California’s Water War: Part II: Balancing Agricultural and Domestic Water Demands
—–Powerpoint Presentation: UP206A-Final_LiuS
California’s freshwater is a limited resource sought after by both the agricultural community and people. The livelihood of both crops and people depends directly on this availability of water; hence, the two groups are constantly fighting for water use rights. Growing water demands means we must prepare for a future water shortage. To tackle this issue, my project seeks to balance the future agricultural and domestic water demands across California’s counties and the potential impacts on California’s agriculture industry using year 2000 as a baseline.
Temperate weather and rich soil make California an ideal location for farming a wide range of crops year-round (AgClassroom.org). In 2000, California’s agriculture industry generated $15.5 billion, nearly one-fifth of the total agricultural output across the United States ($90 billion). In particular, the commercial vegetable and fruit and nut crop categories were particularly dominant in California’s agricultural make-up. Both crop categories generated more than 50% of the total national output (USDA, 2001). In order to maintain the success of the agriculture industry, a sufficient and reliable water supply is necessary. Agricultural lands across California are dependent on irrigation, thus the irrigation efficiency is critical to maximizing plant growth and dollars generated.
California population projections made by the California Department of Finance indicate that there will be a 76% population increase between 2000 to 2050 (CA DOF, 2003). Correspondingly, that means water demand will also increase proportionally assuming that per capita water demands remain the same as 2000 values. Statewide water use for 2000 was 189 gallons per capita per day. This value seems really high, but it includes domestic indoor and outdoor uses. Typically, about half of domestic water use is for outdoor purposes and a large fraction of the indoor uses is from flushing the toilet (Gleick et. al, 2003).
- Select an appropriate spatial scale based on data availability and sufficient resolution. (County-scale was determined to be the most appropriate.)
- Find baseline year 2000 data for agricultural water use, domestic water use, population, weather station measurements, crop evapotranspiration rates, etc.
- Find county scale population projections for X years in future.
- Scale up the agricultural and domestic water demands based on the projected population change by county.
- Quantify the imbalance between the water demand and the total amount of water available.
- Propose potential planning scenarios for the future and assess the potential impacts on California.
Evapotranspiration, a major component in the water cycle, was used in this study to help proxy the agricultural water demand. Evapotranspiration (ET) is the combination of evaporation from the surface and plant transpiration (a byproduct of photosynthesis). With a sufficient amount of water water and solar radiation, the photosyntehsis process will occur and plants will build plant mass. Essentially, plant mass is what we are after because, the greater the plant mass the greater the amount of money is generated. To maximize profitability, farmers want to maximize the amount of plant growth (evapotranspiration) and minimize growth time. Growth time is unique to each crop and cannot be easily altered, so the amount of plant growth is an easier target. Optimization of plant growth relates to the efficient application of irrigated water and evapotranspiration from that applied water. Based on this relationship, I came up with an irrigation efficiency index which is the ratio of evapotranspiration from applied water to amount of applied water.
Irrigation efficiency = evapotranspiration from applied water / applied water
Per capita water use = (public supply + domestic [private] supply)/total population
% population change = (population 20XX – population 2000)/ population 2000 *100
Annual ETo = Σmonthly ETo
% irrigated area = irrigated area / total area
% non-irrigated = 1 – % irrigated
Projected water demand in year 20XX = 2000 water use * (1 + % change in population between 2000 and 20XX)
- Freshwater is a finite resource. The total available water is assumed to be the sum of agricultural and domestic water use in 2000. This volume serves as a cutoff for projected water demands.
- Water conservation efforts were not considered. This ignored the 20% reduction in urban water use by 2020 statewide goal in Senate Bill X7-7 (2009). It also did not consider technological advancements in irrigation.
- Climate variability and climate change were not considered. Climate variability refers to the large-scale impacts of the Pacific Decadal Oscillation (PDO) and El Niño-Southern Oscillation (ENSO) which affect meteorological variables such as solar radiation, precipitation, air temperature, etc. These are key controlling factors for evapotranspiration.
- Other water uses (industrial, thermoelectric, mining, etc) assumed constant and unaffected. This is unrealistic because population growth should also yield some sort of increase in each aspect. For simplification, I have chosen to not include those changes for this project.
- Freshwater sources and transport (via aqueducts) were not considered. One of the issues within the first California Water War was the allocation of California’s water resources. In that first water war, water was diverted from Owens Valley to Los Angeles at a cost to the farmers who suffered significant losses. Where the water comes from and where it goes were big questions that I did not consider for this project.
- Desalination, or the removal of salt from ocean water, is a potential solution to our water shortage, but the present associated costs are astronomical and not practical for meeting large-scale demands. For this reason, I did not consider it.
- Recycled water was also not considered. The concept of toilet to tap has seen harsh criticism by the general public; however, local water districts are bolstering their self-reliance and increasing recycled water production for non-domestic applications.
- Future domestic water demands will be fulfilled before agricultural demands.
