sign in sign up
<<
Home , Permaculture, Polyculture

Mixed Vegetable www..org.uk Trials the results

In association with ‘Mixed veg’ can mean so much more than a quick-fix packet from the freezer. For 50 people around the UK last year, it took on a new meaning – a new way of growing vegetables all by Dr Naomi van mixed up together. der Velden Rationale - Learning Photo: C. Atkin from nature? Look around at the natural world to see where choose to grow. You will notice that they tend to grow in mixes of different plants rather than large areas of a just one species. Often species grow together in relatively consistent and discernible mixes. For example, we might expect to find bramble and bluebells in lowland oak woodlands. Plants naturally form communities. There is some good scientific evidence that communities of wild plants growing together may be more productive than any one of the component species growing on its own would be 1. Can growing food plants in communities, Figure 1 - A trial layout High-diversity plot is in the foreground, and behind it the low-diversity plot. rather than , also be more productive (as well as better for the together compared to the same species Studies have found that some species are environment)? grown separately. You might have heard more productive when grown in mixes, People have studied production of of the classic “three sisters” of Central but that others are not. However, very plants when several different types America – , and squash – few studies have looked at mixes of more (usually two or three species) are grown which are said to grow well together. than three crop species, so we thought we’d give it a go.

1 “16-species plots attain 2.7 times greater than monocultures” from: Tilman et al. 2001 Diversity and Productivity in a Long-Term Grassland Experiment. Science 294 (5543) 843-845. Mixed vegetable

We conducted our own study to compare low- and high-diversity mixes to find out: Are more diverse (12 species) vegetable mixes more productive than less-diverse (three species) mixes? How easy do gardeners find these systems to use? The vegetables used are as listed below and in Figure 1. We chose common vegetables from a range of families and which would complement the spaces available (e.g. some roots, some climbers, some low, some high) above and below ground. How did we investigate this? Fifty sets of seeds were kindly provided by four lovely seed companies and sent to households around the country. Figure 2 - Plant roots and shoots occupy different areas , lettuce, sweet corn, peas, radish, & runner beans. Everyone made two plots – one smaller (Roots drawn after Weaver & Bruner, 1927 “Root development of vegetable ”) one for three species - Low diversity - For example, different plants have and one larger one for the more diverse Some plant : different rooting zones so could be mix - High diversity (see Figure 1). Plant niches, competition and combined to maximise use of the Larger plants (beans, sweet corn, kale) co-operation. area, rather than competing in a single were first grown in pots then planted layer (see above). The growing of plants in mixtures out. Others (, peas) were sown in is effectively a balance between These plants might be described as clusters, and the remainder were mixed competition for resources, and occupying different niches. together with some soil and scattered “co-operation” to use all available across the plot. They can also benefit each other, resources as efficiently as possible. for example by fixing nitrogen or Participants were asked to record providing food for pest predators or information about when they visited for pollinators. their plots, how long they spent there, and also to weigh each species each time it was harvested. We measured productivity of the edible plant parts (the bits you’d buy) by What have we found Twenty four participants were able to looking at: out? complete the study and return their data. Weight harvested per area (square-metres) Which plot-type was most productive? Weight harvested per minute of time spent In terms of the total amount produced on the plot per area, the high diversity plot was What’s in the plots? slightly more productive (see top graph Length of time plots were productive for in Figure 3). (longevity of production) Low diversity: On average, 3.1 kg were produced from Participants were also asked to fill in Peas, radish and perpetual spinach a square metre (m2) in the low diversity (beet). questionnaires before and after so we plot and 3.5 kg in the high diversity could find out what they thought of plot. These slight differences are not growing in this way. High diversity: significant2 because there is a lot of variation in how much was harvested by As above, plus onions, lettuce, Thirty-one people returned the initial different people (see maps in Figure 4). beetroot, sweet corn, coriander, questionnaire and twenty-one the final runner beans, kale, marigolds and questionnaire of whom nineteen had When considered by the amount of time rocket. also returned data. people put into the plots, then the low diversity plot was more productive for the effort (see lower graph in Fig. 3).

