Methods

A field study utilizing a randomized block design was chosen to quantify the effect of low-salinity shrimp farm effluent on olive trees. This preliminary trial examined three effluent/well water/fertilizer combinations; 1) normal farm management, 2) 100% effluent water irrigation and 3) 100% well water irrigation. Each treatment was applied to 40 olive trees (Olea europaea var. Manzanillo), planted in rows for four months beginning in March 2001.

As olive trees are long lived, with much of the growth occurring in the early years, a young orchard was chosen as the study site. The selected grove is the southern most of all and as a result, is situated closet to the shrimp farm. Trees are planted in rows running in an east-west direction, approximately 10 m on center. Trees in the grove are flood irrigated every 10 - 12 days, as needed, with irrigation water applied from the eastside of the grove. The space between adjacent rows is commonly planted to wheat or sorghum while the trees are immature.

Due to its proximity to the service road, the southwestern corner of the field was chosen as the study site (Fig. 3). The southern most row of trees was not included in the study area to avoid the potential of an edge effect. Four blocks were laid out, each containing three rows of 10 trees. Rows were randomly assigned to one of three pre-selected treatments: 1) normal farm management, which included irrigation with well water and the application of anhydrous ammonia as fertilizer; 2) 100% shrimp farm effluent water as the sole irrigation and fertilizer source; and 3) a negative control consisting of 100% well water with no additional fertilizer applied.

Figure 3.  Experimental plot at Wood Brother’s farm in Gila Bend, AZ used to test the effect f irrigating with low-salinity shrimp farm effluent. ‘A’ indicates normal farm management, ‘B’ is irrigation with 100% effluent water and ‘C’ indicates the negative control 100% well water.

In rows assigned to the 100% effluent treatment and the negative control, water diversion berms were constructed (Fig. 3). Soil was dug from between the number 10 and 11 trees in these rows to connect the irrigation furrows on either side. Removed soil was subsequently used to create diversionary berms, effectively isolating the treatment trees from the main field’s flood irrigation by directing this water into the inter-row spaces that had been planted to wheat during this trial. Rows assigned to the normal farm management treatment, did not have diversionary berms.

Trees in the experimental plot were flood-irrigated every 10 to 12 days as needed, following the schedule of the main field. Observation of the irrigation methods and conversations with the farm management were used to determine the volume of water applied during each irrigation event. It was determined, based on the size of the experimental rows, that each would need to receive approximately 3500 L of water per irrigation event. Due to the distance from the shrimp production ponds it was not practical to pump water to the study site, therefore water for both the 100% effluent and the negative control treatments was hauled from their respective sources to the experimental rows in a 3800-L polyethylene tank. Water for the 100% effluent water treatment was collected from the shrimp farm’s drainage ditch with a portable pump. Well water for the negative control treatment was taken from the shrimp farm’s water supply lines.

During each irrigation event, duplicate irrigation water samples were collected corresponding to the three treatments for macronutrient analysis. Water was collected from both the shrimp farm’s drainage ditch and water supply lines as well as from the irrigation ditch supplying the main olive grove. Samples were analyzed for total nitrogen, nitrate-nitrogen, total phosphorus, potassium and salinity. The method used to test each parameter is listed in the Table 1. A HACH DR-890 (HACH Co., Loveland, CO) was used to measure total nitrogen, nitrate-nitrogen and total phosphorus. Potassium was measured with a Turner Model 340 (Sequoia-Turner Corp., Mountainview, CA) spectrophotometer and salinity was measured with a YSI Model 32 Conductance Meter (Yellow Spring Instruments, Yellow Springs, OH).

Table 1. Analytical methods used to test the macronutrient levels of irrigation water.

In addition to the chemical analysis of the three irrigation water sources, tree growth, soil salinity and soil macronutrients were also monitored. Individual trees were measured for height and stem diameter each month. Tree heights were measured from the ground to the apical meristem of the longest branch, leaves were not used in measuring tree height. Stem diameters were measured 20 cm above the ground with dial calipers. Diameters were taken at the widest point at this height.

Soil samples were collected at the beginning and end of the study to measure nitrate and phosphorus concentrations and soil salinity. Samples were taken with a 1.5-cm soil corer to a depth of 0.5 m. One sample was collected from each of the experimental rows, in a staggered pattern (Fig. 3). Nitrate was extracted from the soil with a 2 M KCl solution. Phosphorus was extracted with Olsen’s Solution (0.5 M NaHCO₃). For both nitrate and phosphorus, the filtrate was collected and analyzed with the same techniques used for the irrigation water (Table 1).

Irrigation water samples (six per sample date) were analyzed separately and results were grouped by source (shrimp farm drainage ditch, shrimp farm supply line or olive grove irrigation ditch) for statistical analysis. Soil data and tree growth data were grouped by treatment for statistical analysis. The statistical software package JMP IN v4 (SAS Institute Inc., Pacific Grove, CA) was used to analyze all data. A one-way ANOVA was applied to the irrigation water data, followed by a linear contrast to separate the means. Tree growth data was analyzed using a repeated measures ANOAVA. Soil salinity and nutrient data were analyzed using a paired sample t-test.