Figure 1: 2000 Total Crop Value and Population across the United States
This layout demonstrates that California is one of the top agriculture states in the nation. Illinois and Missouri are also in that top bracket of $5.1 to $15.5 billion. Total population was put in the background just to see if there would be any sort of correlation between the two variables.
Figure 2: Fractional Change in Population from 2000 to 2050 across California’s Counties
County population projections are shown here between 2000 and 2050. The population density was inserted in the background. More densely populated counties (the urban centers) are projected to have a smaller percentage population growth. This makes sense because some regions are already built-out and cannot handle that population increase. Counties that are projected to grow significantly have low 2000 population densities. Sierra County is the one and only county to have a negative population projection.
Figure 3: 2000 Per Capita Water Use and Top 20% Per Capita Water Use Counties across California
Per capita water use (pcwu) is an indicator of the rate of water used within each region. It normalized the total water (volume) used in each county based on it’s population for an equal comparison. To touch on the topic of water conservation, I further zoomed in on the counties which had the highest pcwu. These tended to be concentrated inland and in the central part of the state. In order to find a trend, the percentage irrigated area was also added to the display. There did not appear to be a clear correlation between the percentage irrigated area and the pcwu.
Figure 4: California Irrigation Management and Information System Weather Stations and % Irrigated Area in each California County
CIMIS weather stations across California are used to estimate meteorological parameters which are used to calculate evapotranspiration. Measurements are controlled by key important factors include: Longitude/Latitude, crop surface, gauge elevation, climate, location (urban/rural), etc. These variables can greatly impact the calculated ET.
These calculated ET values are used to generate county-scale ET estimates for 20 crop types (grain, rice, corn, alfalfa, pasture, potato, etc). Model outputs include: total evapotranspiration, effective precipitation, consumed fraction, evapotranspiration from applied water, and applied water.
Figure 5a: January 2000 ETo Across California
Figure 5b: February 2000 To Across California
Figure 5c: March 2000 ETo Across California
Figure 5d: April 2000 ETo Across California
Figure 5e: May 2000 ETo Across California
Figure 5f: June 2000 ETo Across California
Figure 5g: July 2000 ETo Across California
Figure 5h: August 2000 ETo Across California
Figure 5i: September 2000 ETo Across California
Figure 5j: October 2000 ETo Across California
Figure 5k: November 2000 ETo Across California
Figure 5l: December 2000 ETo Across California
Monthly evapotranspiration maps demonstrate the large variability in water demand throughout the year. In the dry (desert) region in southeastern California, ET rates are higher year-round in comparison with those along the coast and in the northern part of the state. Rates tend to be the highest during early spring through early summer just following the rainy months (January and February). Toward the end of the year, the rates taper off and return back to low ET levels in the 0-4 inch range.
These spatial evapotranspiration rates are important because they show us where freshwater is needed and when it is needed. Transporting freshwater from the source is a big challenge; this includes bringing water from northern California to southern California or from the Colorado River. Other water wars between states along the Colorado River is yet another problem we face in California.
SKILLS: graduated symbols, measurement/analysis (centroid of CIMIS stations), inset map, time-based analysis (monthly ETo), original data (CIMIS stations)
SOURCES: UCLA Mapshare County, State, and Country Boundaries, California Irrigation Management and Information System (station locations and monthly reference evapotranspiration (ETo) data)
**** I wanted to generate a video, but ArcMap kept crashing each time I tried. Sorry.
Figure 6: 2000 Annual Total Evapotranspiration and Precipitation and ET Hot Spots
This layout shows the annual total evapotranspiration and a hot spot analysis to demonstrate the significance. In the background, I also added a kriging interpolation of the annual total precipitation to see how correlated the precipitation input and the evapotranspiration output would be, but it turns the precipitation data was probably missing data. For example, a couple precipitation totals in the mountains were near zero which just doesn’t make sense.
I had hoped to create a model to perform the kriging interpolation and mask in hopes of showing the spatial distribution for each of the monthly ETo plots, but since the data quality was questionable, I chose to just show it for the annual plot. In this case, a model would not only help speed up the process, but also automate the spatially interpolation of an “environment” beyond just the location of the points. This was necessary to cover the entire area of California.
Figure 7: Irrigation Efficiency (left) and Evapotranspiration Water Source (native vs. irrigated)
As previously defined, the irrigation efficiency shows how effective the crop is at utilizing the amount of water irrigated. Greater efficiency indicates that the irrigated water is taken up by the crops effectively. The efficiency is also affected by climate and crop type. Northwestern California is more efficient than southern California probably because of the difference in crop type.
The right plot shows where the evapotranspiration water comes from–native (precipitation) or irrgated. Irrigation water significantly greater than native water throughout the state, especially in eastern part of the state.
Figure 8: Evapotranspiration from Applied Water for Two Major Crops: Alfalfa (left) and Corn (right)
Alfalfa and corn are both significant crops in California. Alfalfa is the most produced crop in terms of volume and corn generates the most money. Total evapotranspiration across the state shows where these crops are primarily grown.
