Development of sustainable vegetable production systems for south Florida and Virginia based on use of cover crops and precision irrigation

Development of sustainable vegetable production systems for south Florida and Virginia based on use of cover crops and precision irrigation

Summary

Virginia Component

Virginia Component. In 2004, yield and pest management of organic-grown Irish potato were assessed in four tillage treatments: Marketable Yield
Control (no cover crop, with conventional inversion tillage); …………18.7 t/ha
Cereal rye/hairy vetch cover crop mixture, with no tillage (R/HV); …. .23.3 t/ha
Barley/hairy vetch, no-tillage (B/HV); and …………………………….24.2 t/ha
Austrian winter pea/hairy vetch, no tillage (AWP/HV). ………………..21.5 t/ha
Marketable tuber yield was 18.7 for the control, 23.3 for R/HV, 24.2 for B/HV, and 21.5 t/ha for AWP/HV. Average yield in no-till plots (23 t/ha) was 23% higher than the control (18.7 t/ha).

Yield differences among treatments are believed to have occurred because of superior weed control and enhanced plant-available nitrogen in no-till plots. Incidence of Colorado potato beetle (CPB) was very low and is attributed to high-residue mulch in no-till and farmscape plantings in all treatment plots.
.) and use one of the styles available below.

Florida Component

Progress was made in developing an alternative pepper production system without the use of methyl bromide soil fumigant. Excellent yields of peppers were obtained by growing a nematode-resistant summer cover crops and covering the beds with a thick layer of compost. However this system failed to suppress yellow nutsedge, which had to be had-weeded twice.
Also progress was made in developing an alternative tomato production system without the use of methyl bromide soil fumigant. In a field lightly infested with plant parasitic nematodes but heavily infested with weeds, equally high yields of tomatoes were obtained with plastic and organic mulch. Various summer cover crops were grown on these plots during the fallow period. In the second year S-metolachlor (which is effective against nutsedges) and napropamide (which is effective against broad leaf weeds) were applied at the label rates to the areas designated to receive the organic mulch. The herbicides were sprayed onto the surface of the beds and then watered in with 0.64 cm of water. On two subsequent dates a tank mix at label rates of halosulfuronmethyl (effective against broadleaf weeds and nutsedges) and sethoxydim (effective against grasses) was applied.
Two precision irrigation systems were developed and evaluated for use on highly porous soils with poor water-holding capacities. Both systems are soil moisture based automatic, high-frequency/low-volume (HFLV) drip irrigation systems suitable for producing tomatoes in Krome gravely loam soil. One system utilized switching tensiometers, which require frequent servicing, and the utilized a solid state probe, which is essentially maintenance-free. The performance of each these two precision systems was compared to that based on evapotranspiration and on once per day irrigation commonly used by local growers. The use of these precision systems resulted in savings of irrigation water of more than 70 percent, and reduced the amount of nitrate nitrogen and phosphorus lost through leaching from the root zone by a somewhat similar amount. The tensiometer-based system resulted in higher tomato yields than the solid state probe-based system, and this suggests that the calibration of the latter against tensiometer performance needs to be improved. In addition these system need to be programmed to deliver appropriate amounts of liquid fertilizer during periods of low evapotranspiration.

Objectives/Performance Targets

Virginia Component

Objectives: Assess the effects of an organic-based (high residue no-till cover crop mulch) on tuber and pest incidence of Irish potato. Disseminate research finding and facilitate adoption of high residue no-till systems for production of organic potatoes.

Performance targets: We anticipate that (1) cover crops would produce a high residue mulch that would suppress weed growth and lower incidence of Colorado potato beetle below crop-yield limiting levels; and (2) tuber yield would be equal or exceed that obtained in the control plots.
1. Develop sustainable production systems for tomato, and pepper each based on use of nematode- and pathogen-resistant cover crops (cowpea, sorghum sudangrass, sunn hemp, velvetbean) instead of chemical soil sterilants such as methyl bromide.

2. Assess the effects of a cover crop based system on (a) crop yields and (b) population densities of plant parasitic nematodes, weeds and other serious pests.

3. Conduct research to reliably attain major gains in crop yields through science-based management of irrigation, fertigation and improvement of soil quality. Study the feasibility of using automated irrigation system and soil moisture sensors to maintain the optimum moisture level in the root zone, and prevent leaching of nutrients.

