Previous PageTable Of ContentsNext Page

Precision Agriculture for improved environmental outcomes

Rob Bramley1, Peter Thorburn2, Patricia Hill1, Frederieke Kroon3 and Kerstin Panten1,4

1CSIRO Sustainable Ecosystems, PMB 2, Glen Osmond, SA 5064. Email Rob.Bramley@csiro.au
2
CSIRO Sustainable Ecosystems, 306 Carmody Road, St Lucia, QLD 4067.
3
CSIRO Sustainable Ecosystems, PO Box 780, Atherton, QLD 4883.
4
Present Address: Julius Kuehn Institute (JKI), Bundesallee 50, D-38116 Braunschweig, Germany

Abstract

Using examples from the sugar and dairy industries, this paper shows how the use of spatial data to better inform agricultural management can also make a valuable contribution to reducing the risk of negative environmental impact, whether applied in the practice of Precision Agriculture (PA; paddock and sub-paddock scales) or at whole farm or regional scales. However, for maximum environmental benefit to accrue through PA, existing regional or whole-of-industry management guidelines need replacing by those for site-specific management - a change which requires a marked enhancement to current agronomic understanding.

Key Words

spatial variability, targeted management, scale (paddock, farm, region), nitrogen loss, water quality

Introduction

Agriculture is increasingly under pressure to meet public demands for improved environmental performance, sustainability of practices, and accountability for the traceability, quality and safety of its products (Ancev et al., 2005). Given that “improved environmental performance” is often taken to mean that the impacts of agriculture on water bodies (groundwater, rivers, lakes and the sea) will be minimised, an obvious question is how might it be achieved and demonstrated?

Because of the inherent variability of land, the input-output relationships driving agricultural production systems vary spatially, often over short distances (a few metres). PA promotes better understanding of these relationships and provides a means of targeting management so that “the number of correct decisions per unit area of land per unit time with associated net benefits” (McBratney et al., 2005) is increased.

Intuitively, if a producer makes a “correct decision” in regard to an input such as nitrogen (N) fertilizer, its efficiency of use should be enhanced, with the consequence that the chance of N leaching to groundwater or being exported off-site via runoff is reduced. Some industries have been quick to use this idea as a basis for promoting both the adoption of PA and, in cases where there has been some adoption, the environmental credentials of the industry (eg Wrigley and Moore, 2006). But is such an approach justified? We address this question here with a focus on the environmental component of “net benefits” (McBratney et al., 2005) in examining a role for PA in protecting water bodies from negative impacts of agriculture.

Precision Agriculture, sugarcane production and protection of the Great Barrier Reef

The Great Barrier Reef (GBR) contributes an estimated A$6.9billion per year to the Australian economy, principally through tourism activities in the GBR Marine Park. Intensive cropping, predominantly located in coastal floodplains and dominated by sugarcane production, is a major land use in several catchments which drain into the GBR. The close proximity of intensive land use to the GBR, coupled with strongly seasonal tropical rainfall, raises the likelihood that land use signals may be evident in both fresh and marine waters, especially during peak flood events.

Bramley and Roth (2002) demonstrated that compared to grazing and forestry, sugarcane production in the valley of the Herbert River had a significant impact on riverine water quality, as evidenced by higher concentrations of N, phosphorus (P) and total suspended solids in stream-waters draining land under sugarcane. They concluded that, in regard to minimizing the off-site export of nutrients and sediments, there was much room for improvement in land management. The Reef Water Quality Protection Plan (State of Queensland and Commonwealth of Australia, 2003), and consequent development of Water Quality Improvement Plans (WQIPs) for all coastal catchments draining into the GBR are one response to this. WQIPs set target levels for sediment, nutrient and pesticide loads in waters entering the GBR. Questions therefore arise as to what actions individuals or industries might take in order to ensure that the objectives of WQIPs are met?