2 Wilcoxon signed rank test for 24 samples, z = 1.154, p = 0.130 . The difference between means is considered statistically significant if the value of p is less than 0.05, so in this case, this is not significant. The error bars in the graphs also overlap each other quite a lot which is always a clue that the difference in the average isn’t too important. The results

On average, 3.4 kg were harvested from a square-metre of the low diversity plots Yield per square metre The vertical black lines (“error bars”) for every hour of effort put in, and the show standard error and indicate the high diversity plots yielded 2.3 kg. amount of variability in the data. Although this difference is quite large, Small lines mean that most people it is not quite statistically significant3. It harvested close to the average should be noted that people tended to amount, larger lines mean that some

record time to the nearest five minutes, p r o d u c t i v y ( k g m - 2 ) people got lots and some people so the amount of time spent on the got a little. low diversity high diversity low-diversity plot might have been over- estimated. Yield per hour Overall, there’s little difference in productivity between the two types of plot when we account for area and time. This could be because of the crops chosen or other factors that mean there

is a lot of variation in the data. p r o d u c t i v y ( k g e h ) You can see in the maps in Figure 4 that low diversity high diversity some quite high levels of production Figure 3 - Productivity of low and high were reached – up to 10.6 kg of food diversity mixes when considered per area (top graph, kilograms produced from one square metre) and from one square metre of ground! in terms of how much effort was put into growing (lower graph, kilograms produced per hour of effort from one square There are lots of reasons why this might metre of land). One kilogram (kg) is 2.2 lbs. be – better , better weather, more time, more experience. Productivity of each plot. The amount of time put into plots was significantly related to the amount of yield people got in the high diversity plot4 and a little so in the low diversity plot 5. People who got a good yield on the high diversity plot also tended to get a good yield on the low diversity plot6, perhaps because their site was better or because they are more experienced gardeners. It is interesting to see from the maps in Figure 4 that areas typically considered more marginal, like Scotland, Wales, and the far north of England, generally had some quite high production. Perhaps success with different types of the crops used might also be important?

To give some context, 3.5 kg per metre-square works out at 35 tonnes per hectare! UK wheat yields are about 7-8 tonnes per hectare and that’s Figure 4 - Yellow shows lower production and 0 - 0.75 kg on the best growing land! red shows higher production in grams, harvested from one square-metre, the values are: 0.76 - 1.5 Mind you, it’d take quite a lot 1.6 - 3.0 of time to harvest a hectare 3.1 - 5.0 (that’s an area 100 by 100 metres) using this method! 5.1 - 10.6

3 Wilcoxon Z statistic = 1.680, p = 0.093, 20 people returned data on time spent on the plots. 4 Spearman’s Rank correlation test: Rho = 0.446, p = 0.046 The relationship is considered statistically significant if the value of p is less than 0.05. Rho is a value between +1.000 and -1.000. If it is +1, then there is a perfect straight line fit between increasing time and increasing yield. If it was -1, then increasing time would always lead to a reduced yield. 5 Rho = 0.430, p = 0.052. 6 Rho = 0.845, p < 0.001 (Here you can see Rho is closer to +1.000 meaning that there is a stronger relationship between yield from high diversity plots and low diversity plots, than there is between experience and yield. This is also reflected by the p value being much less than 0.05) Mixed vegetable polycultures

things that could cope with some shade Which crop types did (e.g. spinach beet, lettuce, peas, radish, Longevity of crop