4. Develop enterprise budgets of the cover crop-based production systems vs. those based on use of methyl bromide. Determine social or economic constraints to adoption of advantageous systems, and identify appropriate measures to facilitate adoption if warranted.

5. Disseminate research findings and facilitate adoption of sustainable vegetable production systems in Florida, Virginia and other southern states.

Accomplishments/Milestones

Virginia Component

Research – first-year organic tuber yields were excellent; no-till treatments averaged 23% higher than conventional plots. Suppression of weeds and incidence of Colorado potato beetle were highest in no-till plots.

Florida Component

1. Development of an alternative pepper production system.
Objective: To assess the effects of cover crops, compost and methyl bromide-chloropicrin fumigants on pepper yields.

Materials and Methods: A randomized experiment with 4 replications was used to compare the effects on pepper yields of three cover crops, fallow alone and in combination with compost application or fumigation with methyl bromide + chloropicrin. Cover crops (sunn hemp, velvetbean or sorghum sudangrass) and fallow were the main treatments on plots each 24.3 m long. The treatments were applied to raised beds each 0.9-m wide and 1.8-m center-to-center. A distance of 5 m separated the various treatments. After the cover crops had been terminated, they were incorporated into the soil by means of a rototiller. The beds were re-formed and 6-12-12 fertilizer was incorporated at 1679 kg ha-1. Each 24.3 m-long plot was subdivided, and a section of bed 9.9 m long was fumigated at the label rate with MC-33 and covered with polyethylene mulch, another 9.7 m-long section received no fumigation but was covered with plastic mulch, and a 4.9 m-long section was covered with a layer of compost 8.9 cm thick. The latter section was not fumigated and was not covered with plastic mulch. Two parallel drip lines 30 cm were installed on the surface of the bed beneath the polyethylene and over the compost layer.
The cover crops were seeded on June 14, 2004. Seeding rates for velvetbean, sunn hemp and sorghum sudangrass were 34, 45 and 45 kg ha-1 of germinated seed, respectively. Cover crop biomass samples were taken on August 5, 2004, and the cover crops were flail-mowed on August 17, 2003. The beds were fumigated on September 23, 2004, and immediately provided with two drip lines and covered with the plastic mulch. Compost had been applied to the appropriate sections of the beds on September 21, 2004. On October 12, 2004 pepper cultivar ‘Camelot’ transplants were placed into the bed in two rows each 10 cm from the shoulder of the bed with 30.4 cm between plants. The beds were fertigated through the drip lines beginning at the beginning of fruit set twice per week with 10-0-46 at a rate of 7.8 kg ha-1 of N. The plots covered with compost were hand weeded twice because of dense infestations of yellow nutsedge. Pepper fruits were harvested four times between December 16, 2004 and January 28, 2005.

Results: As shown in Table 1, the lowest yields of pepper were obtained on fallow. As expected, fumigation with methyl bromide-chloropicrin (MC-33) improved the yield, but the addition to fallow of the 8.9 cm-thick compost layer without fumigant increased the yield to a similar extent. Both the sunn hemp and velvetbean treatments increased the yields over that obtained with fumigation. However the combinations of sunn hemp and compost and sunn hemp and velvetbean resulted in the greatest yield increases.

Table 1. Marketable yields and average number of fruits per 30 ‘Camelot’ pepper plants as affected by various treatments.

Treatment Total Marketable Yield t/ha Total Number of Fruit (mean)
Fallow 8.1 41.8
Fallow+ compost 12.7 86.5
Fallow + MC-33 13.1 60.7
Sunn hemp 12.8 60.3
Sunn hemp + compost 20.4 98.5
Sunn hemp + MC-33 24.3 67.0
Velvetbean 11.9 57.3
Velvetbean + compost 25.6 115.8
Velvetbean + MC-33 14.4 66.0
Sorghum sudangrass 8.7 47.8
Sorghum sudangrass + compost 17.2 85.5
Sorghum sudangrass + MC-33 12.4 56.5