Thorburn et al. (2007) have recently proposed and tested a strategy for N fertilizer management for sugarcane based on the maintenance of nutrient balance through replacement. They proposed that the amount of N to be applied this year be based on that removed in the previous crop, plus an amount unavoidably lost to the environment; these combined N losses equate to 1 kg N t-1 of cane. They suggested that this strategy may deliver significant environmental benefits over conventional practice without compromising profitability. Because this approach depends on a knowledge of yield in the previous year, it lends itself to application where yield mapping forms part of the implementation of PA. We were interested to see whether additional benefits might accrue through integration of this strategy with variable rate fertilizer application.

The analysis, conducted over a 10 year period (1996-2005) is based on a 6.7 ha block of sugarcane in the Herbert River district (Bramley and Quabba, 2001); we had a yield map for this block for 1998. Yield maps for other years were simulated using linear scaling of the 1998 block mean against the known annual district means (Figure 1a) ignoring the tendency for yield decline through the ratooning cycle. Annual rainfall data accessed from the Australian Bureau of Meteorology were used as the basis for modifying the range of variation present in the simulated yield maps on the assumption that within-paddock variation was likely to be greater in dry, compared to wet years. A random number generator was used to generate estimates of the coefficient of variation (CV) in the simulated yield maps between values of 20 and 40%. These estimates were ranked and assigned to individual years in rank order of annual rainfall. The standard deviation of mapped yield was calculated from these CVs and the annual mean estimated block yield. The normalised 1998 map was then back-transformed to give simulated maps for 1996-2005. The k-means clustering of all yield maps were used to identify 2 management zones (denoted L and H) for targeting of N fertilizer (Figure 1a).

Based on fertilizer usage statistics obtained from a major sugar industry supplier, ‘standard practice’ was assumed to be uniform application of 190 kg N ha-1 y-1 - the mean rate used in the district over the 10 year study period. The yield maps (Figure 1a) were then used to estimate the potential loss of N to the environment, and its spatial variability within the block, based on crop removal of 0.9 kg N t-1 yield in the previous year (Thorburn et al., 2007). In addition to ‘standard practice’, other fertilizer strategies evaluated were: the N replacement strategy of Thorburn et al. (2007) assuming uniform application of 1kg N ha-1 t-1 mean yield achieved in the previous year, with crop removal remaining at 0.9 kg N t-1 (Nrep); modification of the Nrep strategy with the paddock divided into 2 management zones (Figure 1a; zone based); modification of the zone-based strategy with N applied at rates of 0.7 or 0.8 kg N ha-1 t-1 yield in previous year to the higher and lower yielding zones but with 0.6 kg N t-1 removed in the crop (eff); and the ‘eff’ strategy when implemented using continuous variable rate fertilizer technology (VRT; 1998 only). The VRT was based on a fertilizer application map (10 m pixels) derived by locally averaging N requirement derived from the 1997 yield map (2 m pixels). Aside from assuming a more efficient use of N, the ‘eff’ strategy also recognises that one consequence of the poorer performing parts of the paddock being low yielding is that they are also areas of less efficient N use and so may be larger contributors of N to the environment. Partial gross margins were estimated for each strategy as income from sales of cane (sugar price of US$0.12 lb-1 = $340 t-1), less the costs of fertilizer (urea @ $460 t-1) and harvesting ($6.50 t-1). For simplicity, the sugar content (ccs) of harvested cane was assumed invariant in both space and time and was set at 12.

