Effects of fertilizer nitrogen (N) application rate and Enhanced Efficiency Fertilizers on sugarcane productivity, efficiency of N use and loss of N in runoff (NESP TWQ 2.1.8, UQ)

The dataset consists of tables of means, with statistical differences indicated where they are significant, for measured crop performance, fertilizer N recovery and use efficiency at 6 field sites from Mackay to Freshwater. Runoff losses of N are also shown from sites at Freshwater and Silkwood. **This dataset is currently under embargo until the project end date. Optimizing fertilizer nitrogen (N) application rates to both sustain high levels of productivity and minimize any impacts on the surrounding ecosystem is challenging, especially under monsoonal wet season conditions in northern Australia. The inability of existing application strategies and fertilizer N products to achieve synchrony of mineral N supply with crop demand, or prevent rapid formation of nitrate-N (that is vulnerable to loss via gaseous or aqueous loss pathways) increases risks of inefficient N use. A blend of enhanced efficiency fertilizers (EEFs) with different modes of action has the best chance of lowering the risk of N losses and increasing crop N recovery, providing an opportunity to reduce fertilizer N rates without increasing the risks of productivity loss. Six field trials were established from Mackay to the wet tropics, with data collected from consecutive ratoon crops at each site. Yields and indices of N use efficiency were developed for crops receiving urea-N at rates equivalent to that derived from the local SIX EASY STEPS guidelines, or as urea or a blend of EEF’s applied at N rates calculated using a block-specific yield target (PZYP) based on mill records. Methods: The project established a total of six field sites after the 2016 crop harvest, with all experiments commencing after harvest of the 1st or 2nd ratoon. The experimental design and plot size varied with site. In Silkwood, Freshwater and the Burdekin, plots consisted of large scale strips 6-8 cane rows wide and the length of the cane block, with yield (and in the case of Silkwood and Freshwater, runoff water quality) collected from the whole treated strip. The Burdekin trial contained three replicate strips of each treatment, but due to the extensive water quality monitoring equipment requirements at Silkwood and Freshwater, treatments were not replicated. At all other sites, trials consisted of smaller plot, replicated experiments in a randomized block design. Plot size was at least 6 cane rows wide 30m long, and all treatments were replicated four times. The basis of fertilizer rates was either the District Yield Potential (DYP, currently used to determine the fertilizer N rates in 6ES) or the Productivity Zone Yield Potential (PZYP, used to determine N rates aligned to a spatially relevant yield target). The PZYP was calculated from the mean yield from block or satellite records over two or more crop cycles, plus 2 standard error of that mean. As all sites were established in ratoon crops, plant crop yields were generally excluded from this calculation, especially where those yields were markedly higher than yields of the ratoons. In situations where large variation in yields occurred between La Nina and normal or drier seasons (e.g. in the wet tropics), separate PZYP targets were calculated to reflect the expected seasonal forecast (i.e. lower PZYP targets in forecast La Nina conditions). Each site hosted a Nil N treatment each year (fertilizer N was withheld for that growing season), but these plots/strips were relocated to new plot/strip locations annually. Having the Nil N treatment always located on a plot with a history of fertilizer N application provided a realistic assessment of the soil N supply which the fertilizer N application was designed to augment. Crop harvest and fertilizer application were conducted as per grower normal practice at each location, although at all sites there were no crops harvested in the 1st round. This was considered desirable, as it was expected that the best chance to assess the risks of reduced N rates and the efficacy of EEF’s would be under conditions where fertilizer N losses were more likely to occur (i.e., where the onset of the monsoonal wet season occurred before the crop had finished the majority of biomass N accumulation). Fertilizer N sources The same fertilizer N sources were used at each site. The fertilizer N standard was taken as granular urea, which was applied during the month following harvest of the preceding ratoon. This was compared to an EEF blend consisting of 1/3 urea coated with the nitrification inhibitor 3,4-dimethylpyrazole phosphate (DMPP, marketed commercially as Entec®) and 2/3 polymer-coated urea with a reported 90-day release period (product of Everris Pty Ltd and marketed as Agromaster Tropical®). This blend was chosen as the best possible combination of products that would protect fertilizer N from risk of loss – initially by retaining the N in the NH4-N form, and subsequently by slowing the release of urea-N into the soil solution. Both products were applied using either stool-split (Burdekin, Freshwater, Mackay and Silkwood) or subsurface side-dress (Tully) fertilizer applicators, although it should be noted that on occasions the stool-split applicators did not always effectively close the fertilizer trench and cover the fertilizer band with soil. This suboptimal application strategy contributed to some confounding of the benefits of EEF use in some seasons due to greater loss risks to both the atmosphere