best? coriander, rocket,6 beetroot), it seems production diversity plot also tended to get a good yield on the low diversity plot , perhaps because their site In thewas low-diversity better or because plot they – Peashave green and fingers. It is thatinteresting there to was see froma lot the of mapscompetition in Figure 4 in that the As well as looking at the total amounts spinachareas typically were successful considered morefor most marginal, people like Scotland,High-diversity Wales, and the plot. far north of England, produced, it’s also important to look at generally had some quite high production. Perhaps success with different types of the crops used (seemight left alsohand be sideimportant? of Figure 5) and gave The density of crops is very important when crops are being harvested. high yields (Figure 6). Radish grew for in determining yield – too far apart and In theory, having a greater diversity most people and produced reasonably Which crop types did best? crops are isolated and susceptible to the of different crops might mean more well (given it’s small!) In the low-diversity plot – Peas and spinach were successfulelements. for most people (see left hand side of continuous food production, and fewer Figure 5) and gave high yields (Figure 6). Radish grew for most people and produced reasonably In thewell high-diversity (given it’s small!) plot - Well, rocket Grow lots of one crop close together and “gluts” where we’re overwhelmed with a grew. Vigorously. This was meant to be the individual plants can’t obtain all the particular crop. a coverIn the crop high -thatdiversity we plotpulled - Well, up rocketand ate grew. Vigorously. This was meant to be a that resources they need. From our data, the High-diversity plot earlywe onpulled to giveup and room ate early for on other to give things… room for other things…but it’s so tempting to just cut a bit off and keep harvesting it. Kale, onions, and beans also harvested well for most people. has higher yields earlier in the season but it’s so tempting to just cut a bit off It is possible that our High-diversity crop mix was too dense to perform at giving a greater spread of when food is andSweet keep corn, harvesting beetroot, coriander,it. and lettuce really didn’t grow for many people (Figure 5) and even those who got a harvest didn’t get very much (see theits right best. hand If weside look of Figure at when 6). people were available (see Figure 7). Kale, onions, and beans also harvested harvesting crops, we also see some 7 8 9 Looking at individual species, we can wellFor for the most three people.species grown in both plots, spinach andinteresting peas and radishresults all that grew support much bet this.ter in the Low-diversity plots than the High-diversity plots (see Figure 6). also see some important differences. Figure 8 shows that peas were harvested for longer from the Low-diversity plot, and spinach was ready much earlier. The radish was generally ready around the same time in both. Spinach, lettuce, and beetroot in the High-diversity plot all started to increase towards the end of our growing season (last harvest on 31st October), suggesting that these all started to do well after some of the other crops (rocket, beans, onion) were harvested (Figure 9). Figure 5 Number of people who harvested each crop (max 24). Organised by most successful Figure 5 - Number of people who harvested each crop (max 24). crops in Low- then High-diversity plots. Organised by most successful crops in Low- then High-diversity plots.

Sweet corn, beetroot, coriander, and lettuce really didn’t grow for many

people6 Rho = (Figure 0.845, p < 5) 0.001 and (Here even you canthose see Rho who is closer to +1.000 meaning that there is a stronger relationship between yield from high diversity plots and low diversity plots, than there is between experience and yield. This is also reflected gotby a the harvest p value being didn’t much get less verythan 0.05) much (see7 Spinach. the right Difference hand is highlyside statisticallyof Figure significant. 6). Wilcoxon signed-rank statistic 3.574, p < 0.0005. So there is a huge difference in the average amount of spinach people are getting from each plot and we can see that much more Forcomes the threefrom the speciesLow diversity grown plot. The in Wilcoxonboth signed-rank is a paired test which compares production from each person’s Low- and High-diversity plot, and so eliminates some of the variations that might occur from people growing in 7 8 9 plots,slightly spinach different ways.and peas and radish all 8 Peas. Wilcoxon signed-rank statistic 4.171, p < 0.0005. Difference is highly statistically significant. grew9 much better in the Low-diversity Radish. Wilcoxon signed-rank statistic 3.782, p < 0.0005. Again, the p-value is much, much less than 0.05. plots than the High-diversity plots (see Figure 6). Overall, more diverse plots seem to yield a little more for the land area than less diverse plots, although diversity of vegetables harvested might be at the

expense of yields of individual species. Figure 6 6 Average - Average productivity productivity of the of plants the plantson the two on theplots two Measured plots Measuredin kilograms in per metre- squkilogramsared, per organised metre-squared, by most organised productive by most productive crops in crops the Lowin the- Low- then then High High-diversity-diversity plots plots We can see that some crops grew well The vertical black lines show standard error and indicate the amount of variability in the data. Small lines mean but others, like sweetcorn, consistently that most people harvested close to the average amount (e.g. spinach in the High-diversity plot), larger lines meanThe thatvertical some peopleblack got lines lots andshow some standard people got a little(e.g. (e.g. spinach in thein theLow -High-diversitydiversity plot). plot), failed to produce. It is interesting to see error and indicate the amount of larger lines mean that some people got