Figure 1b shows the implications for potential N loss to the environment of the various fertilizer strategies (except VRT) over 9 harvest seasons from 1997. As can be seen, much of the paddock may potentially leak approximately 1 t N ha-1 over 9 years under uniform N application (standard practice), whereas for each of the strategies based on N replacement, at least some parts of the block have no N leakage at all. Note however, that because the Nrep strategies except VRT depend on calculation of mean paddock yield, irrespective of whether this is partitioned into zones, the Nrep and zone based strategies yield almost identical results in terms of total surplus (Table 1) and so Figure 1b essentially illustrates the effects of the different strategies on spatial variation in N use efficiency. However, the results presented in Figure 1b strongly suggest that not only does ‘standard practice’ represent an inefficient and environmentally risky use of N, but that the Nrep strategy also assumes a higher N requirement of sugarcane than is in fact the case. Indeed, Thorburn et al. (2007) present field data in support of this view. Thus, application of N following the ‘eff’ strategy results in further reductions in potential N loss, but importantly, does not impact on the financial performance of the paddock (Table 1). Also apparent from Figure 1b is that a lack of perfect knowledge about inter-annual variation in yield potential, driven primarily by variation in climate, results in the possibility of N being in deficit in parts of the paddock in some years. Surprisingly, VRT in 1998 led to a greater N surplus than the ‘eff’ strategy. However, comparison of maps for 1998, analogous to those in Figure 1b with that for the VRT strategy (not shown) suggests that VRT may result in less spatially variable N use efficiency than the ‘eff’ strategy; the long term agronomic implications of this are unclear.

a

b

Figure 1. Estimations of (a) yield (1996-2005) and (b) N surplus (1997-2005) in a 6.7 ha sugarcane paddock.

Table 1. Summary of implications for profitability and environmental performance of selected N fertilizer management strategies in a Herbert River sugarcane paddock, 1997-2005.

Standard Practice

Nrep

zone based

Eff

VRT

 

1998

1997-05

1998

1997-05

1998

1997-05

1998

1997-05

1998

                   

N applied (kg)

1,273

11,457

704

5327

725

5368

547

4071

502

N surplus (kg)

680

6,792

108

644

129

678

10

-168

104

Gross margin ($)

11,065

85,620

11,601

91,602

11,616

91,734

11,794

93,031

11,164

Best practice dairying is dependent on access to spatial data

Intensive dairying is a major land use in the Gippsland region of Victoria. However, in a somewhat similar situation to that confronting the sugar industry in Queensland, the region is bounded at its most downstream end by the Gippsland lakes, a significant wetland area of major ecological significance. As with the GBR, there are concerns that agriculture, especially intensive dairying, may negatively impact on the lakes.

The Farm Nutrient Loss Index (FNLI; Melland et al., 2007) was developed as an educational tool to help Victorian dairy farmers understand about nutrient budgeting and loss. Recently, Hill (2008) has implemented a modification of the FNLI (Figure 2a) in GIS to facilitate examination of the hazard of nutrient loss in agricultural landscapes. Initial testing was done at Ellinbank, a 180 ha research dairy farm operated by the Victorian Department of Primary Industries at which best management practice is considered to be implemented with respect to all facets of dairy production.

Risk may be defined as the integration of likelihood and consequence. Hill (2008) considered the likelihood of N loss at Ellinbank as being a product of the availability of N, and the potential for available N to be lost to the environment. N availability was calculated by using per paddock estimates of N inputs and outputs, using detailed farm records of fertilizer N applied, the amount of milk produced, the liveweight gain of the cattle, and estimates of N fixation by legumes, N in imported feed and applied via dairy shed effluent, and exported in manure and urine and through exported feed or plant material (hay, silage, etc). The difference between total inputs and outputs gives the surplus that is potentially available for loss to the environment (Figure 2b). The potential for the available N to be lost to the environment is controlled by soil properties and landscape position, which control the likelihood of loss through runoff (Figure 2c), leaching / deep drainage, waterlogging, subsurface lateral flow and gaseous emission (volatilization of urea and nitrous oxide emission). The consequences of loss were based on flow pathways (on and off-farm) and resource condition targets. Thus for example, N getting into the La Trobe River and thence the Gippsland Lakes was deemed far more serious than accession to the farm dam. Integration of the risk of N loss with the calculated N surplus enabled assessment of the risk of N pollution for the Ellinbank property; in Figure 2, we consider loss by runoff only. Two features of Figure 2d are immediately apparent. First, pollution risk, when considered at farm scale is highly spatially variable. Second, even though it is regarded as a ‘best practice’ farm, Ellinbank poses a significant risk with respect to offsite pollution arising from N with the majority of the property having a pollution risk towards the high end of Hill’s (2008) range.

a. Schema for risk assessment

b. N surplus (load)

c. Runoff likelihood

d. N loss risk via runoff

Figure 2. (a) Schema and (b, c) selected components of the assessment of (d) the risk of N loss through runoff.