2. Development of an alternative tomato production system.
Objective (2003-2004): To assess the effects of cover crops, and compost and on tomato yields.
Materials and Methods: A field experiment to compare the effects of four summer cover crops (sorghum sudangrass, sunn hemp, velvetbean and cowpea) plastic mulch, and three rates of compost (25 Mt ha-1, 50 Mt ha-1, 50 Mt ha-1 for 2 consecutive years and 100 Mtha-1 the growth and fruit production of a subsequent winter crop of tomatoes was conducted at the Tropical Research and Education Center, Homestead, Florida. The cover crops were seeded in June 2003, and incorporated into the soil in October.
A split block design was selected for this experiment, the main plots were cover crops and the subplots were different mulches. Each plot consists of 15 m long and 0.9 m wide raised bed and each treatment was repeated 4 times. In June 2003 raised beds 15-cm high, 91 cm wide, and 182 cm between centers were formed in a field of shallow gravelly soil. Since the soil was known to be nutrient-deficient, 900 kg ha-1 of 6-6-12 (N-P2O5-K20) dry fertilizer was rototilled into the beds. Four summer cover crops were sown on June 6, 2003 with a Tye no-till drill (AGCO Corp., Lawrenceville, GA). Three of these cover crops were legumes: sunn hemp cv ‘Tropic Sun’), cowpea cv ‘Iron clay’, velvetbean, and one was sorghum sudangrass (Sorghum bicolor × S. Sudanese (Piper) Stapf). The seeding rates were sunn hemp, 56 kg/ha; cowpea, 112 kg/ha; velvetbean and sorghum sudangrass, 45 kg/ha. To destroy the apical dominance in sunn hemp as a means of promoting branching and increasing biomass production, the sunn hemp were cut at 30 cm above the ground in August, 2003. All the cover crops were terminated by flail mowing in October 2003 and the residues were incorporated into the soil. tomato, cv. ‘Sanibel’ seedlings were transplanted to the beds at 51 cm spacing.
The T-tapes were connected with layflat hose for irrigation. Halosulfuronmethyl post emergence herbicide was applied once to control the weeds on the beds with organic mulch. Standard practices to stake, tie and prune tomatoes, and to control foliar insects, diseases, and weeds between adjacent beds were applied (Maynard and Olson, 2000). At tomato flowering, a representative plant from each plot was taken to measure tomato biomass.
The fruits were graded following Florida Tomato Committee Standards (Brown, 2000). The fruits were separated into extra large, large, medium and culls after each harvest and the marketable and total fruit yields were calculated.
Soil samples were taken when the cover crops were seeded, and when they were terminated, and finally at tomato flowering. These samples were used to identify soil nematodes and to measure soil nutrient contents.
Tentative conclusions from results obtained in the 2003-2004 experiment:

1. Sunn hemp significantly improved tomato total marketable yield (Fig. 1).

{Contact Southern SARE office for Word Doc with figures and tables.}

Fig. 1 Tomato total marketable yields with cover crops
Cover crops: SH, sunn hemp; VB, velvetbean; CP, cowpea cv ‘Iron & Clay’; SS, sorghum sudangrass; and FA, fallow. Ct/acre is 25-pound cartons/acre.

2. Sunn hemp and velvetbean, especially sunn hemp significantly improved tomato marketable yield from the 1st harvest (Fig. 2), which can increase the growers’ income because of prices usually are favorable early in the season.

Fig. 2 Tomato marketable yields from the 1st harvest with cover crops
Cover crops: SH, sunn hemp; VB, velvetbean; CP, cowpea cv ‘Iron & Clay’; SS, sorghum sudangrass; and FA, fallow. Ct/acre is 25-pound cartons/acre.

3. Organic mulch applied for 2 consecutive years significantly improved the total marketable yield if applied it in two years (OM22) compared with a single year but there was no a significant difference found between different rates of application (Fig. 3).

Fig. 3 Tomato total marketable yields with different mulches
PM is plastic mulch. Rates of application of compost: OM1, 25 Mt ha-1; OM2, 50 MT ha-1; OM22, 50 Mt ha-1 applied in 2 consecutive years, and OM3, 100 Mt ha-1. Ct/acre is 25-pound cartons/acre.

4. Tomato marketable yields were significantly increased by the application of organic mulch in two years or with plastic mulch (Fig. 4).

Fig. 4 Tomato marketable yields from the 1st harvest with various mulches
PM is plastic mulch. Rates of application of compost: OM1, 25 Mt ha-1; OM2, 50 MT ha-1; OM22, 50 Mt ha-1 applied in 2 consecutive years, and OM3, 100 Mt ha-1. Ct/acre is 25-pound cartons/acre.