that the most productive crop, spinach, variabilityOverall, more in diverse the data. plots seem to yield a little morelots for theand land some area thanpeople less diversegot a littleplots, (e.g. did well only on the Low-diversity plot. although diversity of vegetables harvested might be atspinach the expense in theof yields Low-diversity of individual plot).species. SmallWe can lines see that mean some that crops most grew wellpeople but others, like sweetcorn, consistently failed to produce. It is Most crops need a certain amount of harvestedinteresting to closesee that to the the most average productive amount crop, spinach, did well only on the Low-diversity plot. sunshine, and although we tried to pick Most crops need a certain amount of sunshine, and although we tried to pick things that could cope with some shade (e.g. spinach beet, lettuce, peas, radish, coriander, rocket, beetroot), it seems that 7 Spinach. Difference is highly statistically significant. Wilcoxon signed-rankthere statisticwas a lot3.574, of p competition < 0.0005. So there in the is aHigh huge- differencediversity in plot. the average amount of spinach people are getting from each plot and we can see that much more comes from the Low diversity plot. The Wilcoxon signed-rank is a paired test which compares production from each person’s Low- and High-diversity plot, and so eliminates some of the variations that might occur from peopleThe growing density in slightly of crops different is very ways. important 8 Peas. Wilcoxon in determining signed-rank statisticyield – 4.171, too farp < 0.0005.apart and Difference crops isare highly isolated statistically and 9 significant. Radish. Wilcoxon signed-rank statistic 3.782, p < 0.0005. susceptibleAgain, the p-value to the is much, elements. much less Grow than lots 0.05. of one crop close together and the individual plants can’t obtain all the resources they need. It is possible that our High-diversity crop mix was too dense to perform at its best. If we look at when people were harvesting crops, we also see some interesting results that support this. Longevity of crop production As well as looking at the total amounts produced, it’s also important to look at when crops are being harvested. In theory, having a greater diversity of different crops might mean more continuous food production, and fewer “gluts” where we’re overwhelmed with a particular crop.

From our data, the High- diversity plot has higher yields earlier in the season giving a Figure 7 Continuity of production. Amount harvested each month. Solid line shows High-diversity plot, dashed line the Low-diversity.

Figure 6 Average productivity of the plants on the two plots Measured in kilograms per metre- squared, organised by most productive crops in the Low- then High-diversity plots The vertical black lines show standard error and indicate the amount of variability in the data. Small lines mean that most people harvested close to the average amount (e.g. spinach in the High-diversity plot), larger lines mean that some people got lots and some people got a little (e.g. spinach in the Low-diversity plot).

Overall, more diverse plots seem to yield a little more for the land area than less diverse plots, although diversity of vegetables harvested might be at the expense of yields of individual species. We can see that some crops grew well but others, like sweetcorn, consistently failed to produce. It is interesting to see that the most productive crop, spinach, did well only on the Low-diversitygreater plot. spread ofgreater when spread food of whenis available food is available (see Figure (see Figure 7). 7 ). Most crops need a certain amount of sunshine, and although we tried to pick things that could cope with some shade (e.g. spinach beet, lettuce, peas, radish, coriander, rocket, beetroot), it seemsLooking that at individualLooking atspecies, individual we species, can also we can see also some see some important important differences. differences. Figure Figure 8 shows 8 shows that peas that peas there was a lot of competition in the High-diversity plot. were harvested for longer from the Low-diversity plot, and spinach was ready much earlier. The The results were harvestedradish for longer was generally from readythe Low around-diversity the same time plot, in andboth. spinach Spinach, lettuce,was ready and beetroot much in earlier. the High -The The density of crops is very important in determining yield – too far apart and crops are isolatedradish and was generallydiversity ready plot all around started to the increase same towards time thein both.end of o Spinach,ur growing lettuce,season (last and harvest beetroot on 31st in the High- susceptible to the elements. Grow lots of one crop close together and the individual plantsdiversity can’t plot allOctober), started suggesting to increase that thesetowards all started the toend do wellof o afterur growing some of the season other crops (last (rocket, harvest beans, on 31st obtain all the resources they need. It is possible that our High-diversity crop mix was too dense to October), suggestingonion) were that harvested these all (Figure started 9). to do well after some of the other crops (rocket, beans, perform at its best. If we look at when people were harvesting crops, we also see some interesting results that support this. onion) were harvested (Figure 9).

Longevity of crop production As well as looking at the total amounts produced, it’s also important to look at when crops are being harvested. In theory, having a greater diversity of different crops might mean more continuous food production, and fewer “gluts” where we’re overwhelmed with a particular crop.

From our data, the High- diversity plot has higher yields earlier in the season giving a Figure 7 Continuity of production. Amount harvested each month.

SolidFigure line shows 7 High - Continuity-diversity plot, dashed of production line the Low-diversity. Amount harvested each month. Solid line shows High-diversity plot, dashed line the Low-diversity. What about the quality of the food produced?

