This study was conducted to evaluate a method being used in California to monitor N loss from field scale fertilizer applications for estimates of canopy N loss during defoliation. The methodology employs citric acid traps to measure NH₃ concentrations at various heights above the canopy for field estimates. A micro-site was established using the same active denuder samplers at heights within and above the canopy. These sites were monitored prior to and through the post treatment period following a Def and Prep application. Standard mass balance micrometeorological techniques were used to estimate the integrated NH₃ flux by combining measurements of wind speed and NH₃ air sample concentrations from the height- dependent sampling locations. The experiment was timed to span a short (2-week) period pre- to post-defoliation application based on the hypothesis that the enhanced senescence induced by a chemical defoliant would increase plant emissions of NH₃. Information obtained on NH₃ emissions will be used to help develop a complete N budget for California cotton.

Prior measurements of NH₃ emissions from cotton fields fertilized with anhydrous NH₃ injected on a silty clay with pH of 8.5 and a similar site with pH of 7.8 produced emission factors of 5.6 and 3.9 % of applied fertilizer, respectively (Potter et al). The defoliation site was on a Lethent silty clay with pH of 7.9. The methodology and sampling procedure produced data that showed differences in diurnal NH₃ emission values and distinct differences related to the canopy defoliation. At this site, the canopy sampling showed a 76% reduction of NH₃ emissions from the pre-treatment canopy to 5 days post treatment. By this time the field had reached 65 % defoliation (visual rating) and was well desiccated (75%). NH₃ volatilization peaked at 2 days post-defoliation. This value was 11% higher that the pre-treatment value. There was at this time, a distinct shift in the diurnal emissions. Prior to defoliation the day/night emission were nearly equal (51/49% respectively). After defoliation, the nighttime emissions were reduced to only 6% of the total diurnal value. This trend in reduced night emissions was extended to the 5^th day of the sampling period. At 65 % defoliation (5 days post treatment) the total diurnal emissions were reduced 70% from the pretreatment values.

This field study showed that canopy level NH₃ emission can be monitored to provide estimates of NH₃ volatilization loss from a cotton field. The sampling procedure was sensitive enough to show differences in diurnal NH₃ flux values and changes associated with plant defoliation and desiccation. Additional monitoring could provide estimates of seasonal N losses and specific losses from various fertilizer application methods and soil types. This information could help explain discrepancies in unaccounted N for a complete cotton N budget.