Conclusions and future directions

Both of these analyses provide messages for farmers and catchment managers which warrant more robust evaluation. The sugar example indicates that PA may make a positive contribution to minimising off-farm losses of N without reducing the profitability of cane production (Table 1) whilst the dairy example highlights the value of spatial information more generally. Both examples also demonstrate that uniform management strategies, which in effect manage for the average condition, are a poor strategy from both production and environmental perspectives. However, it is apparent from both studies that incorporation of spatial data into agricultural management will deliver little benefit, if any, if it is not combined with good agronomy. Thus, the fact that Thorburn et al. (2007) found that the N replacement strategy did not lead to loss of yield suggests that sugarcane can do well at less than 1 kg N t-1. Of course, through yield mapping, PA also enables identification of areas of inherently higher and lower yield and, thus, areas of higher and lower N use efficiency. Similarly, there is room for improved agronomic management on dairy farms – with respect to management of fertilizers and dairy shed effluent in particular (Hill, 2008). Possible next steps might also include a stronger move towards PA through recording pasture production and milk yield on the basis of stocked areas rather than paddocks, and more targeted management for both production and environmental goals. Such a strategy could lead to intensification of some areas but quarantining of others. However, as our sugar example indicates, access to a means of better targeting fertilizer does not deliver the benefit by itself; it only satisfies the objective of putting the right amount in the right place at the right time if the right amount, place and time are known. On-farm experimentation is therefore to be greatly encouraged.

References

Ancev, T, Whelan, BM and McBratney, AB (2005). Evaluating the benefits from precision agriculture: the economics of meeting traceability requirements and environmental targets. In Proceedings of the 5th European Conference on Precision Agriculture. Ed JV Stafford. pp. 985-992. Wageningen Academic Publishers, The Netherlands.

Bramley, RGV and Quabba, RP (2001). Opportunities for improving the management of sugarcane production through the adoption of precision agriculture – An Australian perspective. Proceedings of the International Society of Sugar Cane Technologists 24, 38-46.

Bramley, RGV and Roth, CH (2002). Land use impact on water quality in an intensively managed catchment in the Australian humid tropics. Marine and Freshwater Research 53, 931-940.

Hill, PA (2008). Managing nitrogen losses in rural landscapes - the use of spatial information and risk management frameworks in balancing production and conservation in a dairy case study. PhD Thesis. Australian National University, Canberra. In press/under examination.

McBratney, A, Whelan, B, Ancev,T and Bouma, J (2005). Future directions of Precision Agriculture. Precision Agriculture 6, 7-23.

Melland, A, Smith, A and Walker, R (2007). Farm nutrient loss index. An index for assessing the risk of nitrogen and phosphorus loss for the Australian grazing industries. Department of Primary Industries, Ellinbank, Victoria.

State of Queensland and Commonwealth of Australia. 2003. Reef water quality protection plan; for catchments adjacent to the Great Barrier Reef World Heritage Area. Queensland Department of Premier and Cabinet, Brisbane.

Thorburn, PJ, Webster, AJ, Biggs, IM, Biggs, JS, Park, SE and Spillman, MF (2007). Towards Innovative Management of Nitrogen Fertiliser for a Sustainable Sugar Industry. Proceedings of the Australian Society of Sugar Cane Technologists, 29, 85-96.

Wrigley, T and Moore, S (2006). Public Environment Report 2006. Canegrowers, Brisbane.

Previous PageTop Of PageNext Page