Fresno State Logo                  Welcome to Wateright

  Agroforestry Systems for On-Farm Drain Water Management *
G.S. Jorgensen, Kenneth H. Solomon, and V. Cervinka **

CATI Publication #930102
© Copyright January 1993, all rights reserved

Drainage and Salinity Problems
Some agricultural lands within California's western San Joaquin Valley suffer from waterlogged soils or rising water tables. Nearly 350,000 hectares are currently affected by a high water table (within 1.5 meters of the ground surface), and 400,000 hectares could be affected by the year 2000 (San Joaquin Valley Drainage Program, 1990). Compounding these problems is the fact that soils of the Valley's west side are derived from marine sediments, often containing elevated levels of salts and other elements found in sea water (arsenic, boron, and selenium). As this land is irrigated, these elements are dissolved and become constituents of the drainage water. And even with the high-quality water provided by the state and federal water projects, imported total salt (1,600,000 tons/year *** ) and associated trace elements (boron, selenium, others) are major problems.

The San Luis Unit of the Federal Central Valley Project (CVP) and the State Water Project (SWP) began providing water to 400,000 hectares in west Fresno County in 1968. As part of the San Luis Unit, a master drain was mandated to move agricultural drainage water to the Sacramento/San Joaquin Delta. Construction of the drain began in 1968, and it eventually collected drain water from about 18,000 ha. But due to funding problems and environmental concerns over the effects of drain water discharge into the Delta and San Francisco Bay, the San Luis Drain was never completed. It was terminated near Los Banos at a regulating reservoir, Kesterson, which had become part of a national wildlife refuge.

Reproductive abnormalities and deaths of aquatic birds were discovered at Kesterson National Wildlife

Refuge, and were found to be related to the trace element selenium, a contaminant associated with agricultural drain water in the area. The drains emptying into the San Luis Drain were plugged in 1986 by order of the Secretary of the Interior. Kesterson has since been covered, and lands once served by the San Luis Drain no longer have an outlet for agricultural drainage water. Growers in the affected area must now manage the drain water on-farm.

The drought in California the past 6 years has intensified the water quality aspect of the drain-age problem. Water supplies in the CVP and SWP reservoirs have been insufficient to supply the normal allotment to many users. To make up the deficit, many growers are relying on poorer quality ground water, recirculating the salts and other elements back into the crop rootzone.4

Drain Water Management
Several options exist for growers and drainage districts to manage agricultural drain water. Among these options are: source control, ground water management, drainage water treatment, and drainage water reuse, which includes agroforestry.

SOURCE CONTROL--This involves minimizing the contribution of irrigation water to the underground water table by improving irrigation systems and present irrigation management practices. Estimates (San Joaquin Valley Drainage Program, 1990) of the amount of water added to the underground water table from unnecessary deep percolation range from 0.18 to 0.23 meters/ year (assuming 0.09 meters/year are required for leaching).

GROUND WATER MANAGEMENT--In areas where relatively good quality shallow ground water exists, this water could be extracted and applied directly to crops, blended with high- quality surface water, or the ground water levels could be managed to facilitate direct crop uptake and use of this water.

TREATMENT--Various treatment options have been studied, and while some have shown promise, none are yet feasible. Treatments include bacteria based processes, microbial volitization, geochemical immobilization, heavy metal absorption with iron filings, ion exchange, and reverse osmosis. The treatment systems examined suffer from either poor economics or difficulty in maintaining sustained operation of the equipment and systems.

Agroforestry is a biological system for managing agricultural drainage water (Cervinka, 1990). It represents a crop-integrated approach where increasingly saline water is applied to successively more salt-tolerant plants. As the salt concentration increases, the volume of drainage water requiring ultimate treatment or disposal is decreased (see FIGURE 1, from San Joaquin Valley Drainage Program, 1990).
Figure 1.
For example, good quality water is applied to a salt-sensitive crop such as carrots. The drain water from below this crop is captured by a subsurface tile drain system, and when feasible, applied to a more salt tolerant crop such as cotton. The ensuing drainage water, now further reduced in volume and concentrated in terms of salt and other elements, is then applied to salt-tolerant trees such as eucalyptus. The drainage water is again captured and applied to halophytes such as saltbush ( Atriplex canescens ). The resulting highly concentrated drain water is captured yet again for final disposal in evaporation ponds, deep well injection, or further treatment. As the cost of transport and some treatment options are volume sensitive, the cost of the ultimate disposal of the drainage water may be reduced by employing agroforestry. The agroforestry system offers a management disposal option that is less problematic than large-scale evaporation ponds, which may be classed as toxic sites, with potential benefits for wildlife habitat.

