California Drought: Farms

California is now entering the fifth year of a major drought.  On January 17, 2014, the Governor of California declared a statewide drought emergency. By many measures, 2014 and 2015 have been the worst years of the drought.  Based on the U.S. Drought Monitor, as of December 29, 2015, over 95 percent of California’s $43-billion agricultural sector was experiencing severe, extreme, or exceptional drought, with the livestock sector more directly exposed to exceptional drought than the crop sector (table 1).1/

However, in California, exposure to local water shortages, which is what the Drought Monitor measures, are only part of how the drought is affecting farms.  California agriculture relies heavily on irrigation, and much of the irrigation water is supplied by large-scale State and Federal water projects that store and transport water across hundreds of miles. Therefore, the degree of drought exposure based on local weather does not fully capture the potential impacts. Many other factors, such as surface water availability, groundwater access and cost, irrigation technology choice, crop insurance enrollment, livestock feed availability, and water rights, influence the vulnerability of farms to ongoing drought.

Table 1: Exposure to drought in California
  Market value (2012) Drought exposure (Drought Monitor category) as percent of market value
  ($ billion) Exceptional (D4) Extreme (D3) Severe (D2)
All farms in California 42.6 72.7 14.2 8.3
Crops 30.2 70.8 16.9 8.4
Livestock and products 12.1 78.7  7.7 8.3
Source: ERS calculations of exposure are made using county-level value of production data from the 2012 Census of Agriculture and the December 29, 2015 county-level Drought Monitor data. Market value numbers are from the USDA 2012 Census of Agriculture.

Data on irrigation in California come from a variety of sources. The most comprehensive numbers on the extent of irrigation are from the USDA Census of Agriculture, conducted every 5 years. In 2012, about 7.9 million acres in California were irrigated, representing a third of the total land in farms and the majority of harvested cropland. The best State-level estimates on water sources used for irrigation, as measured at the farm level, come from USDA’s Farm and Ranch Irrigation Survey (FRIS) conducted following the Census of Agriculture.  Differences in water sources used are a major determinant of farm-level vulnerability to the ongoing drought.

For irrigation in California, farms may rely on on-farm surface water, off-farm surface water, groundwater, or some combination of these sources.  As reported in the 2013 FRIS, farmers in California applied 9.8 million acre-feet of groundwater, 1.8 million acre-feet of on-farm surface water, and 11.9 million acre-feet of off-farm surface water.2/ Only 3 percent of irrigated acres used on-farm surface water (fig. 1) as their sole source of water. About 19 percent of the irrigated acres in California were irrigated solely with groundwater and so are not as influenced by the reduction in surface water supplies. Another 52 percent of irrigated acres were irrigated from multiple sources, primarily a combination of off-farm surface water and groundwater. For these farms, reductions in surface water deliveries may be partially or wholly offset during drought years through increased pumping.  Within this acreage, there is wide variation across California in the extent to which local aquifers have been pumped or recharged over recent years.3/ Lastly, about 26 percent of irrigated acres in 2013 in California were irrigated solely with off-farm surface water. These farms are particularly vulnerable to drought conditions in the watersheds where source water supplies are located and may not have the ability to irrigate when off-farm surface water deliveries are curtailed.

Figure 1

Off-farm surface water in California is collected, stored, and allocated through a complex mix of infrastructure and institutions.  Farm-level impacts of reductions in surface water supplies will differ widely depending upon where a particular farm is located and its established water right priority. The two largest over-arching mechanisms for delivering surface water are the State Water Project and the Federal Central Valley Project. The State Water Project manages reservoirs, aqueducts, and pumping plants and delivers water to 29 local water agencies according to the terms of long-term water supply contracts. The Federal Central Valley Project delivers water to hundreds of different contracts, including some of those that also receive water from the State Water Project.  On average, 70 percent of annual State Water Project supplies go to urban users and 30 percent to agricultural users. In contrast, the Central Valley Project, managed by the U.S. Bureau of Reclamation, allocates, on average, about 70 percent of its delivered water to agriculture (5 million out of 7 million acre-feet).

In February 2014, according to the California Department of Water Resources, snowpack was at approximately 24 percent of normal levels, averaged across the State.  On January 31, 2014, the State Water Project, which provides surface water to about 750,000 acres of agricultural land, had announced that it would not be able to deliver any water to these users, the first so-called “zero allocation” in the project’s history. (However, storms in February and March of 2014 ultimately allowed the State Water Project to deliver 5 percent of contracted amounts.) The Federal Central Valley Project, which supplies about 3 million acres, delivered only 20 percent of contracted water in 2013 and no water to agricultural users in 2014. As of April 6, 2015, snowpack was at 6 percent of normal, a record low. The Federal Central Valley Project indicated that 2015 would be another year with no water delivered to agriculture users , although that announcement addressed agriculture users holding contracts with the Central Valley Project, which constitute mostly the junior water rights in the system.  As of March 2, 2015, the State Water Project was projecting 20 percent of deliveries in 2015. Moving into 2016, snowpack is slightly above average for early in the year.  However, it is still too early for either project to have indicated whether or how much they might be able to increase deliveries in 2016.

