NIAB - National Institute of Agricultural Botany

Orson's Oracle

Hard to believe

Posted on 05/11/2014 by Jim Orson

I’m not fond of museums, but I am a nightmare when I actually visit one. I read everything and so make very slow progress. I remember my family hauling me out of the Elvis Presley museum in Las Vegas because I had barely progressed beyond about 20% of the exhibits whilst they had done the whole lot and had had a cup of coffee.

The same slow progress occurred a few days ago. We were invited to a ‘do’ just down the road at the Imperial War Museum at Duxford and were given the freedom to see the exhibits for most of the day. Needless to say, I went round just one of the many hangars. In this there was one of the two prototypes of Concorde. It seems that it did its cold weather testing above the Equator. This really sounded counter-intuitive so I read on. Apparently the coldest temperatures at 50,000 feet are above the Equator and the warmest temperatures at that height are above the North Pole!

Issues that have occurred in my day job have sometimes sounded counter-intuitive but again have been true. The prime example was when NIAB TAG, in its most recent 60 trials, collated the nitrogen response curves for feed wheat and found that the economic optimum nitrogen doses did not appear to decrease with increasing Soil Mineral Nitrogen (SMN) levels in the soil. In fact, the best predictor of the optimum dose for the yield of feed wheat in individual trials was to apply the average optimum dose identified in the 60 trials rather than follow any of the current recommendation systems.

I must admit that this worried my colleagues and I because it seemed cFertiliser spreadingounter-intuitive and so we read published papers which contained the results of similar trials. True enough, the overall conclusions were the same for situations similar to those in our trials. Interestingly, this also included the database used for the current edition of RB209.  It should be noted that all our trials were done on long-term arable soils where there had been no recent history of organic manure use; we did not have sites on true sands or the true silts. None of our sites had SMN levels above 100 kg N/ha. Please note that I am talking about feed wheats where the level of protein is immaterial.

Our results have been a continuing fascination to me. Subsequently, I developed an alternative recommendation system and tested it against the results of our trials and also the trials carried out by other organisations. It gave almost the same level of accuracy of prediction for the optimum dose in each trial as applying the average dose identified in our database.

The alternative recommendation system I developed was not very novel because it was the same as that used in RB209 except that SMN was assumed to be used at 50% efficiency rather than the assumption of 100% as used in RB209. This simply resulted in the recommended N dose being reduced by 15 kg/ha between index 1 and 2 and also between 2 and 3 rather than the 30 kg/ha in RB209. There is now so much information to suggest that the efficiency of use of SMN is well below 100% that this evidence can no longer be ignored in the next edition of RB209.

Recently we have re-opened an internal email debate on our results and discussed them with a soil scientist who has penned many peer-reviewed papers on nitrogen application to wheat. This correspondence released the genie of canopy management.

Remember, canopy management in wheat? It was the vogue for many years and much research funding was spent on it. It argued that wheat yield was not related to nitrogen dose but was determined by using the nutrient to build an optimum crop canopy to trap solar energy efficiently. I think it failed because the SMN was assumed to be used at 100% efficiency and because by the time the size of the final dose of N could be calculated, it was often too late in the season for it to be fully exploited by the crop.

Let’s go through the maths of canopy management but assume that SMN is used at 50% rather than 100% efficiency:

  • To trap sunlight efficiently in wheat there needs to be a Green Area Index (ratio of the green surface area of the crop per m2 to one m2 of soil surface) of at least 6 by around ear emergence;
  • Each unit of GAI requires around 30 kg N so the total of N required in the crop by around ear emergence is 180 kg N/ha;
  • There is already some N in the crop by the spring, typically around 30 kg N/ha;
  • There is also some in the soil (SMN), typically around 50 kg N/ha which we now assume is used at 50% efficiency by the crop;
  • This means the N demand to be met by applied bag N is 180 kg N/ha less the 30 kg N/ha already in the crop and the 25 kg N/ha from the soil;
  • Therefore the demand for N from the fertiliser is 125 kg N/ha;
  • RB209 assumes that bag N is used at 60% efficiency and so the dose needed to satisfy the crop requirement of 125 kg N/ha is that number divided by 0.6;
  • This equates to a dose of applied bag N of 208 kg/ha.  