and in runoff. Fertilizer N recovery, crop yield and indices of fertilizer N use efficiency Fresh and dry biomass and crop N content were determined from hand-cut biomass samples collected from 7-10 months after fertilizer application, on the assumption that at this stage, the crop N content would be at a maximum, and most relevant to the yield-determining processes. Crop N was partitioned between leaf/cabbage/dead leaf and stalks at that time. In situations where biomass sampling was conducted a little earlier than desirable (e.g. due to an impending cyclone), smaller numbers of whole stalk samples were again collected for dry matter and N concentration immediately prior to harvest (to determine the partitioning of N between harvested and non-harvested portions of the crop), and stalk N concentration from the final harvest was used in combination with cane yields to estimate crop N removal. Yields were determined by commercial harvest in the case of the large strip plots, with the bins collected from each strip weighed and ccs determined at the mill. In the case of the small plot trials, yields were determined from small plot hand harvesting and ccs was determined by near infrared spectroscopy. A number of indices of N use efficiency were calculated from these data including – -Agronomic Efficiency of fertilizer N use (AgronEffN) = Fertilizer N rate/(YieldN1 – YieldN0) = kg fertilizer N required to produce an additional tonne of cane yield. In this calculation, YieldN1 is the cane yield at fertilizer rate N1, while YieldN0 is the yield with no N applied. -Nitrogen uptake efficiency (NUpE) = (Crop N1 – Crop N0)/Fertilizer N rate = the additional crop N uptake/kg fertilizer N applied. In this calculation, N1 is the biomass N content for N rate 1, while N0 is the biomass N content with no applied N fertilizer. -Nitrogen Utilization Efficiency (NUtE) = Yield/Crop N content = t cane produced/kg of crop N uptake. This figure is a very useful indicator of trial sites where yield is constrained by factors other than N (e.g. waterlogging). Runoff and drainage losses of N Surface water runoff and drainage below the root zone (1 m depth) were monitored in four of the fertilizer rate treatments at Silkwood and Freshwater sites. In addition, strategic sampling in the farm drain around the Silkwood block was undertaken. Surface water samples were collected by automated samplers, with sampling undertaken across the hydrograph (Freshwater) or as an integrated composite of runoff from each individual runoff event (Silkwood). Runoff samples were analysed for sediment, total nitrogen, urea, ammonium-N, and nitrate-N (in addition to other constituents). Drainage samples were collected from 5 lysimeter barrels in each of the treatments with runoff monitoring (totalling 20 barrels) on a weekly to monthly basis at Silkwood. Drainage samples were analysed for nitrate-N and ammonium-N concentrations. Limitations of the data - Data collectively represent variable soil types and seasonal conditions, so extrapolation from particular sites to other seasons, regions or soil types should be undertaken with extreme caution. Similarly, the performance of the EEF fertilizer blend used in this study is specific to the products used, and extrapolation to a broader range of EEF technologies would not be appropriate. Format: The data consists of a series of spreadsheets containing crop summary data for each successive crop season at all sites. As of March 2019, the dataset is complete for 2 crop seasons at all sites. Data are presented as treatment means with statistical significance (P<0.05) indicated where appropriate. Data Dictionary: -EEF – Enhanced Efficiency Fertilizer. In the context of these trials, refers to a blend of 1/3 urea coated with the nitrification inhibitor DMPP (Entec ®) and 2/3 polymer coated urea (Agromaster Tropical ®) with a reported 90d release period. -Agronomic Efficiency of fertilizer N use (AgronEffN) = Fertilizer N rate/(YieldN1 – YieldN0) = kg fertilizer N required to produce an additional tonne of cane yield. In this calculation, YieldN1 is the cane yield at fertilizer rate N1, while YieldN0 is the yield with no N applied. -Nitrogen uptake efficiency (NUpE) = (Crop N1 – Crop N0)/Fertilizer N rate = the additional crop N uptake/kg fertilizer N applied. In this calculation, N1 is the biomass N content for N rate 1, while N0 is the biomass N content with no applied N fertilizer. -Nitrogen Utilization Efficiency (NUtE) = Yield/Crop N content = t cane produced/kg of crop N uptake. This figure is a very useful indicator of trial sites where yield is constrained by factors other than N (e.g. waterlogging). Data Location: This dataset is filed in the eAtlas enduring data repository at: data\NESP-TWQ-2\2.1.8_On-farm-nitrogen-management\data **This dataset is currently under embargo until the project end date

Principal Investigator
Bell, Michael, Prof University of Queensland
Co Investigator
Moody, Philip, Dr Queensland Government · Department of Science, Information Technology, Innovation and the Arts (DSITIA)
Co Investigator
Skocaj, Danielle, Dr Sugar Research Australia
Co Investigator
Masters, Bronwyn Qld Dept Natural Resources, Mines and Energy
Co Investigator
Fries, Jakob Qld Dept Natural Resources, Mines and Energy
Co Investigator
Dowie, Jayson Farmacist
Co Investigator
Webster, Tony CSIRO
Co Investigator
Turner, John Farmacist
Point Of Contact
Bell, Michael, Prof University of Queensland m.bell4@uq.edu.au

Data collected from 01 Dec 2016 until 31 Mar 2021


Data Usage Constraints
  • Attribution 3.0 Australia