Study of ramial chip mulch and organic fertilizers on wild blueberries

 In 2014, we laid out and established the research plots, and performed all the various treatments for each.  Data collection began in 2015, with the first of our two harvests for the study.

 In late April 2014 I began setting up and laying out our sixteen test plots. Each one was 12’x50′, and I left a 4′ alley between each plot to minimize spillover or “drift” of mulch or fertilizer from plot to plot.  The treatment plots were laid out at random, so neighboring plots vary from each other across the entire study area, which totals about 1/2 acre.  I originally used small plastic flags on fiberglass posts to mark the boundary corner of each plot.

Prior to laying out and establishing the plots, most of the study area had been mowed (a standard pruning technique), while the rest of the five acre field was burned (the alternative pruning technique).  Our study originally proposed having the test plots all in the similarly managed area of the field, to reduce the number of variables that could affect our results.  Unfortunately, the 16 test plots could not readily fit within the mowed area of the field, so I established four of the plots in an area that was burned (one replication of each treatment).  The different pruning techniques may or may not alter the plant growth and yield –but either way, I have clearly noted on the test map of the plots which ones had been mowed and which ones had been burned.

Over a four week period in May, I cut saplings (predominantly gray birch) on our farm from areas where they’d become densely established over the past 20 years, thereby opening up former field space.  Most of the  saplings were smaller than 3″ in diameter, not yet leafed-out, and everything up through the tiniest of branches was chipped, resulting in what is known as “ramial chips.”

The ramial chips were then spread to a depth of 2″ to 4″ on twelve of the 16 test plots (our control treatment in the study, replicated across four plots, has no mulch or fertilizer applied).  In low spots and hummocky areas, I often allowed the mulch to accumulate to depths up to 8″, thereby helping to level out the surface of the field, a condition which helps make future harvesting and pruning work more efficient.

Just prior to spreading the mulch on eight of the plots (the ones where I had planned the replications of two different treatments of fertilizer applications), I spread the appropriate amounts of fishmeal and aragonite.

One treatment over four plots had a rate of 12 lbs./plot (or the equivalent of 70 lbs./acre of N) of the fishmeal and 2 lbs./plot (or 150 lbs./acre of Ca) of the aragonite.  The second treatment over four plots had half those rates (6 lbs./plot of fishmeal or 35 lbs./acre of N and 1 lb./plot of aragonite or 75 lbs./acre of Ca).  The fertilizer was weighed in a hanging scale, and spread evenly over the plots by hand.  Then, the ramial chips were immediately spread over the top of the fertilizer.

I found the work of producing and spreading the ramial chip mulch to be accomplished in less time than I had estimated.  Although it is time-consuming and difficult work, I am happy in light of developing enterprise budgets for this study that the labor costs could be lower than originally thought.

By the time I was chipping and spreading the ramial chip mulch on the last few test plots, the gray birch saplings had begun to leaf-out, which could alter, among other things, the C:N ratio of the ramial chips, and therefore change their chemical, biological and/or physical attributes in the study.  This may or may not have had an affect on the results of our project, but either way, I noted these plots on the test map of our field.

Our original proposal included gathering baseline data during the first summer of the study (2014), including soil test analyses, foliar nutrient analyses, and Brix readings of foliage from each plot.  Due to illness and an extremely busy farm season (it was a banner year for the wild blueberry harvest), I was unable to perform these tests.  This is unfortunate, as it will not be possible to compare the results of later tests with the starting data.  However, I did have some data (soil and foliar analysis) from this field from 2010, which I will be able to use, as well as the harvest data from the field going back to 1998.  Our final analyses and outcome for the project is primarily comparative, treatment to treatment, so the baseline data, though important and interesting, is not imperative; all of the plots are in the same limited area within a field that has been managed in the same manner for almost 20 years.  I do plan on performing all of the tests on all of the plots from now on, as originally proposed.

Finally, for 2014, I realized by late summer that the marking flags I had chosen were not going to be sufficient for a four year study.  Some of the flags had already torn, and most had faded dramatically in the sun; there were also several that had blown over or been lifted out.  So, I purchased some iron rod rebar, cut it to 2′ lengths, spray painted these new stakes in four different colors (to mark and correspond with the four different treatments), and installed them in replacement of the original corner boundary markers.  These stakes are more permanent for our study, and give instant identification as to which treatment one is looking at for each plot (as they are now noted on the test map).

2015 saw the first harvest year in our study allowing us to begin the gathering and analyzing of data.  

