Use of precision agriculture by grain growers: some leaps, many small steps

| Date: 23 Feb 2011


Use of precision agriculture by grain growers: some leaps, many small steps


Rick Llewellyn, CSIRO

Introduction
There have been major advances in technology for precision agriculture and considerable effort aimed at increasing its application in Australian cropping over the past decade. The adoption of some technologies that can give immediate benefits has been relatively rapid and widespread (e.g. autosteer). However, the application of precision agriculture technologies to within-paddock spatial management remains relatively low. The focus of this paper is on this latter category, where inputs are varied according to within-paddock spatial variation and spatial information is used to improve decision-making and management. A snapshot of the current status of PA adoption is presented using available data on variable rate fertiliser management and yield mapping across regions. This is followed by consideration of what will be required for widespread adoption of spatial management in the future.


A snapshot of current adoption
The results presented in Table 1 capture just two aspects of precision agriculture on a national scale that includes 5 selected SA regions (see report cited below for further details). The questions used a relatively ‘low bar’ for adoption to reflect the fact that the adoption process is still in early stages. Although a range of inputs can be varied, variable fertiliser rates is the focus of the results presented here.
The results show that 21% of SA growers were varying fertiliser rates according to identified paddock zones which is close to the national average. This is likely to include many growers who are using a ‘manual’ approach to applying varying inputs within paddocks. Nationally, a majority of growers with yield maps don’t use variable rate fertiliser. Fifty-nine percent of growers who varied fertiliser rates within paddock did not have a yield map.
Table 1. Use of within-field variable rate fertiliser application and yield mapping by Australian grain growers, 2008-09. 


State
Region
Have at least one crop yield map (%)
Vary fertiliser rates on identified field zones in at least one field (%)
Vary fertiliser rates on identified field zones in at least one field AND have at least one crop yield map
New South Wales
 
Central West
19
20
6
Northern
28
19
9
Southern
26
17
10
 All
25
18
8
Queensland
Southern
26
14
2
South Australia
 
Central
20
13
7
Lower Eyre Peninsula
32
20
10
Mallee
17
24
9
Upper Eyre Peninsula
20
32
5
Western Eyre Peninsula
8
15
3
 All
19
21
7
Victoria
Loddon
24
20
9
Mallee
24
35
18
Wimmera
23
11
4
 All
24
23
11
Western Australia
 
Northern
41
16
11
Central
40
22
9
South west
29
21
9
All
37
20
10
 
All respondents (n=1170)
25
20
8

NB. Data from interviews conducted in 2008 with 1170 randomly selected primary cropping decision makers (cropping greater than 200 ha) within major grain growing regions (Llewellyn & D’Emden 2010).


The highest use of yield maps was in the WA Northern and Central regions (41%), with Lower Eyre Peninsula also high (32%) (Table 1). Nationally, 20% of farmers stated that they were using variable fertiliser rates on identified zones in at least one field. This varied between 11% in the Victorian Wimmera to 35% in the Victorian Mallee. In SA only 7% were using VR fertiliser on identified zones in at least one field and had at least one crop yield map.
The use of yield mapping and variable fertiliser rates is positively associated with larger area of land managed and higher proportion of land cropped (Table 1). Use of a consultant and higher education is associated with greater likelihood of yield mapping. However, the ability to explain adoption among the very early adopters is generally weak as personal traits other than those below tend to play a more important role than for the later majority of adopters. In addition to the factors shown in Table 2, regions that were positively associated with greater use of varying fertiliser use within identified paddock zones and yield mapping were the Victorian Mallee and Northern WA wheatbelt.


Table 2: Factors significantly associated with use of yield mapping and varying fertiliser rates on identified zones by Australian grain growers

 
Yield mapping
Vary fertiliser rates
Vary fertiliser rates and yield mapping
 
Higher cropping %
+*
+*
+**
Larger farm area
+***
+*
ns
Higher education
+***
ns
ns
Use cropping consultant
+***
ns
+**
Vary fertiliser rates in-paddock
+***
 
 
Have a yield map
 
+***
 

*P<0.10; **P<0.05; ***P<0.01; ns – not statistically significant; +/- indicates direction of influence. Based on logit regressions (all models P<0.001).