5. The total extra large fruit yield was significantly increased by growing and incorporation of sunn hemp (Fig. 5).

Fig. 5 Tomato total extra large fruit yields with cover crops
Cover crops: SH, sunn hemp; VB, velvetbean; CP, cowpea cv ‘Iron & Clay’; SS, sorghum sudangrass; and FA, fallow. Ct/acre is 25-pound cartons/acre.

6. The extra large fruit yields from the first harvest were significantly improved by sunn hemp and velvetbean compared with the fallow (Fig. 6).

Fig. 6 Tomato extra large fruit yields from the 1st harvest with cover crops
Cover crops: SH, sunn hemp; VB, velvetbean; CP, cowpea cv ‘Iron & Clay’; SS, sorghum sudangrass; and FA, fallow. Ct/acre is 25-pound cartons/acre.

7. Application of organic mulch for two years or just using the plastic mulch significantly improved not only the total extra large fruit yield (Fig. 7)

Fig. 7 Tomato total extra large fruit yields with different mulches
PM is plastic mulch. Rates of application of compost: OM1, 25 Mt ha-1; OM2, 50 MT ha-1; OM22, 50 Mt ha-1 applied in 2 consecutive years, and OM3, 100 Mt ha-1. Ct/acre is 25-pound cartons/acre.

8. Application of organic mulch for two years or just using the plastic mulch significantly improved the extra large fruit yield from the first harvest (Fig. 8).

Fig. 8 Tomato extra large fruit yields from the 1st harvest with different mulches

In general these data suggest that cover crops, sunn hemp and velvetbean, especially sunn hemp and organic mulch can improve tomato yields and the quality in the subtropical area of Florida. However densities of plant parasitic nematodes were very sparse, but weed pressure was quite high in the experimental field.

Experiment in 2004-2005
Objective (2004-2005):
Materials and Methods: The experimental design and the materials and methods used in the 2004-2005 were identical to those used in 2003-2004 with the exception that after incorporation of the cover crops into the beds mixture of S-metolachlor (Dual Magnum, which is effective against nutsedges) and napropamide (Devrinol, which is effective against broad leaf weeds) were applied at the label rates to the areas designated to receive the organic mulch. The herbicides were sprayed onto the surface of the beds and then watered in with 0.64 cm of water. On two subsequent dates a tank mix at label rates of halosulfuronmethyl (effective against broadleaf weeds and nutsedges) and sethoxydim (Poast, effective against grasses) was applied.

Results (2004-2005: The yields obtained in this second year were much higher than in 2003-2004. However the pattern of results were different.

For tomato marketable yield, no significant differences between the cover crop treatments were observed either at the first harvest or with respect to total marketable yields (Fig. 9).

Fig. 9. Tomato total marketable and the first harvest yields from the various cover crop treatments.

With respect to yields with the various mulches (Fig. 10), the most important result was that yields with compost used as an organic mulch were as high as those obtained with plastic mulch.

The lowest total marketable yield was obtained with 25 Mt ha-1 of compost. However very good yields were obtained at the higher rates, although yield did not correlate well with the amount and number of applications in consecutive years.

Fig. 10. Tomato total marketable and first harvest yields. OM-1: 25 Mt/ha, OM-2: 50 Mt/ha, OM-3: 75 Mt/ha, OM-32: 50 Mt/ha for 3 years

c. Precision fertigation of tomatoes using an automatic drip irrigation system with soil moisture feedback.

Objectives.

1. Assess the effectiveness of the soil moisture based automatic, high-frequency/low-volume (HFLV) drip irrigation system developed at the Tropical Research and Education Center (TREC), University of Florida/IFAS in producing tomatoes in Krome gravely loam soil.
2. Assess the practicality of combining the system with an on-line passive fertigation system.
3. Estimate the potential reduction in nutrient leaching from the root zone into the shallow Biscayne aquifer of the area.

Materials and methods.

