As implementation of the 2014 Sustainable Groundwater Management Act (SGMA) proceeds, it’s no secret that the San Joaquin Valley will have to adapt to a future with less water for irrigation. Our research shows that overall irrigation supplies may decline by as much as 20% by 2040. Land uses will have to change, and some have raised concerns that SGMA’s implementation could put smaller farms at a disadvantage, given their more limited resources and capacity. To gain insight on these issues, we conducted a detailed geographical analysis of cropping patterns and water conditions by farm size on the San Joaquin Valley floor, using county real estate records on ownership of agricultural parcels (individual properties of varying sizes) to identify farms.

In this first blog post, we explore where San Joaquin Valley farms of different sizes are and what they grow.  We provide more details in an accompanying dataset. A subsequent post will delve into water issues facing farms of different sizes in the SGMA era.

A comparison with USDA’s Agricultural Census

Our dataset enables a fresh look at farm size when compared to the other main source, the USDA Agricultural Census. The Ag Census gathers numerous types of information from farmers every five years (e.g., estimates of income, tenancy, race/ethnicity), but it tends to undercount smaller farms due, among other reasons, to non-response. And the Census is only reported at the county level—too coarse a scale for understanding SGMA issues, since water conditions vary considerably within counties. In contrast, although we are not able to distinguish between lands that are owner-operated and those that are rented, our unique dataset allows us to look at numerous issues of interest at the parcel level: where farms of different sizes are, what they grow, the water situation in their area, and how that could change under SGMA.

We consider irrigable acreage (including both cropped and idled land) as the measure of farm size. We identify roughly 34,500 distinct farms—a much larger number than the 2017 Ag Census (about 19,000 irrigated farms); this principally reflects a much larger number of very small farms (those with less than 50 acres) in our data (see the dataset for a comparison). We also count fewer very large farms (those with more than 2,000 acres), which likely reflects the fact that some farms—especially in the small- and medium-size categories—are rented to and managed as part of larger farming operations. (Some small farmers also rent their land, but likely from owners of small holdings.)

Most farms are small, but larger farms occupy the most acreage

Farms in the San Joaquin Valley come in all sizes, from large enterprises spanning multiple basins to single parcels with fewer than 10 acres. Valley-wide, more than three-quarters (77%) of all farms are small, with less than 100 acres of irrigable land. But most cropped acreage (58%) is held by large farms, those with 500 or more acres. (Ag Census data reports 66% of the valley’s farms as small, and 75% of irrigated acreage in large farms.)

While small farms are present across the valley, they make up a much bigger share of all farms—and a larger share of cropped acreage—along the valley’s east side, especially in the Eastern San Joaquin, Modesto, Turlock, and Kings groundwater basins. In contrast, larger farms have a greater presence in the west and south (see the figure below).

Larger farms are more likely to consist of multiple parcels of land—distinct properties that are not necessarily adjacent. On average, farms with fewer than 100 acres have just 1.5 parcels, while farms with over 500 acres average 16 parcels. Larger farms are also more likely to hold lands in more than one groundwater basin. Valley-wide, roughly 1,600 farms are in multiple basins, including nearly 600 large farms. Multi-basin farms account for roughly one-third of all farm acreage in the valley, with an especially large presence in Kern, Tulare Lake, and Westside basins. Farms in multiple locations face more complicated operating conditions—including policy differences across water districts and counties. But their geographic diversity can also make it easier for them to diversify farm portfolios and adapt to changing conditions.

Smaller farms are more specialized

With less acreage to work with, it is not surprising that small farms are less diverse in their crop choices than larger farms. On average, small farms grow just over one crop per farm (out of a total of 23 crop categories), while the largest farms average around three crops (see figure below). Small farms are also more likely to invest in perennial fruit and nut orchards, although some of the smallest farms specialize in alfalfa and pasture and other field crops. (Some small farms, including those of immigrants from Southeast Asia in the Fresno area, specialize in a mix of fresh produce for local and urban markets—a level of diversity that is hard to pick up in the land use data.) Larger farms are also heavily invested in perennial trees and vines, but they are much more likely to grow a variety of field and grain crops (e.g., cotton, wheat, safflower) and some vegetables (e.g., tomatoes).

With small acreages, the chance of earning enough income to support a household is much greater with perennial fruits and nuts and some specialty vegetables, which bring in higher returns. Some of our regional advisors have suggested that farms below 100 acres generally would not provide adequate income for a household unless they were planted in such crops. Of course, not all small farms operate as a primary business; many are owned by people who maintain small acreages for some supplemental income or as a rural lifestyle choice.

Farms of all sizes will face pressures in adapting to SGMA. The experience of small farms may vary from that of larger farms due to differences in crop choice and their ability to diversify their operations across different settings. Also crucial will be the water conditions they face—including the degree of cutbacks that may be needed to address groundwater overdraft as SGMA implementation progresses. We examine this issue—along with the implications of policies to allocate and trade water—in an upcoming blog post.

This work was done by a PPIC team that also included Alvar Escriva-Bou, Spencer Cole, and Joy Collins. We are grateful for the contributions of Jeff Allenby and Hallah Elbeleidy from the Lincoln Institute of Land Policy’s Center for Geospatial Solutions, who worked with us to create a dataset on agricultural parcels. This work was supported by the Lincoln Institute of Land Policy and the USDA Secure Water Future grant at UC Merced.