Since reported delivery percentages are made relative to allocations (contracts), which can vary from year to year, the recent shortfalls in deliveries are easier to see in the context of historical deliveries. To provide this comparison, we tally the total deliveries to selected agricultural categories by the State Water Project, and to the overall Central Valley Project.  Deliveries from both projects  (as a percentage of average historical deliveries) were down modestly in 2012 and 2013, and then dropped dramatically in 2014. Projections for 2015 suggest that deliveries were similar to those in 2014.  In this historical context, the current drought has had a somewhat larger impact on surface water deliveries than the 1991-1992 drought (fig 2) or the 1977 drought (not shown). While surface water from these projects is delivered through much of the State, the impacts of these reductions are most pronounced in the Central Valley of California.  Farms in Southern California receive much of their surface water from the Colorado River, which has not been as heavily impacted by the current drought.

Figure 2

A major uncertainty about the impacts of the ongoing drought in California is the extent to which farmers will be able to offset shortfalls in surface water deliveries through increases in groundwater withdrawals. Historically, groundwater extraction is much higher in drought years, illustrating that  many farmers in California are able to offset a portion of surface water delivery reductions by increasing their groundwater withdrawals (fig. 3).  Given low recharge rates in some areas, the aquifers are very slow to recover the depleted water resources. Even if groundwater extraction can be used by some farmers, it may lower water-table depths and reduce well yields, leading to increased groundwater extraction costs in future years. In some areas, past groundwater extraction has contributed to land subsidence (i.e., sinking) or saline intrusion, which limits current groundwater extraction. Further, as producers in the Great Plains learned in 2012, groundwater irrigation systems are not always able to provide enough water to meet crop demands during extreme and exceptional droughts.

Figure 3

Two factors expected to have a large influence on eventual drought impacts on different types of California farms are irrigation application rates and crop insurance participation. As examples of how these factors can vary, we look at selected commodity categories or commodities – orchards, forage, vegetables, rice, corn, and cotton (table 2). Together, these six categories used an estimated 22 million acre-feet of irrigation water in 2012, or over 80 percent of total irrigation water in California that year. For all these commodities, irrigated production accounts for the majority of acreage in the State. Irrigation ranges from 81 percent of forage production to essentially 100 percent of vegetable, rice, and cotton production. Crop insurance participation varies across commodities. We can estimate the rates and liability using the 2015 summary of business data from USDA’s Risk Management Agency (RMA) and 2012 acreage data from USDA’s National Agricultural Statistics Service (NASS). Based on these numbers, insurance participation ranges from about 8 percent for forage to between 70 and 97 percent for crops like almonds and tomatoes.

Table 2: Factors influencing drought vulnerability for selected crops in California
  Est. 2012
Water use
(million acre-feet)
Total acreage (1,000)
Percent irrigated
Insured acreage (1,000)
2015 Liability
($ million)
Land in orchards 8.3 3,139 98    
Almonds   936   735 2,902
Grapes   940   555 1,551
Walnuts   329   153 435
Forage 5.1 1,670 81 134 76
Alfalfa   990      
Other hay   600      
Land in vegetables 3.2 1,126 ≈100    
Tomatoes   295   286 667
Rice 2.5 562 100 496 436
Corn 2.0 668 95 168 61
Cotton 1.1 368 100 274 213
Grapes include table grapes for insurance calculations.
Forage includes all hay, haylage, and chop silage hay for acreage and irrigation calculations.
Corn includes grain and silage for acreage and irrigation calculations.
Source: Water use totals are ERS calculations based on 2012 acreage (from USDA Census of Agriculture) and 2013 water application rates (from the USDA Farm and Ranch Irrigation Survey). Acreage totals are from the 2012 USDA Census of Agriculture. The insured acreage and liability numbers are from the RMA Summary of Business data.

1/ The Drought Monitor is a compilation of several different drought indices and additional information, combined on a weekly basis by a rotating group of experts from USDA, NOAA (National Oceanic and Atmospheric Administration), and the National Drought Mitigation Center. The monitor divides drought conditions into five categories of drought, in order of ascending intensity: abnormally dry (D0), moderate drought (D1), severe drought (D2), extreme drought (D3), and exceptional drought (D4).
2/ Source: 2013 FRIS, table 5.
3/ See, for example, USGS reports on water availability in the Central Valley Aquifer.