Editor's note: A special thanks to Dr. Monica Ozores-Hampton at the Southwest Florida Research and Education Center for coordinating The Immokalee Report, of which the article below is part.

Applying BMP (best management practice)-recommended water and phosphorus (P) inputs for seepage-irrigated tomato and watermelon in South Florida can reduce water use and P leaching to groundwater without adversely affecting fruit yield.

These are the findings of a 3-year comprehensive UF/IFAS study at the Southwest Florida Research and Education Center in Immokalee. The study determined BMP effectiveness with regard to yield, water use and water quality by evaluating response to different irrigation and fertilizer (N-P-K) inputs.

One tested production system represented industry-average fertilizer and water use whereas a second system used BMP-based inputs, both with seepage irrigation. A third system used BMP-based fertilizer rates with subsurface drip (BMP-drip), which is an alternative to the traditional seepage system.

The industry fertilizer rates and irrigation management were determined from a grower survey and on-farm measurements, respectively. The BMP-based P rates were based on a Mehlich-1 (M1) soil P test, whereas industry rates represented the average of survey results.

For tomatoes, they were 162 pounds of P₂O5 per acre, and for watermelon, they were 170 pounds P₂O5 per acre.

The industry system was maintained at a higher soil moisture than the BMP systems, which were managed by measuring soil moisture in the bed. The water table depths for the industry system were shallower compared with the BMP system. Concentrations of P in the plant, soil and groundwater were measured.

Tomato yields measured across three P rates and four growing seasons were not different, indicating that no adverse impact of reduced fertilizer rates occurred. Yield effects for watermelon varied across the two seasons tested.

During one season, there was no difference in yield; during the other season, yields for the industry average fields were higher, mainly due to a wetter-than-average spring that resulted in an extremely high water table that leached potassium and nitrogen from the beds.

However, no effect of P fertilizer rates on yield could be established. Although P concentrations in the leaf tissue for the BMP system were lower than the industry average, they were within the sufficiency range.

Despite a 35 percent higher P rate applied in the industry system (162 pounds P₂O5/acre) compared with the BMP (120 pounds P₂O5/acre) systems, the M1-P values remained similar as long as both systems received P inputs.

Soil M1-P stabilized around 55 parts per million and showed a gradual decrease after P application stopped for the two BMP systems, but M1-P steadily increased in the industry system, reaching a high of 145 ppm by the end of the study.

M1-P was positively correlated with groundwater P. Results showed that total P (TP) concentration in the groundwater for both BMP systems (BMP = 2,098 parts per billion and BMP-drip = 2,048 ppb) were 33 percent lower than the industry average of 3,090 ppb. Although the sub-drip system used almost 50 percent less water compared with the industry system, it did not offer any groundwater quality benefit.

Our results showed that adoption of UF/IFAS-recommended P rates as a BMP did not appear to reduce crop yield but improved water quality. Reductions in groundwater P observed with the BMP rates will also reduce the P loads in farm drainage.

The study was funded by the Florida Department of Agriculture and Consumer Services, the Southwest Florida Water Management District and the Southwest Florida Vegetable Growers Research and Education Fund.

Sanjay Shukla is an associate professor of agricultural and biological engineering at the Southwest Florida Research and Education Center in Immokalee. Gregory Hendricks is a post-doctoral researcher at the same center. Tom Obreza is a professor in the Soil and Water Science Department, Gainesville.