This is extraordinarily close to the average economic optimum dose of 205 kg N/ha in our 60 trials when one kg of N costs the same as 5 kg of wheat. The optimum dose of bag N increases as the price of N gets relatively cheaper. This demonstrates that achieving a more complete canopy, either earlier in the season and/or during grain fill, is worthwhile when the cost of N falls relative to the price of feed wheat.

I recognise that I’ve ignored N from rainfall and from net mineralisation in this calculation. Much of the rainfall that is received by the crop occurs by the early spring (when N availability in the soil and crop is assessed) and net mineralisation is limited in the situation of our trials. In the context of our trials, these are relatively background constants and so do not affect the level of reduction in N recommended between indices 1 and 2 and between indices 2 and 3.

It is comforting that canopy management for a typical situation suggests the same optimum dose as the average of our 60 trials. However, some soil scientists maintain that canopy management should not be considered when predicting the optimum N dose of very high yielding crops. They assume that as yields rise, grain N (protein) gets diluted and there must be a limit to this dilution. All I can say is that our trials have shown that doses of around 205 kg N/ha can support surprisingly high yields of wheat. In fact, our results intimate that only a very slight upward adjustment of N dose is needed for yields of feed wheat above 10–11 t/ha. This is because although the N removed from the field in very high yielding feed wheat can be higher, such crops are much more efficient in taking up and utilising nitrogen than average yielding crops.

I’m sure that there must be logical reasons for the temperatures at 50,000 feet to be at their lowest above the equator. Not to be outdone, we’re slowly developing a better understanding of what drives the optimal doses of N for feed wheat. Whilst this results in the improved predictability of the ‘average’ situation, our database suggests that there is a large and unpredictable variation between sites in the efficiency of use of both SMN and applied nitrogen. Trying to establish reasons for such a variation remains a significant challenge to researchers.      

Please remember that milling wheats are a different kettle of fish because additional nitrogen, above that required for yield, is necessary in order to meet the specified grain N (protein) content for higher yielding crops.

Apologies for the length of this blog being as tortuous as my progress round Duxford....

Leave a comment / View comments


Precautionary principle cuts both ways

Posted on 17/10/2014 by Jim Orson

I had a flu jab recently. The vaccine is to protect against the strains of flu mostly likely to occur this coming winter. It got me thinking about the long standing debate over the last few decades about prophylactic versus rational decision-making for pesticide use. Nobody now supports prophylaxis, but there is an issue regarding the definition of rational and responsible decision-making.Flu jab

About fifteen years ago, a Government Committee on Pesticides visited Morley Farm, the home of NIAB TAG in Norfolk. I described the approach to pesticide use on the farm and mayhem literally broke out when I described weed control in sugar beet. The adopted FAR approach meant that every week, for at least three weeks after crop emergence, the crop was sprayed with a tank-mix of herbicides, all at very low doses. Weeds typically had a maximum size of not much bigger than early cotyledon when they were sprayed. After three applications the intensity of treatment was eased according to observations on weed emergence.

The committee was horrified by such an approach because there was no opportunity for us to identify both the species and number of weeds emerging in order that we could take a more ‘rational’ decision on whether or not to spray. This, they claimed, was an essential requirement for responsible pesticide use. I tried and tried to get them to realise that all our trials had shown that, overall, FAR involved significantly lower levels of herbicide use than in the more conventional programmes but their minds were closed. My horror of such an attitude was further reinforced a few months later when at a conference I gave a paper mentioning the FAR approach and the level of pesticide savings involved. In the audience a high ranking scientist in MAFF was vehemently shaking his head in disagreement.

