During the summer of 2001, romaine lettuce was grown on these fields by an organic vegetable grower and was harvested in early October 2001. Upon completion of the transition period and after discussions on experimental details with the grower, a replicated five-year rotation trial with 5 treatments was established in a 0.4 ha section of the ranch. The soil type of the fields is Santa Ynez fine sandy loam with 2–9% slopes. Water infiltration is very slow due to a claypan at 45–150 cm deep. Organic matter content in the topsoil is as low as 10 g kg−1 and the pH is 6.7–7.0 . The baseline V. dahliae population in the topsoil was below the detection limit. With a typical Mediterranean climate, the majority of rainfall is concentrated in winter . The average annual precipitation at the Elkhorn Slough National Estuarine Research Reserve was 498 mm and mean daily temperature ranged from 3.6 to 21.7◦C during the experiment.Fresh strawberry fruit yield was measured for each cultivar once or twice per week from 40 designated plants during the harvest period. Mortality of strawberry plants was examined from May to the end of the season by counting living and dead plants in the middle two beds of each plot. During year 5, 20 transplants were sampled on November 23, 2005, and 4 whole plants of strawberry were dug out from the middle two beds of each plot on January 8, March 13, June 2, and October 5 in 2006. After washing out soil from roots with running water, samples were dried at 60 ◦C in a convection oven to constant weight. The dried samples were weighed for dry biomass, plant bench indoor ground to pass through a 0.5-mm sieve with a Wiley mill, and analyzed for total C and total N content using a Vario Max CNS analyzer .

Twenty mature fruits from each plot were sampled on June 20, 2006, frozen at −20◦ C, and lyophilized by a freeze dryer. Freeze dried samples were passed through a 0.5-mm sieve and analyzed for total C and N content using the fore-mentioned method. Fresh moisture content of fruit was calculated from the difference between preand post-lyophilized weight of fruits. Cumulative total fruit yield, total N, and fresh moisture content of mature fruits were then used to estimate N uptake by fruit on June 2 and October 5, 2006, which was combined with whole plant biomass-N to calculate plant biomass N plus harvested fruit N on these dates. Since the focus was on strawberry disease management, broccoli wasmanaged as a cover crop for disease control in this experiment and broccoli floret yield was not recorded. About 1 kg of fresh broccoli and cover crop samples from each plot was brought to the lab and biomass and total C and N content were determined by the above-mentioned methods. Approximately 1 kg of composts and organic fertilizers used in the trial were sampled, oven dried, ground, and analyzed for total C and total N in the same manner. Also ∼100 g subsamples were taken to gravimetrically determine moisture content by heating at 105◦ C for 48 hours in a convection oven.Ten to twenty soil cores at 0–15 cm were sampled by soil probes at the middle two beds of each plot to make a composite sample. Samplings were conducted 2–3 times per year for the entire trial period. About 100 g of composite sample were air dried in the lab for six weeks and then ground and plated onto V. dahliae semi-selective medium using the Anderson Sampler dry sieve technique to test the number of viable microsclerotia of V. dahliae. NP-10 plates were incubated in the dark for three weeks, then examined for typical microsclerotia formation by using a dissecting microscope. During year 5 , topsoil and subsoil were sampled monthly in the same manner. After mixing well in a bucket, about 5 g of soil were taken from the composite sample and transferred on site into a pre-weighed screw top plastic tube containing 25 ml of 2M KCl.

Each tube was tightly sealed and kept on ice in an ice chest and transferred to the lab. In the lab, samples were reweighed, shaken for one hour with a reciprocal shaker, filtered to obtain clear supernatant, and kept at −20 ◦C until analysis. NH4-N and NO3-N concentrations in the KCl extracts were determined by flow injection analysis and the sum of NO3-N and NH4-N was expressed as inorganic N in the 0-30 cm soil layer. Separately, a ∼100 g subsample was taken from a composite soil sample to determine gravimetric soil moisture content. The remainder of soils sampled at the end of the strawberry harvest season in October 2006 were air dried, passed through a 2-mm sieve, and analyzed for pH and electrical conductivity . Ten gram subsamples were further ground to pass through a 0.5 mm sieve and analyzed for total C and N content using the above-mentioned methods. Soil bulk density was also determined at the end of the harvest season in October 2006 using the soil core method for 0–15 cm and 15–30 cm depths of the strawberry beds. Four cores were sampled per depth and bulked to form composite samples.During years 1 and 4, plants were regularly observed for disease symptoms by the grower and researchers and tested as needed. At the end of year 5, plants showing dieback, stunting, or collapsing symptoms were collected from all plots on October 5, 2006. For some of the Seascape plots, plants in general showed few signs of distress. Plants were taken to the lab and isolations made on roots and crowns. Plant material was first washed free of dirt and debris. Crowns and roots were then surface sanitized by soaking in a 0.1% bleach solution for 3 min and then thoroughly rinsing in sterile distilled water. Using aseptic technique, crowns and roots were dissected and symptomatic pieces of tissue were placed into separate Petri dishes containing acidified corn meal agar , PARP semiselective medium , or NP-10 . Isolations were evaluated on October 16, 2006.Prior to analysis of variance , data were log transformed to ful- fill ANOVA’s assumptions when needed. Split plot two-way ANOVA and cultivars as variables was applied to test statistical differences among treatments. Repeated measure analysis was conducted to examine treatment effects on V. dahliae population in soils. Tukey’s HSD post-hoc test at P = 0.05 was used for mean separation.