Overall, the High-diversity plot had a What about the qualityMost of the of foodwhat produced?was harvested was of high quality for more sustained availability of crops, Most of what was harvestedgarden was produce of high with quality relatively for little damage. Peas did Figure 8 Timing of harvests for crops in both plots. Solid Figure 8 - Timinggarden of harvests produce for with crops relatively well, spinach little damage reasonably. Peas well, did and radish fared rather with different crops peaking at different line shows High-diversity plot, dashed line the Low- diversity.in both Amount plots harvested Solidwell, line each showsspinach month High-diversity in reasonably grams from lessone well, well. and ( Figureradish 10fared). Therather low -diversity plot tended to times. squareplot, -metre. dashed line the Low-diversity.less well. (Figure 10). Thehave low slightly-diversity better plot quality tended food, to but the overall

It’s also likely that several crops only Amount harvested eachhave month slightly in grams frombetter one quality differences food, but between the overall the plots are not significant10. square-metre. 10 really started to grow after other crops Overall, the High-divedifferencesrsity plot had between a more sustained theIdeally, plots are we not would significant like to compare. these results to those availability of crops,Ideally, with different we wouldcrops peaking like atto compare these results to those (like rocket) were removed. different times. It’s also likely that several crops only from own-grown crops under “normal” growing Figure 8 Timing of reallyharvests started for to grow cropsfrom after in ownother both -cropsgrown plots. (like cropsrocket)Solid methods.wereunder “normal” growing It is important to understand howline shows these High-diversityremoved. plot, It is important dashedmethods. to line understand the Low how- these systems can work to provide a sustained yield of food. systems can work to provide adiversity. sustained Amount harvested each month in grams from oneLooking at damage to the crops shown in Figure 11, square-metre. Perhaps this is a good way of getting lots of food but yield of food. avoiding gluts. HoweverLooking if you at really damage want lots to of the peopl cropse felt shown that thein Figure Low diversity 11, plot suffered less What aboutspinach, the then quality givingpeopl it of plenty thee feltof food space that produced?to thegrow Low will be diversity plot suffered less Perhaps this is a good way of getting important. damage than the High-diversity plot and the usual way Most of what was harvesteddamage was than of high the Highquality-diversity for plot and the usual way lots of food but avoiding gluts. However in whichFigure they 9 Timing grew offood. harvests The for Highother crops-diversity plot Overall, the HighproduceIt should-dive withalsorsity be relativelyplot notedin had whichthat awe morelittle gavethey andamagesustained grew end date food.. for Peas perhapsour The did inHigh sufferedthe High-diversity-diversity a little plot plot. more Amount pest harvested and disease damage. if you really want lots of spinach, thenavailability of crops,experiment with and different askedperhaps everyone crops suffered to peakingharvest everythinga atlittle more pesteach monthand diseasein grams from damage. one square -metre. well, spinachthat was reasonably remaining at well,the end and of October. radish Some fared crops rather giving it plenty of space to grow willdifferent beless times.well. ( FigureIt’s also 10 likely). The that low several-diversity crops plotonly tended to really started towould grow have after kept other growing crops for longer. (like rocket) were important. have slightly better quality food, but the overall removed. It is important to understand how these differences between the plots are not significant10. It should also be noted that we gavesystems an can work to provide a sustained yield of food. PerhapsIdeally, this iswe a goodwould way like of to getting compare lots ofthese food results but to those end date for our experiment and askedfrom own-grown crops under “normal” growing everyone to harvest everything thatavoiding was gluts. However if you really want lots of spinach,methods. then giving it plenty of space to grow will be remaining at the end of October. important. Figure 9 - Timing of harvests for other Some crops would have kept growing forLooking at damage to the crops shown in Figure 11, Figure 9 Timing of harvests for other crops It shouldpeopl alsoe felt be thatnoted the that Low we diversitygave an end plot date suffered for our less in cropsthe High in the-diversity High-diversity plot. Amount plot harvested longer. Amount harvested each month in grams from one experimentdamage and than asked the everyoneHigh-diversity to harvest plot everything and the usual wayeach square-metre. month in grams from one square-metre. that wasin which remaining they atgrew the endfood. of October.The High Some-diversity crops plot wouldperhaps have kept suffered growing a little for longer. more pest and disease damage.

What about the quality of the food produced?