Hence, at around the turn of the millennium it was obvious that in official circles, rational decision-making necessarily involved thresholds. This was despite a large multi-site field trial project in the 1990s investigating the role of weed thresholds for broad-leaved weed control in winter cereals. The results showed that the annual use of lower-than-recommended doses of herbicides used less herbicide over time than the use of thresholds to decide whether or not to treat. In addition, there were high costs associated with doing the necessary field surveys to assess weed numbers for the threshold approach. On a more theoretical level, modellers had also reached the same conclusion for wild-oat control.Sugar beet

It’s now clear to all in the industry that thresholds (including those based on forecasting) are not the universal approach to rational decision-making on pesticide use. This may be because once a threshold has been exceeded, effective and timely control may not be achieved, either because of weaknesses in the pesticides to be employed or because of weather delays. The FAR approach initially seemed wrong to many but was the right solution bearing in mind the limitation of the selective herbicides available.

Very effective pesticides will increase the prospect of thresholds being adopted.  For instance, those who prefer a situation where weed numbers and species are assessed before making a herbicide application in sugar beet should really be on the side of GM glyphosate tolerant beet which currently offers the only hope for the practical adoption of such an approach.

The precautionary principle is clearly behind the widespread inoculation against flu. By the time an epidemic has started and the specific strain identified it is too late to make the vaccine and inoculate the most vulnerable. Those who argue against the registration of some pesticides and GM crops also invoke the precautionary principle to support their arguments. However, the very same people fail to see that farmers may also have to invoke the precautionary principle rather than thresholds when taking carefully assessed risk-based decisions on pesticide use.

The important thing to bear in mind is that there are various approaches that farmers must take when assessing the need for pesticide application. Thresholds are the right approach in some circumstances, particularly for insecticide application. In others, thresholds are both impractical and unwise and alternative precautionary, but risk-based strategies have to be adopted. Unfortunately, such risk assessments are now heavily influenced by the reality of pesticide resistance which results in farmers having to take a more precautionary approach. It should be noted that risk assessments are also necessary for the applicability of adopting cultural measures that may reduce a farm’s reliance on pesticides.

Leave a comment / View comments


Getting high tech

Posted on 09/10/2014 by Jim Orson

I’m getting high tech on my allotment. Well, to be more truthful, I’ve been using low tech means to achieve what is made possible on a larger-scale by ‘precision farming’.

This autumn, instead of digging the whole allotment I’ve not tilled it at all in some places, and in others I’veMustard cover crop dug the small areas where I’ve detected significant resistance to the penetration of a fork. In the intermediate areas I’ve eased the soil but not turned it over. This is almost real-time detection of soil strength determining the extent of cultivation required. I’ve then spread barley grains over the soil surface and covered them with around 5-8 cm of nicely rotted and very crumbly compost.

This is my first attempt at cover cropping and I hope to destroy the barley in December with glyphosate. The intention is that when I direct sow in the spring I’ll have the perfect seedbed for small seeded species. They should establish quickly because the surface layer will have been conditioned by the compost and the layers beneath will have been conditioned by the barley roots rather than slumping under the pressure of winter rainfall.

I realise that this is not original thinking and it mimics much of what a few farmers are now adopting. There’s little doubt that cover crops are now ‘in’ and they may have true benefits for many farmers. It seemed very different only a few years ago when NIAB TAG started research on cover crops at Morley (funded by The Morley Agricultural Foundation and the J C Mann Trust). At that time there was widespread suspicion of cover crops but this research has now proven to be a great asset. It takes many years to generate meaningful results and so it is a significant benefit to be ‘ahead of the curve’.

The results so far suggest that cover crops can just about financially justify themselves in terms of the increased value of the following crops that are receiving the recommended dose of applied nitrogen. The pay back from break crops increases when sub-optimal doses of nitrogen are applied to following crops. In addition, water infiltration rates into the soil have increased resulting in a significantly lower risk of surface ponding and run-off. This means a lower erosion risk and perhaps an opportunity to cultivate earlier in the spring. It’s still early days and, like many other soil conditioning approaches, it is possible that the benefits will increase over time.

Cover cropping is a large subject to research. There are lots of factors to take into account, particularly time of sowing, method of sowing and choice of species. The work at Morley suggests that legume-based cover crops are more beneficial to following crops than those based on brassica species such as mustard or fodder radish. There are two possible downsides to these types of cover crops; firstly the margins are slim and the seed costs are not cheap and secondly they may carry disease to related crop species.