We successfully gathered useful data from the first of the two harvest cycles that this study encompasses (the second harvest cycle will be in 2017).  The data includes harvest yields from each test plot, which have then been analyzed, as well as Brix readings from the ripe fruit taken from each plot.

In mid-May, we set four hives of honey bees on the field for a period of several weeks while the blueberry plants were in bloom, and had good weather plus notable healthy numbers of native/wild pollinators to ensure good pollination for the year.  Just prior to harvesting, in late July, we strung the boundaries of each of the sixteen plots (with white string) to clearly mark the edges of each.  On August 3rd, we harvested the blueberries from all of the alleys between and surrounding the plots; then on August 4th and 5th, we harvested the plots.  The berries from each plot were harvested into clearly marked field totes, which were then gathered together and brought to our winnower shed.  The field totes from each individual plot were then processed through our fresh-pack line (i.e., winnower), and the total weight of processed berries from each plot then measured and recorded.  At this time, a small sample was taken from each plots’ winnowed berries, which were then placed in labeled zip-lock bags in our cooler; several days later, we used our refractometer to measure and record the Brix level of each sample.  All of these numbers have been gathered and analyzed in several matrixes, and now serve as the foundation set of data which will be compared and contrasted to our pre-test baseline data and the second set of test data yet to be collected.

Due primarily to continued personal illness (Lyme disease), we were not able to perform all of the data gathering from our original proposal this year.  Namely, we did not collect weekly Brix readings of the fruit for up to five weeks (from early ripening through harvest), but rather only collected the one Brix reading at harvest.  Also, we were not able to collect Brix readings from the plant tissue from each plot.  

In 2016, after a successful prune (by fire), we successfully gathered useful data from the second of the prune cycles that this study encompasses (the first prune cycle was in 2014).  The data includes wild blueberry foliar analyses from each of our 16 test plots, as well as soil samples of each plot that were analyzed by two separate soil testing labs.

In March, we spread straw over the entire field (about 4 acres in total), spreading it evenly at a rate of 40 bales/acre.  Then on the first of May, we burned the field, setting the straw alight with a drip torch and containing the burn with a small crew using backpack water-pumps.  It was a good, even burn, which resulted in a very vigorous and healthy sprouting and regrowth over the summer.  It was a rather dry summer, with less than average rainfall, which may impact the size and number of buds set for next year’s harvest —we will just have to wait and see, and compare next summer’s yield data with this sort of natural phenomenon in mind.

It was quite apparent that the ramial chip mulch, that had been spread on most of our test plots (except the control plots) back at the beginning of the project in 2014, had begun to break down and disintegrate into the soil;  we are hopeful that this continuing decay means that the mulch will no longer hinder the physical act of harvesting next summer's crop (as it did in 2014) and that perhaps it will not be tying up as many nutrients as one expects it does in the earlier stages of decomposition.

On July 12, when the new sprouts were in the “tip die-back” stage, we cut about 100 stems of the wild blueberry plants from each plot, as required by the Univ. of Maine testing service, which were then bagged and labeled and shipped to the lab.  The results (a nutrient analysis of the plants, which is used by wild blueberry growers for recommended soil amendments and fertilizing rates) were received in November.

In late September, we used our soil core sampling tool, and took two boxes of soil samples from each plot.  These boxes were labeled and then shipped (one sample from each plot) to two separate soil testing labs —one at the Univ. of Maine, the other one at Perry Agricultural Lab in Bowling Green, MO.  As we proposed in our original grant application, the standard soil test (Univ. of Maine) will give accurate pictures of the total nutrients present in the soil, while a Ream’s-type test (Perry Agricultural Lab) will give more a picture of what amounts of nutrients are readily available for the plant roots to take up, and checks both ammonia and nitrate nitrogens.  The results of these tests, including soil pH, Cation Exchange Capacity (CEC), and soil nutrient levels, were both received in early November.

All of this data has been gathered and analyzed, and now serve as the second set of data of foliar and soil nutrient levels which will be compared and contrasted to our pre-test baseline data (to be presented in our Final Report, next year).

In 2016, we purchased a stainless steel hydraulic press, needed to extract juice from the wild blueberry plants for refractometer readings of their Brix level.  Due to the manufacturer’s (Pike Agri-Lab) delay in needing to construct and assemble the unit, we did not receive it until late summer —too late to do the weekly sampling and testing that we had intended and wanted to do from our proposal.

In 2017,  we had our second harvest for this project, and collected data on the lbs. of organic wild blueberries harvested under each treatment, and the Brix levels of the ripe berries from each treatment.