A field in which sorghum sudangrass was grown as a summer cover crop was utilized for this experiment at the University of Florida’s Tropical Research and Education Center (TREC). Beds were formed 1.83 m apart and pre-plant dry fertilizer (6-6-12) at 1000 kg/ha was rototilled into each bed. On 9/30/04, fumigant (66:33 (vol: vol) mix of methyl bromide: chloropicrin (MC-33)) was injected into the soil at 392 kg/ha during the final formation of the raised beds, and during this same operation two sets of two drip lines (4 drip lines per bed, arranged so that 2 lay adjacent to each other 15 cm to the right of the center of the bed and 2 lines adjacent to each other to the left of the center of the bed) and plastic mulch were installed. One set of drip lines was used to deliver water and the second set was used to deliver liquid fertilizer. The irrigation lines used were T-TAPE TSX 508-12-450, flow rate: 5.6 l/min/100 m at 5.6 m pressure, drip spacing=0.3 m, diameter: 16 mm). Tomato seedlings of the cultivar, ‘FL 47’, were transplanted on 10/15/04 into the plastic mulched raised beds. The transplants were installed in one row per bed with plants spaced 0.46 m apart. Tomatoes were cultured and protected according to local agronomic practices in Homestead (FL)

Figure 1. Field plot layout for tomato experiment in Block 5 at TREC during 2004-2005.

The field was divided into two areas (Fig.1), i.e., (1) the Experiment Plot from which water, nutrient and leaching data was obtained, and (2) the Demonstration Plot for clientele to view the system working as it would in commercial practice. Four irrigation treatments were selected (Table 1). Each treatment was replicated 3 times. Each bed 50 m long was operated/measured independently.

Table 1. Irrigation treatments for tomato experiment.
Treatment Scheduling method Scheduling Device
C1 (switching tensiometer) -25 cbar Tensiometer
C2 (QIC+ECH20) 400 mV QIC/ECH2O
C3 (Kc-High (100% needs)) Kc-High (100% needs) Historical weather data
C4 (Typical grower’s schedule) Typical grower’s schedule Custom

Treatments C1 and C2 utilized high-frequency-low volume irrigations as determined by soil moisture requirements. In the C1 treatment a switching tensiometer in each bed was set at -25 cbar. In the C2 treatment utilized the quantified irrigation controller (QIC) developed at the University of Florida (Muñoz-Carpena et al., 2004) combined with a capacitance probe (ECH2O, Decagon Inc.). The irrigation set point for the QIC controller was set to 400 mV, which is equivalent to -25 cbar for Krome very gravely loam soil. Both treatments were controlled by an irrigation timer (model no. EXP-12 LX+, Rain Bird Corp.) set to irrigate up to 4 times/day, 12 minutes per irrigation (maximum of 48 min/day).
The soil moisture devices were interfaced to cut-off irrigation whenever the moisture level exceeded the desired set-point. Treatments C3 and C4 were controlled by the timer, C3 was adjusted throughout the season to irrigate according to historical long-term ET and tomato crop coefficient (Kc) values published for Miami-Dade area (Simonne et al., 2004). The maximum needs for the season corresponded to 48 min/day of irrigation for our system. Treatment C4 was selected to represent the common commercial producer practices in the local area, with a fixed schedule throughout the season equivalent to 1h/day of irrigation.
Water use in each treatment was continuously recorded by a positive displacement water meter equipped with a magnetically actuated reed switch (PSM-T 1.6 X 1.3 cm (5/8 X 1/2 inches), ABB Water Meters, Inc., Ocala, FL) connected to an event data logger (H7-002-04, Onset Computer Corporation, Bourne, MA). Weekly readings were also manually taken from the counters in each water meter.
The irrigation schedule is summarized in Table 2. Notice that numbers given for C1 and C2 correspond to maximum irrigations if the soil was dry and signals from the soil moisture probes would not cut irrigation off. Actual frequency of irrigation is lower.

Table 2. Irrigation schedules for the different treatments (numbers in parenthesis represent time per irrigation in minutes (4 irrigations per day)
DAILY IRRIGATION TIMES (min/day)
Treatment Week
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
C1 Establishment 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12)
C2 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12) 48(12)
C3 17(5) 21(6) 21(6) 42(11) 42(11) 42(11) 38(10) 38(10) 38(10) 38(10) 38(10) 31(8) 31(8) 21 (6)
C4 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Fert (Expt.)1 10 14 14 19 24 24 24 24 24 24 24 19 14 10
Demo (C2)2 12 12 12 12 12 12 12 12 12 12 12 12 12 12
1 Experimental plots were fertigated from the 50% dilution tank. 2 Demo plot was fertigated from 15% dilution tank until 01/07/05 and from 11% there after.

