Sustainable Management of Soil Nitrogen and Organic Matter in Low-External-Input Organic Crop Production in the Caribbean.

  Cover crops (CC) as sources of organic matter and nitrogen are of particular interest to low-external-input organic crop production systems where external inputs are not economically feasible and may not be locally available at all.  A two-year experiment was established on a certified organic vegetable farm in St. Croix, U.S. Virgin Islands to evaluate the effect of four cropping systems on soil nitrate nitrogen (NO₃N), total soil organic matter (OM) content, and CC biomass. Systems tested were: 1) high crop complexity (HIGH) (legume and grass CC bicultures + vegetable crops (VC) with an intercrop); 2) intermediate crop diversity with legume CC (MED-L) + monoculture VC); and 3) intermediate crop diversity with grass CC (MED-G) + monoculture VC); and a control of 4) low crop diversity (LOW) (no CC + monoculture VC). Complexity levels were tested over a 2 year period consisting of 6 sequential crop cycles; 3 per year (Fall to Spring, Spring to Summer, and Summer to Fall).  The legume CC were sunn hemp (Crotalaria juncea L.) [SH] and velvet bean (Mucuna deeringiana, (Bort) Merr) [VB]; and the grass CC were sorghum-sudangrass (Sorghum bicolor x S. sudanense L.) [SS] and pearl millet (Pennisetum glaucum L.) [PM].  Plots with no CC were managed as weedy fallows [WF].  Statistical analysis was performed by system complexity and crop cycle using PROC MIXED (SAS). In 2006, the SH+PM biculture biomass yield was 10,355 kg/ha and was similar to SS, PM, and SS+VB, which yielded 6,967, 6,477, and 7,306 kg/ha of biomass; respectively, and greater than that of the WF, VB, and SH.  In 2007 SS produced 11,433 kg/ha of total biomass, which was greater than the yields of the WF, PM, and SH.  HIGH and MED-G systems had biomass yields of 9,270 and 8,806 kg/ha in 2006 and 9,709, and 10,890 kg/ha in 2007. These yields were significantly greater than for the LOW and MED-L systems that produced 4,900 and 6,150 kg/ha in 2006 and 5,378 and 7,144 kg/ha in 2007; respectively.  There was no effect of system complexity on soil OM or NO₃-N levels. However, there were crop cycle effects for OM and NO₃-N levels.  Both OM and NO₃-N levels exhibited similar trends over the duration of the experiment.  Organic matter increased from cycle 1 to 2 from 4.7% to 5.6%, then decreased to a low of 2.5% in cycle 6.  Similarly, NO₃-N levels decreased over time from 86 ppm in cycle 1 to 28 ppm following cycle 6.  Following cycle 6 at the end of year 2, NO₃-N levels between SH and the WF differed with 32 and 25 ppm; respectively.