Most of what was harvested was of Figure 10 Quality of crops harvested. Figure 10 Quality of crops harvested. high quality for garden produce with Amount of damage in categories: Amount of damage in categories: relatively little damage. a, none – will eat all; b, mild – will eat most; a, none – will eat all; b, mild – will eat most; c, moderate – will eat half; d, severe – will Looking at damage to the crops shown c, moderate – will eat half; d, severe – will eat little; e, devastated – will eat none. in Figure 10, people felt that the Low eat little; e, devastated – will eat none. diversity plot suffered less damage than

the High-diversity plot and the usual way in which they grew food.

The High-diversity plot perhaps suffered Figure 11 Perceived damage on Low -diversity, Figure 10 - PerceivedFigure damage 11 Perceived on Low-diversity, damage on High-diversity Low-diversity, and a little more pest and disease damage. High-diversity and “Normal” growing methods. “Normal” growingHigh methods-diversity Note and that Normal“Normal” methods growing are the usual methods. growing methods of participants and differed between people. Answers basedNote on questionnairethat Normal responses methods for pests are – 24the usual growing The low-diversity plot tended to have Note that Normal methods are the usual growing responses, for disease - methods15 responses. of participants methodsand differed of participants Figurebetween 10 people. Quality and differed of crops between harvested. people. slightly better quality food, but the Answers based on questionnaireAnswers responsesbasedAmount on questionnaire for of pests damage – responsesin categories: for pests –

overall differences between the plots are 24 responses, for disease24 - responses,15 responses.a, for none disease – will - eat15 responses.all; b, mild – will eat most; not significant10. c, moderate – will eat half; d, severe – will

10 eat little; e, devastated2 – will eat none.2 2 10 Chi-squared tests for; Peas Χ Chi2 =- 3.89squared p>0.05; tests Spinachfor; Peas Χ Χ2 = = 4.60 3.89 p>0.05; p>0.05; Radish Spinach Χ2 Χ= 3.09 = 4.60 p>0.05. p>0.05; No Radish significant Χ = 3.09 p>0.05. No significant 10 Chi-squared tests for; Peas Χ2 = 3.89 p>0.05; Spinach Χ2 = 4.60 p>0.05; Radish Χ2 = 3.09 p>0.05. No significant differences between Low- and High-diversity plots. differences between Low- anddifferences High-diversity between plots. Low - and High-diversity plots.

Figure 11 Perceived damage on Low-diversity, High-diversity and “Normal” growing methods. Note that Normal methods are the usual growing methods of participants and differed between people. Answers based on questionnaire responses for pests – 24 responses, for disease - 15 responses.

10 Chi-squared tests for; Peas Χ2 = 3.89 p>0.05; Spinach Χ2 = 4.60 p>0.05; Radish Χ2 = 3.09 p>0.05. No significant differences between Low- and High-diversity plots.