The fertilizer injection times and concentrations were adjusted throughout the season to match the needs of the tomato plants as presented in Table 2. Dissolved fertilizer (4-0-8) was applied automatically through a Venturi injector (model no. 484, Mazzei Injector Corp.) into the drip lines with manufacturer’s injection rate of 64 L/hr when positioned across 10 m pressure regulators with upstream operating pressures of 25 to 35 m. A calibration test was carried out to determine the actual injection rate of the Venturi injectors once they were installed in the system. The injection rates of the Venturi injectors were found to be 52.2 L/min for each of the two combined injectors of the experiment plot, and 58.2 L/min for the single injector used for the demonstration plot. The liquid fertilizer was injected twice per week off the regular irrigation schedule and equally across the field using the second pair of parallel drip lines in each bed to avoid biasing the water application results by the amounts of fertilizer applied (Fig. 2 and 3).

Figure 2. Irrigation system depicting lay flat hose and drip tape layout. Note: Each bed was provided with 4 drip lines. On set of two was used to deliver water, and the second set of two was used to deliver liquid fertilizer.

Figure 3. Layout of control system infrastructure at pump house on Block 5, TREC. S.V. denotes the solenoid valve.

The electrical wiring to control the irrigation and fertigation treatments is depicted in Figure 4. Each treatment in Table 1 was controlled by a different channel in the irrigation programmer box to operate the irrigation valves. Each valve or set of valves that needed to be controlled independently was assigned a separate station. Six stations were used for the various valve control requirements. The Rainbird controller was connected to a relay box to boost the irrigation controller’s capabilities, and to allow multiple valves in the field to be controlled by one station, without an excessive voltage drop.

Figure 4. Wiring scheme of control system to Rainbird ESP-LX+ Irrigation controller in Block 5, TREC. The Upper four rectangles represent the experimental sets of beds, and the lower elongated rectangle represents the demonstration set of beds.
Zero tension lysimeters were buried under the beds at random positions along the beds (see Fig. 1). The capture area of the lysimeter was 1662 cm2 (circle of 46 cm diameter) (Fig. 5). The surface of the lysimeter was a flat plastic catchment pan with a drain at the center. Each lysimeter was buried flush with the surface of the bed-rock, and covered with the soil approximately 30 cm deep. This allowed field cultivation to continue without damage to the lysimeters or cultivation equipment. The lysimeters were positioned beneath the beds at positions shown in Figure 2. The lysimeters were manually pumped to remove all leached water before the experiment started, and to take samples at desired intervals during the experiment. In order to assure that the lysimeters were placed directly beneath the beds, the coordinates of the excavations made in the bedrock for their installation determined in relation to stakes outside the experimental area, and these measurements were taken and recorded on 06/12/04. Subsequently the topsoil was place over the lysimeters and the cover crop was grown, terminated and incorporated into the soil. By means of the coordinates the raised beds were formed directly over the lysimeters. After the tomatoes had been transplanted, the positions of the plants and the drip emitters relative to each lysimeter were examined, and appropriate adjustments were made in a few instances. It was imperative to position the plants and drip emitters directly over the lysimeters to obtain accurate leaching data.

Fig. 5. Field lysimeter. The 19-liter capacity bucket was installed in an excavation in the limestone bedrock. A 46 cm-diameter flat plastic catchment pan with a drain at the center was positioned above the bucket to collect leachate from the root zone. Hoses inserted into allowed the Leachate to be pumped out and replaced by air.

Tomatoes were harvested four times during the period March 1 -18, 2004. Harvested fruits were graded following Florida Tomato Committee standards. Irrigation treatments were established according to Table 1 in a randomized complete block design with four replications. Irrigation scheduling methods consisted of switching-tensiometer based (C1-C2), QIC controller based (C3-C4), historical evapotranspiration (ET) weather based (C5-C8), and practices used by local growers (C10).

