Moa Technology is on a mission to bring new mode of action herbicides into the world, to break weed resistance and ensure food security. My colleagues based in our Oxford laboratories do amazing work to be able to identify classes of chemistry that have the potential to fill that need. After glasshouse screening, the most advanced compounds move into our international field trial program where we test these new herbicides against farmers’ problems for the first time, outside of control conditions.
To do this, we set up field efficacy trials in which moves us away from laboratory science and closer to field ecology. Efficacy trials are set up in real world situations, often in farmers fields, with real crops and with application methods that are much closer to modern on-farm technology; the scale is still small – plots within the experiment are typically 1m2 but that a colossal leap from the plant pots that we use in the glasshouse evaluations.
Whenever I talk to farmers, wherever they are in the world, I can be sure that, at some point the conversation will cover either the weather or the soil (OK, there might be a bias here when they find out what I studied during my PhD, and what I have spent the last 15 years of my career doing!). The point is that every farmer has their own specific challenges for production. Many of these are driven by different weather climates and soils conditions which can vary significantly. Research during the green revolution[1] establish that different world agro-climate zones had different weed problems[2]: we call this field variability. A weed is only a plant growing in the wrong place; Italian rye grass, Lolium multiflora is an important weed for arable farmers in France, Spain and Australia (to name just a few) but for a livestock farmer in the south of England, it’s the pasture grass of choice. So when we take our most advanced compounds into the outside world to test, we need to understand all of this variability in order to stand a chance of being able to differentiate the signal from the noise in the results we record.
Fortunately, there is a rich history of agricultural research both in the UK and globally which we can lean on for support. While we do not need to emulate all of the statistical methods that were developed at the long term agronomy trials at Broadbalk in Rothamsted[3], we do draw on contemporary understanding from plant protection product development by running a global trial program to ensure our compounds perform in a range of climate zones and crop systems[4]. Like any experimental scientific process, we start by defining exactly what we want to know as the outcome of the experiment or the technical question. This can be something as simple as “does this compound have herbicidal activity at one fixed application rate”, but often they become more complex as we seek to maximise the insights we can draw from every trial.
We select our trial locations based on being able to address these technical questions, and doing so in such a way that we can take into account the variability in climates, soil, crops and weeds. Once we have determined the where, we have to design the what. A statistically valid experiment with testable hypotheses which can be carried out in every one of those selected locations. More often than not, we set up a randomised complete block design where we have 4 replicates and a minimum of 6 treatments. This means that we always have at least 12 residual degrees of freedom, which is enough statistical power in the design to be able to discriminate treatment difference on a statistical basis even in the (experimentally) noisy world that is a farmers field. The trials that are designed by our in-house agronomy team to ensure consistency across multiple countries, allowing the results to be collated and analysed as a whole. Every trial has a few “must haves” to make them valid. They all have an untreated control so we know the weeds died because of the herbicide, they all have a global standard – a registered herbicide that is in every trial, so that we know that the trial is agronomically valid. Treatment lists can fill up really quickly which is why we only aim to take the very best compounds from our pipeline into this kind of testing.
As you can read elsewhere on the Moa website, this approach has brough us some stunning successes, but we don’t – we can’t – stop there. It goes back to that conversation I have with every farmer about the weather, every season is different, the weeds don’t stop coming. To develop solutions that truly solve global weed resistance problems we need to test accounting for as much of the field variability as possible. The best way to do this is with counter season testing – in the Southern Hemisphere spring becomes summer when autumn becomes winter in the Northern Hemisphere. This means that we can keep our pipeline of discovery flowing as fast as possible and maximise every opportunity to test. Our Southern Hemisphere testing program in the second half of 2024 will involve further trials of the chemical classes tested in the Northern Hemisphere, as well as some new analogues which the glasshouse team have shown are even better than what we have already tested. How they do that, is a story for another day.
[1] Jenny, Hans. 1958. Role of the plant factor in the pedogenic functions. Ecology. 39: 5-16.
[2] Geiger, Rudolf (1954). “Klassifikation der Klimate nach W. Köppen” [Classification of climates after W. Köppen]. Landolt-Börnstein – Zahlenwerte und Funktionen aus Physik, Chemie, Astronomie, Geophysik und Technik, alte Serie. Vol. 3. Berlin: Springer. pp. 603–607.
[3] Fisher, Ronald A. (1921). “) Studies in Crop Variation. I. An Examination of the Yield of Dressed Grain from Broadbalk”. Journal of Agricultural Science. 11 (2): 107–135. doi:10.1017/S0021859600003750
[4] E. Bouma. 2005 OEPP/EPPO, Bulletin OEPP/EPPO Bulletin 35, 233–238