How easy was it to grow? Most people who joined in and completed the initial survey did have some experience of growing veg before (just 1 beginner out of 31people), so we don’t really know what it would be like for complete beginners. It did seemMixed that people vegetable with more experiencepolycultures of vegetable growing got a little bit more out of the high diversity plot, but it didn’t make a difference for the low diversity one11, but neither of these is statistically significant, so if you grow veg you could give this method a go and get good results! Peas did well, spinach reasonably well, and radish fared rather less well (see How easy was it to How much effort is it? figure 11). grow? Was the yield worth the effort? How much effort is it? Most people who joined in and Overwhelmingly, people thought it was Was the yield worth the effort?Ideally, we would like to compare these results to those from own-grown crops completed the initial survey did have worth having a go, especially with the Overwhelmingly, people thoughtunder it “normal” was growing methods. some experience of growing veg before smaller Low-diversity plot (see pie charts worth having a go, especially with the (just one beginner out of 31 people), so in Figure 12). we don’t really know what it would be smaller Low-diversity plot (see pie charts Overall, people enjoyed trying out this What about the quality of the food produced? like for complete beginners. Most of what was harvested was ofin high Figure quality 12). for method of growing and were keen to garden produce with relatively little damage. Peas did It did seem that people with more give it a go (Figure 13). experience of vegetable growing got a well, spinach reasonably well, andOverall, radish fared people rather enjoyed trying out this But – be careful! little bit more out of the high diversity less well. (Figure 10). The low-diversitymethod plot of tendedgrowing to and were keen to give it a go (Figure 13). plot, but it didn’t make a difference for Although the results look good, they are have slightly better quality food, but the overall the low diversity one11, but neither of by no means the final answer. There are 10 differences between the plots are not significant . Figure 12 Was yieldthese worth is statistically the effort? significant, so if you some important things to consider: Ideally, we would like to compare Butthese – resultsbe careful! to those Reponses from finalgrow questionnaire veg you could (21 give responses) this method a Although the results look good, they are We’re really happy that 24 people from own-grown crops under “normal” growing go and get good results! completed the trial. However, it’s still methods. by no means the final answer. There are some important only a relatively small number of plots. things to consider: Looking at damage to the crops shown in FigureWe’re 11really, happy that 24 people completed the Any one grower can have a large influence on the overall results. people felt that the Low diversity plot sufferedtrial. less However, it’s still only a relatively small damage than the High-diversity plot and thenumber usual wayof plots. Any one grower can have a large We’d like to repeat it again with in which they grew food. The High-diversityinfluence plot on the overall results. We’d like to hundreds of people joining in to get a perhaps suffered a little more pest and diseaserepeat damage. it again with hundreds of people joining in real feel for what is happening! to get a real feel for what is happening! It was a strange year for weather and It wasHow a strange easy year was for weather it to grow?and many people many people agreed it was a difficult agreed it was a difficult growing season. Perhaps growing season. a differentMost yearpeople would who yield joined different in andresults completed the initial survey did have some experiencePerhaps a different of growing year would yield Differentveg beforecombinations (just 1and beginner types of vegetableout of 31people), so we don’t really know what itdifferent would results be like for speciescomplete might give beginners. very different It did results seem – there’s that people with more experience of vegetab Differentle growing combinations got a and little types still a lot which isn’t known about polycultures! of vegetable species might11 give very bit more out of the high diversity plot, but it didn’t make a difference for the low diversity one , but Figure 12 - Would you plant a mixed different results – there’s still a lot which neither of these is statistically significant, Figureso if vegetableyou 13 Would grow bed again?vegyou plantyou coulda mixed give isn’tthis known method about a polycultures! go and get good results! vegetable bed again? In summary,

This is a highly productive method of growing; on average 3.5 kg of food were produced from one square-metreHow and somemuch people effort achieved is it? over 10kg. and allotments have the potential to give us a largeWas amount the yield of food worth from athe small effort? amount of space. They also generally avoid many of the negative impacts on wildlife and the environment that might be associated with other methods of food production.Overwhelmingly, However, this is people a much thoughtmore labour it was-intensive method of production which people tend to do forworth enjoyment having in growinga go, especially as much as with for the the final product, tasty though that may be! smaller Low-diversity plot (see pie charts in Figure 12 ). 11 Figure 10 Quality of crops harvested. Spearman’s rank correlation on experienceFigure 11 and - Quality production of crops per harvested metre-square: High diversity plot Rho is 0.420, p = 0.051 (almost less than 0.05); AmountLow diversityAmount of ofdamage plot damage Rho in categories: isin 0.311, categories: p = 0.159 . Rho would be +1.000 if all the data Overall, peoplea, none a: enjoyed –none will – will eateat all; all; tryingb: mild b, – willmild eat out most; – will this eat most; formed a perfect straight line of increasingc: moderate experience – will eat half; and d: severe increasing – will eat little; yield, and -1.000 if increasing experience but decreasing yield.m ethod So this means of c,growing experiencemoderatee: devastated andcounts– will – will wereeat aeat littlenone half; bit,keen d,but severe people to –with will less experience can still get high yields. give it a goeat (Figure little; e, 13 devastated). – will eat none.

Figure 12 Was yield worth the effort? But – be careful! Figure 13 - Was yield worth the effort? ReponsesReponses from from final questionnaire final (21 questionnaire responses) (21 responses) Although the results look good, they are