It’s for these reasons that some farmers are choosing oats. Oats from the barn are cheap and do not carry over root disease in cereals, except for a relatively rare and localised form of Take-all. I chose barley because I don’t grow cereals on the allotment (!), it was too late in the year to sow a more exotic form of cover crop and I pinched the barley from a trailer at NIAB (white collar crime?).

As with any new approach, claims of its benefits can be exaggerated. There’s little doubt that cover crops can reduce nitrate losses in the winter. However, does this mean that less nitrogen needs to be applied to the following crop? This seems a logical conclusion, but in fact it is not true. A three-site study in Northern France concludes that whilst cover crops have consistently reduced nitrate leaching and increased nitrogen storage in the soil organic matter they have had, even after their adoption over at least a thirteen year period, no impact on the efficiency of bag nitrogen applied to crops. The same also holds true at Morley, over a shorter term of adoption, where the optimum dose of bag nitrogen for crops is the same in rotations adopting or not adopting cover crops.

Once again this proves why we should do field trials rather than get carried away with theories. It certainly helps to knock on the head the balance sheet approach to nitrogen nutrition adopted by some soil scientists. This method, which is also logically based on the amount of nitrogen the crop is likely to remove less the amount at the start of the season in soil mineral nitrogen and in the crop, is far less reliable than some considerably simpler approaches to predicting optimum nitrogen doses.

Leave a comment / View comments


Achieving high yields

Posted on 01/10/2014 by Jim Orson

Cambridge recently had an open weekend. Amongst other things, it includedCambridge Backs the colleges opening their doors and also, in some cases, their gardens and libraries. There was an Elizabethan theme as it is 450 years since Good Queen Bess visited the city; hence some colleges had wonderful Tudor manuscripts on display. One of the events was Evensong at Kings College in the presence of Queen Bess (a professional look-alike!) and her entourage in period costume with music and an order of service appropriate to that age. We really enjoyed the day.

Whilst walking round St John’s Chapel my wife noticed a plaque on the wall in memory of Frederick Blackman, the plant physiologist who died in 1947. This brought back memories of university lectures. Blackman was the guy who worked out the details of gas exchange in plants during photosynthesis - during the day it was carbon dioxide in and oxygen out etc.

Blackman may not even have considered that his work would be relevant to the debate about climate change. For instance, a recent German report says that some species of trees are now growing very significantly faster than in the past due to increased carbon dioxide levels in the air.

Around ten or so years ago crop physiologists at Rothamsted Research proved that in growth chambers (where one factor affecting growth can be changed) an increased concentration of carbon dioxide in the air would increase cereal yields. We’ve not seen this clearly demonstrated in real life because of the complexities of growth and of identifying the precise role of individual factors in a particular season.

Blackman recognised this problem and put forward the law of limiting factors in 1905. This proposed that when a process depends on a number of factors, its rate is limited by the pace of the slowest factor. This is why it’s so difficult to predict yields of most crops in the UK. Solar radiation is a major driver of photosynthesis but plants under moisture stress will not photosynthesise at the potential rate set on a very sunny day. Sunny summers tend to be dry summers and dull summers tend to be wet summers.Wheat

We have only to go back to 2007 and 2012 to see that in wet summers there is insufficient sunshine to support even average wheat yields. This year has been sunnier than average, and there have been very good yields where the thunder storms provided sufficient moisture during grain fill.

However, it’s not even as simple as that. Warm nights mean that the rate of respiration is increased taking the edge off the photosynthetic gains during the day. In addition, extremely hot days, particularly during flowering, can reduce pollen viability (hence reducing grain numbers) as well as encourage early leaf senescence and reduce chlorophyll content of leaves leading to a permanent reduction in the yield potential of the crop. Warmer weather than average reduces the time to harvest and hence reduces the number of days when the crop can put on weight.