In particular circumstances, additional benefits may be possible. Even though selenium in high concentrations is toxic, it is required in lower concentrations to maintain the health of some animals. Many cattle on the east side of the San Joaquin Valley, and elsewhere throughout California and the West, require selenium supplements to maintain adequate blood levels of the element. The atriplex grown in the agro-forestry system has been shown to accumulate the element selenium, and feeding trials have shown that atriplex hay is capable of maintaining required blood selenium levels in forage cattle (Frost, 1990).

Agroforestry plantings at various test sites were initiated in 1985 in a cooperative research effort of the USDA-SCS and the California Department of Food and Agriculture, Agricultural Resources Branch. Experimental plantings are being investigated for the following applications: to intercept drainage water flowing out of high ground into the valley; to lower shallow ground water (passively); and as a method of managing agricultural drain water produced from crop lands (Cervinka, 1990). Extensive work is underway to genetically improve the eucalyptus trees, and to identify other promising salt-tolerant atriplex and grass species. Trees are selected for better salt and trace element tolerance from plantations throughout the San Joaquin Valley and propagated by cloning. Cloning provides an exact replica of the mother plant, and reduces the variability in trees produced from seed.

Agroforestry Plantation, Mendota, California

The most intensively monitored agroforestry plantation is a site near Mendota, California, 75 kilometers west of Fresno. The site consists of 9.43 hectares of Eucalyptus camuldensis , planted in 1985 and 1986, and 2.02 hectares of halophytes (FIGURE 2). Data collected at the site includes chemical composition of the ground water, depth to ground water, soil moisture status via tensiometers and neutron probe, volume and chemical composition of drain water applied to the trees, composition and volume of the ensuing drain water from the trees that is collected and applied to the atriplex, Sodium Absorption Ratio (SAR) of the soil, tissue analysis of the trees and plants, and the composition of the salts collected in the evaporation basins.

The experiment is continuing. However, a summary of what has been learned thus far is given below.

Permeability of the soils is restricted physically by a high clay content and chemically by a high sodium content in the top layers. To help overcome this, irrigation water was not applied in discrete events; rather, water during the irrigation season was applied almost continuously, by adding water to a series of basins. Additional water was applied during the winter months. In soils where permeability is restricted, applications during the winter may be necessary to provide sufficient leaching to maintain salt balance in the soil profile under the trees.

FIGURE 2. Agroforestry Plantation, Mendota, California
Saline agricultural drain water from the Westlands Water District collector system (average EC of 10 dS/m, 12 mg/L boron, 400 mg/L selenium, SAR of 11 mM/L 1/2 ) was used to irrigate the trees. The site has three separate subsurface drain systems: a perimeter drain around the entire plantation to intercept flows into the planting; a system under the trees to collect any water passing beneath the rootzone of the trees; and a system beneath the atriplex and halophyte planting. Individual sumps with water meters measure the volume of drain water from each drain system.

Research Results and Observations from the Mendota Site
CONCENTRATING EFFECT--Drain water collected from under the eucalyptus trees is reduced in volume and more concentrated. The reduction in water volume causes increases in drain water characteristics as follows: EC, 3.2 times; SAR, 6.3 times; boron concentration, 4.2 times; and selenium concentration, 1.8 times.

SOIL SALINITY--Irrigation during the 1987 through 1989 growing seasons was done by the farm cooperator, and was subject to the availability of irrigation labor not otherwise occupied on the farm. The amount of water applied during these years met neither tree water needs nor leaching requirements, resulting in a dramatic rise in soil salinity and eucalyptus leaf tissue boron concentrations (in excess of 2000 mg/L). Soil ECe (averaged over the top 2.4 m of the soil profile) increased from just above 10 dS/m in 1987 to just under 30 dS/m at the beginning of the 1990 season. During the 1990 irrigation season, adequate drain water was applied to supply the needs of the trees plus a 16 percent leaching fraction, after which the average soil ECe fell to 25 dS/m. Additional leaching accomplished by irrigating during the fall and winter months of low evaporative demand was successful in lowering the ECe to 18 dS/m by June 1991 (Tanji, 1992).

