue to spur more solar.
High electricity storage costs are also holding back solar expansion. The landscape will change (literally and figuratively) when batteries become cheap enough for large-scale storage, which will enable solar power to be used when the sun is not shining.
The figures in this article were generated using this R code.