Perhaps the most recent period of ideal weather for wheat yields occurred during the first week of June last year. At that time the crop was not under moisture stress and it was very sunny but relatively cool during the day and cold at night.

So the rate of yield accumulation, as Blackman hypothesised, may well be limited to the pace of one particular factor but the limiting factor may continually change during the growing season. Trying to sort out the precise role of increased carbon dioxide levels amongst all this is impossible. What we really need is sun but not heat and a good supply but not an oversupply of moisture. No wonder farmers are always complaining about the weather; it is rarely perfect.

Leave a comment / View comments


Giant slugs and giant problems

Posted on 23/09/2014 by Jim Orson

I’m sure I heard it first on BBCR4’s Today programme but I can’t find a record of it. Then I met a person outside our house who confirmed it. In fact, he said that his dad had mistaken one for a large stick when taking his dog for a walk late one evening. What am I talking about? Garden slugs: they are ENORMOUS this year.Giant slugs

Over the last couple of weeks I have upped my anti against slugs in the garden. This has meant rather than just relying on metaldehyde pellets, I’ve roamed the lawn in the dark picking slugs and transporting them to a large open area for the birds to eat the next morning. It’s Integrated Pest Management and encouraging bird numbers all rolled into one.

I hope that my metaldehyde use is not contaminating water. The garden is not underdrained and slopes away from any hard surfaces. However, metaldehyde in water has been an issue ever since the river companies developed a method of measuring it with sufficient accuracy to say when levels in water exceed the Drinking Water Directive limit of 0.1 parts per billion (ppb).

The problem with metaldehyde is that the processes installed at waterworks to remove small Directive exceedances of nearly all pesticides don’t do much at all in reducing the levels of metaldehyde. It’s neither strongly adsorbed by the activated carbon beds nor oxidised by the ozone treatment. Hence, unlike most pesticides, the raw water going into the treatment works has to comply with the Directive.

The very active Metaldehyde Stewardship Group has for the past few years been providing guidance in its Pelletwise campaign in order to reduce metaldehyde movement to water. This autumn’s guidance document reflects the increasing pressure on the future use of this pesticide. Quite rightly the guidance has adopted a targeted and conditional approach. There is little doubt in my mind that adopting best practice, particularly in identified high risk fields, will significantly reduce pesticide movement to water.Get Pelletwise

It comes as a surprise to most people that pesticide movement to water through the soil is not influenced by the solubility of the pesticide. Glyphosate is one of the most soluble pesticides available but does not move through the soil to watercourses. Its presence in water is largely determined by spray drift and its wide usage on hard surfaces.

The ability of pesticides to move through the soil to water is mainly determined by how strongly they are adsorbed by the soil (glyphosate is strongly adsorbed by the soil) and their persistence in the soil. There is a relatively simple calculation based on these two measurements which indicates the level of risk of movement through the soil. It’s called the Groundwater Ubiquity Score (GUS). Despite being relatively simple it’s surprisingly accurate, but of course more precise models are used in pesticide registration. It was developed by Gustafson (another reason for the abbreviation GUS), a Monsanto scientist who published the approach in 1989.

The GUS score for metaldehyde suggests that (marginally) it should not move through the soil in sufficient quantities to cause a problem. There are perhaps three possible explanations as to why it is occurring in water.

Firstly, that it is moving to water through different routes. There is little doubt that run-off from fields and the washings from contaminated equipment are a source of metaldehyde in watercourses. However, I understand that movement through soil has been confirmed as a route by the Metaldehyde Stewardship Group.

Secondly, this is an exception that proves the rule for the GUS score. This is a possibility but somehow I doubt it.

Finally, there may be something about the use of metaldehyde that the GUS score does not take into account. This may be the explanation. The GUS score assumes that the pesticide is applied directly to the soil but this is not the case with metaldehyde. It’s necessarily applied in pellet form and so is not immediately adsorbed by the soil. This may influence how long it persists in the soil environment and possibly undermines the methodology used in calculating its GUS score.

I must dash as it is now dark and I must continue with my slug patrol.

Leave a comment / View comments


Page: ‹ First  < 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 >  Last ›