Some of our best researchers...

Av Singh, Ph.D.

For a long time organic farming has remained on the fringes of most university and public institutions research programmes. The newly formed Organic Research Centre of Canada (OACC) has the mandate to place greater emphasis on addressing the research needs of the organic agricultural community. However, this article is not about us...it’s about those who have been (and will continue to be) pioneers in organic research - the farmer. Farmers, conventional or organic, seeking to cut production costs, to maximize yields, or to enhance stewardship of natural resources often experiment with new methods and implement new innovations and ideas into their farming system. Carrying out simple research experiments can provide valuable answers to production problems. But, the key to knowing that your on-farm research is in fact generating meaningful results, is to ensure that the experiment is designed and conducted properly. So, this article will highlight some important tips on conducting on-farm research.

Farmers are great at coming up with ideas for research, which in my mind is the all important first step. From the idea, we have to move towards a viable research objective. If my idea was: “I wonder what would happen to the crop yield if I plowed-down my cover crop later in the season?”, then I would need to phrase it into a testable research question, such as: “To determine the effect of cover crop kill date on the yield of the following crop.” The preceding objective clearly states what is going to be tested (kill date of cover crop) and which effects will be measured (yield of following crop). With a clear objective, we move to the design of the experiment, where replication and randomization are key. Replicating your treatments gives you greater confidence that the results you see are in fact an effect of the treatments rather than just random variation (chance). Without replication, data analysis in a valid statistical manner is extremely difficult, so most scientists advise at least 3 replications. It is equally important to randomize your treatments, thereby eliminating a potential bias in the system. In our example, if we were testing two kill dates for our cover crop and we replicated the treatments 3 times, but we always had the early kill date treatments along the tree line, while the late kill date plots were away from the influence of the trees, we might not get the “real” results. Key times to seek professional assistance (from OACC, universities, Agriculture and Agri-Food Canada, or your provincial agriculture specialists)are designing the experiment and later on during data analyses. A mistake at the design stage can render your data unusable, or even worse, misleading.

By constructing a specific design that includes randomization and replication you have eliminated (or at least reduced) some potential sources of variation (stuff that makes it hard to make heads or tails of your results). However, there are other potential sources of variation and they include treatment applications and data collection. Treat every plot exactly the same except for that part that is intentionally varied (i.e., the treatment). So, in our example just because one of the treatment plots looked like it needs irrigation, you can’t irrigate it unless you irrigate all of them. Similarly, data collection is another potential area to make mistakes. All measurements should be taken under the same conditions using the same methods (i.e., don’t use a disc harrow to kill one cover crop and use a mower on another, unless the method of killing the cover crop is what you want to test). It is also very important to record data from each individual plot, and not to lump all treatment types together because you’ll lose the value of replication.

Lastly, some other general tips that you may find useful in producing confident results would include: 1) Maintain some form of objectivity...the results may not turn out as you hoped or planned. There is a lot to learn from “negative” results; 2) Climate will influence your results...you may want to repeat your experiment over years until you feel comfortable with the results under varying environmental conditions, 3) Your eyes are your greatest tools...visual observations made during the course of your experiment are extremely beneficial in understanding your final results. You should also look for changes outside your test parameters, for example, you may notice that you had better weed control when you used a late cover crop kill date versus an early, or perhaps the opposite was true; and 4) Manage your time wisely...expect to devote extra time to your research during busy harvest seasons, you’ve done all the work, it’s well worth maintaining the integrity of the experiment so your results will be meaningful.

This last point is one that deserves more attention. The workload of a research experiment added on to the already busy life on the farm is often a deterrent for initiating research. In many areas, groups of farmers have banded together (e.g., Practical Farmers of Iowa) to conduct on-farm research about a topic of interest, which not only helps spread the workload, but also creates a network where farmers can bounce ideas off each other. Farmer research teams work especially well when university, government, or nonprofit organization researchers join as part of a “participatory” research team (e.g., Pesticide Free Production Canada). The power of participatory research comes from combining the experience, resources, and perhaps most importantly, creativity of many people to address a common problem. An additional benefit gained by such a process is that the data generated from several farms is often more reliable and practical than a trial conducted at one or two locations. Academic researchers often provide farmers a greater understanding of some of the “science” in their production systems, while researchers benefit, by seeing the “science put into practice” in the real world context of farms. Through participatory research, knowledge from the farmer and the researcher is integrated to produce new knowledge helping others to modify their management practices.

Currently, the OACC in collaboration with other scientists and farmers has initiated several on-farm research projects throughout Atlantic Canada, Alberta, and Manitoba. Students from across Canada will be conducting on-farm research projects ranging from alternatives strategies for pest management (e.g., compost teas in potatoes), varietal selection for organic tree fruit production, transitioning to organic dairy production, agronomic practices to reduce weed invasion in organic grain production, forages as mulches in row and cash crops, and several others.

For those farmers/researchers planning on conducting some of their own on-farm research trials please send the OACC an email describing your ideas and if you have any questions we will be glad to be of assistance. Perhaps in future years we can forge an intellectual partnership and collectively gain a better understanding of the complexities within organic systems.

For more information please call the Organic Agriculture Centre of Canada at 902-893-7256 or email oacc@nsac.ca