Figure 11 Perceived damage on Low-diversity,by no means the final answer. There are some important High-diversity and “Normal” growing methods.things to consider: Note that Normal methods are the usual growing We’re really11 Spearman’s happy rank correlation that on experience 24 people and production completed per metre-square: High the diversity plot Rho is 0.420, p = 0.051 (almost less than 0.05); Low diversity plot Rho is 0.311, p = 0.159 . Rho methods of participants and differed between people. would be +1.000 if all the data formed a perfect straight line of increasing experience and increasing yield, and -1.000 if increasing experience but decreasing yield. So this means experience counts trial. However,a little bit, but people it’s with lessstill experience only can stilla getrelatively high yields. small Answers based on questionnaire responses for pests – number of plots. Any one grower can have a large 24 responses, for disease - 15 responses. influence on the overall results. We’d like to repeat it again with hundreds of people joining in 10 Chi-squared tests for; Peas Χ2 = 3.89 p>0.05; Spinach Χ2 = 4.60 p>0.05; Radish Χ2 = 3.09 p>0.05. No significant differences between Low- and High-diversity plots. to get a real feel for what is happening! It was a strange year for weather and many people agreed it was a difficult growing season. Perhaps a different year would yield different results Different combinations and types of vegetable species might give very different results – there’s still a lot which isn’t known about polycultures!

Figure 13 Would you plant a mixed vegetable bed again? In summary, This is a highly productive method of growing; on average 3.5 kg of food were produced from one square-metre and some people achieved over 10kg. Gardens and allotments have the potential to give us a large amount of food from a small amount of space. They also generally avoid many of the negative impacts on wildlife and the environment that might be associated with other methods of food production. However, this is a much more labour-intensive method of production which people tend to do for enjoyment in growing as much as for the final product, tasty though that may be!

11 Spearman’s rank correlation on experience and production per metre-square: High diversity plot Rho is 0.420, p = 0.051 (almost less than 0.05); Low diversity plot Rho is 0.311, p = 0.159 . Rho would be +1.000 if all the data formed a perfect straight line of increasing experience and increasing yield, and -1.000 if increasing experience but decreasing yield. So this means experience counts a little bit, but people with less experience can still get high yields. The results

In summary, This is a highly productive method of growing. On average 3.5 kg of food were produced from one square-metre and some people achieved over 10kg. Gardens and allotments have the potential to give us a large amount of food from a small amount of space. They also generally avoid many of the negative impacts on wildlife and the environment that might be associated with other methods of food production. However, this is a much more labour- intensive method of production which people tend to do for enjoyment in growing as much as for the final product, High-diversity plot in June tasty though that may be! Comparing the plots, the High-diversity from competition for space and nutrients Some of the seed onions have also crop was slightly more productive, and and less able to cope with additional started to grow well this year. pressures. gave a longer period for harvesting food; Perhaps we need to consider these however, the crops grown in the Low- Getting a balance between different food systems in terms of year-long diversity plot were individually more plants can be important for many production, or continuous production? productive, perhaps because they had reasons. Maddy Harland’s article on the less competition with other species. People enjoyed having a go at this, and Permaculture Magazine website There is a balance between growing a most were inspired to do it again. (22nd May 2012), talks about choosing variety of different food and growing a perennial varieties or annuals which Generally, they found the Low-diversity large amount of any one crop. self-seed, to create a naturalistic self- plot easier to manage, perhaps because sustaining polyculture. It is also likely that the more time you put it was smaller, perhaps because the mix into your garden the more it will reward of species was less confusing, or perhaps There are many ideas here which you you with food. because it wasn’t overwhelmed with might like to explore further. This Having a mix of some species may also rocket early on! summer, several of us are working on different species mixes so that we are reduce damage from pests and diseases, It’s June now, and I am still harvesting prepared for our future trials. but it is not clear how much diversity is spinach from both plots, even though I needed. cut them right back in October. If you have suggestions of crop mixes that have worked for you, we would love Our results suggest that the High- I’m sure the kale would have kept going to hear from you! diversity plot suffered most damage. It through the winter too but I honestly may be that plants were more stressed couldn’t face any more so pulled it up!

The polyculture research team Acknowledgements Dr Naomi van der Velden - Plant ecologist at the Massive thanks to Celia Ashman for all her help in co- University of Cumbria ordinating the trials and all the data collection, and Edgars Andy Goldring, Tomas Remiarz, Roz Brown, Dr Ian Fitzpatrick - our Latvian hard-working student placement. - the mixed veg team Big shout out to Garden Organic, Edwin Tucker & Son Ltd., Please note that we have a research co-ordinator in place - Beans and , and Chase Organics for supplying the Chris Warburton-Brown. seeds for this trial. Send any enquiries to [email protected] Thanks to Chris Evans for advice on appropriate plant mixes. To everyone who participated, and especially those who completed the trials, many, many thanks – we couldn’t have done it without you!