Some observations on PA adoption:
Convenience and the cost of complexity
Some precision agriculture technologies such as GPS-guidance that can readily contribute to simplicity and convenience of farm operations are already widely used. ‘Embodied’ technologies such as these offer direct benefits as soon as they are used (e.g. reduced overlap). A new crop variety with disease resistance is another example of an embodied technology. However, use of PA technologies for site-specific management is still in the early stages of adoption. Whilst many of the technical PA hardware-software complications are likely to be overcome soon, it will always be the case that variable rate technology requires ongoing information analysis and decision-making that involves some form of cost (e.g. management time) to gain value from it. By nature it is information intensive and potentially complex so 1) rapid adoption rates should not be expected; 2) a major role for advisory and support services will be important; 3) growers are likely to take a step by step approach to adoption. Where the well-understood nature of the land system has allowed for more simple ‘convenient’ zoning methods and there has been strong advisory support for spatial management there have been higher levels of adoption of variable rate e.g. Victorian Mallee.
There’s a high road and a low road
Australian farmers are adopting variable rate and PA technology via different paths. For example, there is not a strong correlation between levels of yield mapping and the use of variable rate fertiliser (Table 1). Some growers are beginning with a relatively ‘low-tech’ approach to varying inputs by soil type and then later building on their existing knowledge by increasing their use of PA technologies and spatial information. This approach has contributed to higher early levels of use of variable rate fertiliser in the regions where the landscape allows paddock zoning to be done relatively simply (e.g. dune-swale), particularly where opportunities for cost-effective input reduction can be identified. Another path is the one based on collection of a robust set of multiple layers of spatial information, analysis for potential zoning decisions and then application of variable rate technology. This is more information intensive, involves greater initial complexity and will require greater support services to promote initial adoption (Table 2). There is an important role for support services and consultants in both of these paths to reaching the potential profitable use of precision agriculture technology.
It’s not just about adoption by growers
A majority of grain growers pay an advisor to provide cropping advice. Growers using more complex cropping systems are far more likely to use a paid advisor e.g. Australian growers using no-till are more than twice as likely to use a paid advisor as those that don’t. Growers using yield mapping (and using variable rate fertiliser with yield mapping) are more likely to use consultants but it is not a strong association. This is because most growers use a consultant, but not for PA-related advice. Before the majority of potential PA adopters apply PA technology to spatial management, a much greater proportion of advisors will need to have ‘adopted’ PA-approaches. When it comes to PA adoption, both growers and consultants face a very similar question: is it worth investing in the time, cost and effort to be able to apply PA technologies in my business? There is evidence already that pro-active ‘adoption’ of PA-approaches by strong local sources of advisory support can drive more rapid adoption by growers (e.g. a comparison of adoption levels in the Victorian Mallee to similar agro-ecological districts).
It’s still early days
In many regions there is a common expectation among growers that they will be using PA at some stage in the future (Robertson et al 2011). Declining hardware costs and resolved compatibility issues will help these expectations to be met. It is worth keeping in mind that currently, we are generally still only observing adoption by early adopters of variable rate. The factors that influence very early adoption are not likely to be the same as those that explain future peak levels of adoption in different districts. Adoption by the middle to late majority of adopters will be influenced more by the demonstrated level of profitability or relative advantage for them locally. These differences are not yet being fully reflected in current adoption rates (Table 1). On some farms and in some land systems the level and type of in-paddock variability has the potential to be able to managed much more profitably than in others. After over 20 years we still see very different levels of use of no-till between districts and we should expect to see very different peak levels of variable rate technology use between regions in the future. A R&D challenge is to identify where the greatest potential for highly profitable use of in-paddock spatial management is likely to be, based on the type and level of inherent variability. This will allow investment in developing management and support services to be better targeted. Extensive adoption will require not only greater demand for PA by growers, but greater adoption of PA by agronomy advisors. Generating greater confidence in the relative advantage of PA approaches in particular regions is still needed for both to happen.

References
Llewellyn RS, D’Emden, F. 2010. Adoption of no-till cropping practices in Australian grain growing regions. CSIRO report for GRDC, SA No-till Farmers Association and CAAANZ. 32pp. www.grdc.com.au/notill_adoption
Robertson, MJ , Llewellyn, RS, Mandel, R, Lawes, R, Bramley, RGV, Swift, L, Metz, N, O’Callaghan, C. 2011. Adoption of variable rate technology in the Australian grains industry: status, issues and prospects. Precision Agriculture (in press)

Rick Llewellyn
CSIRO, Waite Campus, Adelaide
rick.llewellyn@csiro.au