Regression analysis was conducted for examining relationships between marketable fruit yield of strawberries in year 5 and the length of break period between strawberries. Statistical analysis package Statistix was used for these analysis.Cultivars Aromas and Seascape were planted in all plots in year 5. Strawberries in all plots grew well without any major pest problems and disease symptoms. The average plant mortality was 1.2%, a level similar to the first four years, and no significant difference in mortality was found between any treatments. In the early harvest stage, rotation treatment did not affect fruit yield whereas, between cultivars, cv. Seascape had a higher yield compared to cv. Aromas . This is a common pattern in years 2 through 4 . In the mid-harvest season, however, yield was highest in treatment E followed by D , C , B , and A . A significant difference was found between treatments E and A. In the late season, no significant differences were found between any treatments. Overall cumulative marketable yield in year 5 ranged from 650 to 890 grams per plant, a comparable range with the first four years. Seascape had a significantly higher yield than Aromas , which was the opposite trend for years 2 through 4; yields for Seascape exceeded those for Aromas in mid- to late season in the previous three years but not for year 5. Among rotation treatments, though a significant difference was found only between treatments A and E, the longer the years between strawberries, the greater the marketable fruit yield. Since the interaction was not significant , greenhouse rolling racks cumulative marketable fruit yield of both cultivars were pooled and relative yields to the average of the seven- year rotation yield for both cultivars calculated. Then the correlation between the relative yields and the length of break periods were analyzed. The result showed a strong positive linear correlation between the two factors . Numerically, marketable yield in treatment A was approximately 20% lower than treatment E. Total fruit yield showed a similar trend with marketable yield ranging from 891 g per plant to 1093 g per plant .The loss of soil productivity when crops are grown repeatedly on the same land resulting in poor plant growth and reduced yields is called “yield decline” , “soil sickness” , or “replant problem” . Such losses, called here “yield decline,” have been reported in many crops worldwide including strawberries . Biotic and abiotic factors can cause yield decline. Further, although one factor may possibly be responsible for yield decline, it is more likely that a combination of factors interact to cause the effect . This study also demonstrated the challenges researchers face when using a participatory process where farmer involvement is a key part of the experiment. Farmers cannot wait until the end of a long-term experiment such as this before making adjustments in their farming practices. Changes occur as needed, and the experimental design needed to shift as well. This definitely added to our difficulty in pointing to single factors causing yield decline.

The present study showed that the integrated practices of broccoli residue incorporation, compost application, mustard cover crop incorporation, use of a relatively resistant cultivar, and rotation with non-host crops allowed the one- to three-year rotations to have a statistically similar marketable fruit yield to the seven-year rotation in organic strawberries grown in this region of coastal California. Although the trial was conducted in a field with a low baseline V. dahliae population and results may be different for fields with greater disease pressure, it should be noted that the V. dahliae population in the soil was kept low during the five-year study and no V. dahliae or other major pathogens were detected from strawberry plants at the end of year 5. The low V. dahliae population in the soil during the trial may be due to broccoli residue incorporations though the application rate of broccoli residues in this trial was lower than many farmers are using today. Nevertheless, a signifi- cant positive correlation of the break period between strawberries and the marketable fruit yield in year 5 existed , and marketable fruit yield in continuous strawberry plots was 20% lower than seven-year break plots . Given these results, the cause of the yield reduction at the shorter rotations appears to be something other than major soil-borne pathogens . It is known that other minor sublethal pathogens such as species of Colletotrichum, Pythium, Rhizoctonia, and Cylindrocarpon can reduce fruit yield of strawberries in California , which may have played a role in yield reduction of the shorter rotation treatments inthe trial. Further, Seigies and Pritts showed that three sowings of brown mustard cover crops enhanced successive strawberry growth compared to continuous strawberries regardless of relatively high levels of fungal infection on the strawberry roots. Use of mustard cover crops in the longer rotation treatments of the present study may have had a similar positive effect. Soil nutrient imbalance after continuous cropping can also cause yield decline. In the present study, fertility management changed from year to year in order to reduce N loss during the rainy winter and by the grower’s preference. To assure adequate nutrient status, we conducted tissue tests for strawberries and monitored soil inorganic N dynamics during the strawberries’ growth period. Although strawberry leaf blade analysis in the mid- and late season of year 5 showed differences in Mg and NO3-N among treatments, none of these appeared to be critically defi- cient. Further, even though plant tissues in shorter rotations showed lower petiole NO3-N, it is difficult to interpret whether it was due to the lack of N supply, reduced N absorption by roots infested by non-lethal pathogens, or both . In fact, no significant difference was found between any treatments in soil inorganic N content during year 5 , suggesting the lower total-N content in leaf blades may be attributed to the poor root health in shorter rotations rather than the lack of N supply.