2. Experimental Results

2.1. Water application

The total amounts irrigation water applied in each treatment up to the early season point (second harvest, 90 days after transplanting (DAT)) and at the end of the season (120 DAT), were subjected to a repeated measurement ANOVA analysis. The means compared using Tukey-Kramer post-hoc test at the 0.05 significance level.
Table 3 summarizes the results of the water application by all treatments. Although less than one third as much water was applied in Treatment C2 (QIC + ECH20) as in Treatment C1 (switching tensiometer), these differences between these soil moisture feedback treatments were not significant statistically (p<0.05). On the other hand the amounts of water applied in accordance with evapotranspiration (C3) or a typical grower’s schedule (C4) were significantly greater than those applied in the soil moisture feedback treatments (C1 and C2). Values in these two broad categories ranged from 3.5 to 28 fold. Table 3. Amount of water applied to tomato beds via drip lines in each treatment (mm)
Treatment Early season (2nd harvest) End of season (4th harvest)
C1 (switching tensiometer) 81 a 129 a
C2 (QIC+ECH20) 17 a 35 a
C3 (Kc-High (100% needs)) 283 b 352 b
C4 (Typical grower’s schedule) 479 c 611 c

Figure 6 shows how a large portion of the water savings achieved by the soil moisture feedback treatments is obtained early in the season. The soil moisture feedback treatments (C1 and C2) resulted in minimal water use up to the end of December (DAT 65), i.e. flat slope, while the fixed calendar treatments apply water greatly in excess of the needs of the tomato plants during this time, even for the ET treatment corrected for crop coefficient during the season (C3). The real-time water application achieved by the soil moisture probe feedback proved to be very efficient in reducing water applications by meeting the plants’ needs, which are determined by local conditions.

Fig. 6. Water applied in each treatment.

2.2. Yields

The results were analyzed using linear ANOVA (SAS GLM Procedure) with Duncan Multiple Mean Separation at significance level of 0.05. The dependent variable was Total Marketable Yield (TM), Extra Large (XL), Large (L), Medium (M), Small (S, only in 4th harvest), Cull (C), while the treatment variable was irrigation at 4 levels (C1, C2, C3, C4).
The test of significance and separation of means procedure (Tables 4a and 4b) showed that treatment effect were not significant (p=0.1971) for TM (sum of XL, L and M) and the model explained 37% of data variation (R2 = 0.3744). Treatment effects were significant for XL (p=0.0008), L (p=0.0144), and M (p=0.0707). The model can explained 55.8% to 75.8% variation among the means. Lastly, the treatment effect was not significant for cull yield.

Table 4a. Tomato early yields in 2004-2005 trial (unit: carton/acre)
Treatment TM XL L M S Cull
C1 1616.8 a 939.0 a b 637.8 a 40.1 b — 58.5 a
C2 1300.1 a 494.0 c 698.3 a 107.8 a — 48.5 a
C3 1294.5 a 823.4 b 394.3 b 76.9 a b — 67.9 a
C4 1630.9 a 1179.5 a 413.7 b 37.7 b — 66.0 a

Table 4b. Tomato early yields in 2004-2005 trial (unit: kg/ha)
Treatment TM XL L M S Cull
C1 54457 a 31627 a b 21482 a 1351 b — 1970 a
C2 43790 a 16639 c 23520 a 3631 a — 1634 a
C3 43601 a 27734 b 13281 b 2590 a b — 2287 a
C4 54932 a 39728 a 13934 b 1270 b — 2223 a

The distribution of grades of tomatoes produced in C4 was somewhat unique, with superior good yield of XL but low yields in L, M and S. Irrigation management did not appear to be a key factor in the distribution of yields between the different tomato grades. As early as the first harvest, tomato yellow leaf curl virus infection was found in some of the tomato plants. This whitefly transmitted virus can drastically affect yields depending on the stage of the plant at the time of infection, so that plants infected before flowering bear no fruit.

2.3. Nutrient leaching

Leachate from the tomato root zone was collected weekly from 4 lysimeters in each irrigation treatment for 14 consecutive weeks (DAT 6, 14, 21, 27, 35, 40, 49, 56, 63, 70, 84, 124, and 139) (Table 5). Volume weighted nutrient concentrations from DAT 14 and 56 were subjected to a repeated measurement ANOVA analysis, and least squares means were compared using Tukey-Kramer post-hoc test. Nutrient loadings were computed as the sum of products of nutrient concentration and leachate volume for all sample dates except the first one (i.e. DAT 6). The first samples were omitted because they contained leachate collected in the lysimeters prior to the initiation of this experiment. The loading means were compared using Duncan multiple means separation method at the significant level of 0.05.