TREE WATER USE--During the 1990 season, evapotranspiration from the trees was estimated using a Bowen ratio energy balance method. Utilizing data from the California Irrigation Management Irrigation System weather station located nearby, a crop coefficient (Kc) of 0.84 was derived (Tanji et al., 1990). Although some sources suggest that trees irrigated with good quality water under optimum conditions may have Kc values in the range of 1.2 or higher (UC Cooperative Extension, Undated), the trees in the study transpired significantly less, perhaps due to increased soil salinity levels.

PLANS--The plantation was harvested and sold as chips for biomass during the summer of 1992. The original planting consisted of rows spaced 1.5 m apart, and was modified to provide 3 m row spacing by completely removing alternate rows. The wider spacing will allow for mechanical cultivation which will benefit both weed control and water penetration. The remaining trees will be allowed to regrow, and will be studied to determine water use rates and regrowth rates. A portion of the planting was removed completely, and replanted with trees that have been selected for their superior performance since the advent of the agroforestry program.

ROOTZONE SALINITY MANAGEMENT-- Irrigation scheduling will be critical to the success of agroforestry. On low infiltration rate soils, the rootzone may be used as a seasonal salt storage area. During the irrigation season, adequate water would be applied to meet the needs of the trees. But without excess water for leaching, the soil salinity will increase. During periods of low evaporative demand, for example winter, irrigations would then continue, in order to leach the accumulated salts from the rootzone.

THE IMPORTANCE OF AGROFORESTRY AND RESEARCH FUNDING--Agroforestry research has suffered from insufficient funds. This is surprising as well as disappointing, since the San Joaquin Valley Drainage Program study identified agroforestry as playing a major role in the management of drainage water within the Valley. All of their considered and recommended policy scenarios require substantial areas of agroforestry. Depending on the scenario, estimates of the area under agroforestry range from 8,500 to 13,000 hectares by the year 2000, and from 13,700 to 21,800 hectares by 2040 (San Joaquin Valley Drainage Program, 1990). Yet research funds available have not been adequate to meet the identified needs of the agroforestry research program, and have in fact been declining for the past few years. More work is required on the tree and halophyte tolerance to salinity and elements such as boron, on tree selection and propagation, on factors affecting the salt and water balance of the system, and on disposal or use of the remaining effluent and salt.

ECONOMICS--The potential value of possible agroforestry products (biomass, honey, essential oils, etc.) will depend on local markets, and may not be high. The biomass produced to date at the Mendota site has neither sufficient volume nor appropriate characteristics to be of any value in the local market. If, however, the agroforestry concept can provide a means of on-farm drain water management, thus alleviating the need for expensive and potentially hazardous evaporation ponds, then perhaps maintaining the San Joaquin Valley's west side as a viable farming region will be sufficient economic justification.

Cervinka, V. 1990. A farming system for the management of salt and selenium on irrigated land (Agroforestry). California Department of Food and Agriculture, Agricultural Resources Branch, Sacramento, CA, May 1990, 17 p.

Frost, B. 1990. Atriplex tested as feed option. San Joaquin Experimental Range Newsletter, Spring 1990, California Agricultural Technology Institute Pub, No. 900304, California State University, Fresno, California, pp. 1, 3.

Karajeh, F.K. 1991. A numerical model for management of subsurface drainage in agroforestry systems. PhD Dissertation, Univ. of Calif., Davis.

San Joaquin Valley Drainage Program. 1990. A management plan for agricultural subsurface drainage and related problems on the Westside San Joaquin Valley, Final Report, 1990. E. Imhoff, Program Manager, Sacramento, CA 183 p.

Tanji, Kenneth K. 1992. Will agrofrestry help solve drainage problems? USCID Newsletter, January 1992, No. 65, US Committee on Irrigation and Drainage, Denver, Colorado, p. 4.

Tanji, Kenneth K., S. Grattan, A. Dong, F. Karajeh, A. Quek, D. Peters, D. Johnson, and G. Jorgensen. 1990. Progress report on water and salt balance, agroforestry demonstration program. California Department of Food and Agriculture, Agricultural Resources Branch, Sacramento, CA, October 1990.

UC Cooperative Extension. Undated. Using reference evapotranspiration (ETo) and crop coefficients to estimate crop evapotranspiration (ETc) for trees and vines. Leaflet 21428, University of California, Division of Agriculture and Natural Resources, Cooperative Extension, Berkeley, California