Figure 9: 2000 Water Use and 2050 Projected Water Demand for Agricultural and Domestic Purposes
Agricultural and domestic water demands across California were scaled up based on population growth projections. Agricultural demands are most concentrated in the Central Valley. Domestic water demands in the urban centers demonstrate a smaller volumetric increase because the projected population increase is smaller than in the other counties. Despite that, the volume of additional water demanded in these urban centers often exceeds that of the smaller counties with a greater projected percentage population growth.
I had a hard time finding data and I ended up using data from published reports. This required that I digitize and/or adjust the data into an ArcGIS-friendly format. The CIMIS gauge data was perhaps the most time-consuming because I had click on each gauge’s website in order to extract the longitude/latitude, period of record, etc.
Limited water supply in the future will create a new balance between agricultural and domestic water demands. Assuming that domestic needs are fulfilled first, this means that agricultural water use will experience a significant reduction of water. Of course, this would yield a relative decreased in crop value produced, but the challenge is really to see which crops will get a water allocation. Based on the evapotranspiration data, I developed planning scenarios (approaches) to see how this would affect California’s agricultural industry.
- Scenario I: Maximize crop production by first watering crops with the highest irrigation efficiency
- Scenario II: Maximize crop profit by first watering crops with the greatest crop value
- Scenario III: Maintain crop diversity by applying a proportional reduction in agricultural water
- Scenario IV: Specialize crops (focus on commercial vegetables and fruits & nuts) by first watering crops with the highest irrigation efficiency
These scenarios were designed to represent different interests. Scenario I would generate the most volumetric output of crops so the general public would have more food. Scenario II would produce the greatest amount of money which would benefit the farmers the most. Scenario III would be a neutral way for the government to cut back on water rights to not bias one crop over another. This would also benefit the public as well in providing a range of crops. Sceanio IV is set-up in a way where California would eventually have a monopoly over the growth of commercial vegetables and fruits and nuts. In this case, we could potentially increase the unit cost and generate more revenue. Each scenario is idealized and actual implementation would require a balance among the different interests.
I chose to explore Scenario III just to get an idea of how much money the agricultural community would lose. I found that there would be a $2.4 billion per year reduction in revenue and food shortage or higher cost for food because it would have to come from other states and/or countries. When applied to the other three scenarios, I assume there would also be a loss in annual revenue.
Using GIS, this project was able to spatially locate where the agricultural or domestic water demand is located and where irrigation is most efficient using evapotranspiration data. To improve the usefulness of this project, it would be ideal to have fewer simplifying assumptions. In particular, freshwater sources location and volume available would be important information to use in allocating water. Such information would also account for the amount of energy/money necessary to divert water X distance away from the source. Of course, this information would probably have to be coupled with a predictive hydrologic model in order to estimate the amount of snow, snow water equivalent, snow melt timing, etc. A more comprehensive analysis of this problem would have to account for the dynamic nature of each of these variables.
AgClassroom.org (2010) A Look at California Agriculture. Web. <http://www.agclassroom.org/kids/stats/california.pdf> 19 Mar 2012.
California Department of Water Resources- Water Use and Efficiency Branch. (2011) SB X7-7 Water Conservation Program Status. <http://www.water.ca.gov/wateruseefficiency/sb7/docs/SBX77-ProgramStatus-07-12-11.pdf> 10 Feb 2012.
Gleick, P., D. Haasz, C. Henges-Jeck, V. Srinivasan, G. Wolff, K.K. Cushing, and A. Mann. (2003) Waste not, want not: the potential for urban water conservation in California. Pacific Institute.
Global ENSO SST Index. University of Washington: Joint Institute for the Study of the Atmosphere and Ocean. Web. <http://jisao.washington.edu/data/globalsstenso/> 9 Feb 2012.
UCLA Spatial Data Repository: Mapshare. <http://gis.ats.ucla.edu/mapshare/> 02 Feb 2012.
- California County Boundaries
- U.S. State Boundaries
- Canada and Mexico Boundaries
Annual Land & Water Use Estimates. (2000) California Department of Water Resources. Web. <http://www.water.ca.gov/landwateruse/anaglwu.cfm> 04 Feb 2012.
California Department of Finance. (2007) Population Projections by Race / Ethnicity, Gender and Age for California and Its Counties 2000–2050. Web. <http://www.dof.ca.gov/research/demographic/reports/projections/p-3/> 10 Mar 2012.
Census Population. United States Census Bureau. (2000, 2010). Web. <http://www.census.gov/> 02 Feb 2012.
Crop Value Summary 2000. (2001) United States Department of Agriculture. Web. <http://usda01.library.cornell.edu/usda/nass/CropValuSu//2000s/2001/CropValuSu-02-15-2001.txt> 08 Feb 2012.
Weather station websites and monthly data. (2012) California Irrigation Management Information System (CIMIS). Web. <http://wwwcimis.water.ca.gov/> 11 Feb 2012.