The irrigation strategy strongly influenced the amounts of leachate obtained from the root zones (p = 0.0026) and the concentrations of ortho-P (p = 0.0076) and TP (p = 0.0031). Irrigation determined by evaporation rate (C3) and typical grower’s schedule (C4) resulted in the most leachate (Table 5). The ECH2O controlled irrigation (C2) generated the least dissolved and total P levels in leachate, while the typical grower’s schedule (C4) generated the most. However the type of irrigation practice seemed to have no obvious impacts on leachate inorganic N levels (Table 1).
Overall, the type of irrigation significantly affected NO3-N (p=0.0381), ortho-P (p=0.0049), and TP (p=0.0017) loadings, and the model explained 49.1% to 67.2% of data variation.

Table 5. Least square means of total leachate volume and volume weighted nutrient concentrations in 2004-2005 tomato trial (unit: mg/L)

Treatment Volume NH4-N NO3-N Ortho-P Total P
C1 27144 a 0.03 a 11.0 a 0.60 ab 0.73 a
C2 18430 a 0.13 a 9.9 a 0.43 b 0.49 b
C3 86570 b 0.10 a 15.6 a 0.51 ab 0.61 ab
C4 77163 b 0.06 a 9.9 a 0.66 a 0.77 a

* Means with the same letter are not significantly different in same column.

Differences in nutrient loading rates among irrigation treatments were most likely caused by the differences in leachate volumes resulting from the various irrigation inputs (Table 6). This was evidenced by the high loadings of ortho-P, TP and NO3-N in the leachates. No significance in NH4-N loadings among treatments might likely be a result of high spatial variability within individual treatments (Table 6).

Table 6. Nutrient loading in leachate in 2004-2005 trial (unit: mg/lysimeter)

Treatment NH4-N NO3-N Ortho-P Total P
C1 2.7 a 262.3 a 15.4 a 19.1 a
C2 2.0 a 181.3 a 8.1 a 9.4 a
C3 145.61 a 1670.5 b 39.5 b 55.0 b
C4 7.3 a 527.0 a 44.3 b 57.4 b

* Means with the same letter are not significantly different in same column.

The results demonstrate that major reductions in nutrient leaching is possible through soil irrigation system control based on soil moisture.

Impacts and Contributions/Outcomes

Virginia Component

Based on data from 2004 and previous research, we are confident that organic potato growers can use high-residue no-till systems to produce profitable potato yields and better maintain the production capacity (sustainability) of their soil than with conventional inversion tillage/cultivation systems.

Presentations were given by Dr. Ron Morse on various aspects of organic no-till production of vegetables – including potato – at three state and regional meetings: Maryland Organic Food and Farming Association (MOFFA) in Annapolis, MD (January 8, 2005); Southern Sustainable Agricultural Workers Group (SSAWG) in New Orleans, LA (January 20-23, 2005); Richmond Area Vegetable and Small Fruit program in Mechanicsville, VA (February 1, 2005).

Carrera, L. M., R. D. Morse, A. A. Abdul-Baki, K. G. Haynes, and J. R. Teasdale. 2005. A conservation-tillage cover cropping system and economic analysis for creamer potato production. Amer. Jour. of Potato Research. (In press).
Morse, R. D. and N. G. Creamer. 2005. Developing no-tillage without chemicals: best of both worlds? In A. Taji and P. Kristiansen (eds.). Organic Agriculture: A Global Perspective. CSIRO Publishing, Collingwood, Victoria, Australia . (In press).

Florida Component

Muñoz-Carpena, R., M.D. Dukes, L.W. Miller, Y.C. Li and W. Klassen. 2004. Design and Field evaluation of a new interface for soil moisture-based irrigation control. ASAE Paper No. 042244. St. Joseph, Mich.: ASAE.

Based on the results of the soil-moisture controlled irrigation systems developed in this project, similar systems are being developed for papaya production by Drs. Kati White, Jonathan Crane and Bruce Schaffer at this Center.

Thirty on people attended a workshop we presented in January 2005 on this technology.

Collaborators: