Thursday, September 17, 2015

How Winter Affects Roses (Revisited)

This is the time of year when everyone in the cold zones should start thinking about winter-protecting their roses.  The sooner you get going, the easier the task will be in a couple of months.  I think this blog post, written in the winter of 2013/14, is worth revisiting for those of my readers who may have missed it. I would particularly call your attention to my articles referenced herein about the annual postassium feast, which should begin in mid-September, in time to "harden off" your roses for winter.

 At the peak of winter here in the cold zones, our roses are "winter-protected" to help them survive the sub-zero temperatures of USDA zones 4, 5, and 6.  Most folks look out at their roses covered (hopefully) with a nice layer of snow and believe their roses are dormant; just waiting to thaw out, break dormancy and start growing again.  But wait....  Did you know that only species roses, such as Rosa Rugosa, Rosa Glauca, Rosa Gallica etc. go through a dormancy cycle and that all modern, repeat-blooming, "remontant" roses do not?  So what's going on here with modern roses in winter?

Before I try to answer that question, I want to say that I recently learned much of this from "Dormancy in Roses", an excellent four-part series in the American Rose, during 2013 and early 2014, by Dr. Gary Ritchie of Olympia, Washington (see footnote below).  I will quote Gary several times in this post and want to give him full credit for his research and opinions.  However, I also want to note that Gary's articles have raised some important issues for me, based on my many years of successfully growing modern roses in Minnesota; in particular, why keeping modern roses frozen hard in the winter is what keeps them alive, rather than killing them outright. This seems somewhat contrary to the conclusion of Part 4 of Gary's article, where he says:

"I've not seen data on specific cold hardiness of modern roses but experience indicates that it is modest at best. So, while we enjoy continuous bloom throughout the summer, we face the annual chore of winter protecting our roses.  Here in the moderate coastal Northwest, this requires no more than mounding up our plants in fall.  But in more extreme climates winter protection can be much more difficult and problematic -- sometimes even requiring burying the plants underground to assure their over-winter survival." 

Here is how I would re-phrase Gary’s quote (above) from my perspective in zone 4/5:

"I've not seen data on specific cold hardiness of modern roses, but experience indicates that, with good winter protection, most modern roses, including budded hybrid-teas, are very cold-hardy, as long as they are allowed to freeze solid and stay frozen all winter.  Here in Minnesota (zones 3, 4, and 5), winter protection begins with planting bud unions four to six inches deep, mounding with dirt or compost in the fall, and subsequently winter-protecting with leaves or hay after the ground freezes in late fall or early winter. Another alternative is the Minnesota Tip method of burying plants underground.  Both methods have as their objective keeping roses frozen throughout the winter; not to keep them from freezing, which is virtually impossible in our zone 4/5 winters."  (JRF Quote)

In other words, the whole purpose of winter-protecting roses in the cold zones, where the ground freezes from several inches to more than a foot down, is to keep roses from repeatedly freezing and thawing. The only exception to this might be the use of insulated R7.5 construction blankets, which are gaining popularity in Minnesota.  My friend and TCRC mainstay, Deb Keiser, who manages the Virginia Clemens Rose Garden in St. Cloud, believes that putting construction blankets down before the ground freezes keeps her roses from freezing in the first place (which is quite an achievement in St. Cloud!). But the principle is the same, whether the ground freezes under the blankets or not:  i.e., to keep your roses from repeatedly freezing and thawing.  This can be problematic here in the Twin Cities (now in zone 5) and even more so in zones 6 and 7, where mid-winter thaws are more frequent.  Take a look at my recent article on winter protection:

Here is how two of my rose beds looked on Christmas Day 2013:

Buck Earth Songs under a foot of snow insulation

Terraced Canadians and Bucks

Now, just in case I have given the impression that I'm not growing hybrid teas in Minnesota winters, here is my winter-protected Elina on a -2 F. afternoon in Edina.  The reason the leaf bag is showing under the snow is that we had a record-breaking 48 F. the day before I took this picture; a 50 degree swing!  And that's what winter protection is all about in zone 4/5: to keep the roses from thawing and re-freezing in these crazy temperature swings!

Above: Elina in a Minnesota Winter

Dr. Gary Ritchie's point about modern roses not going into dormancy is obviously correct. Unlike woody perennials like Rhododendron or lilacs, roses apparently do not have a dormancy "chilling requirement"  in order to generate next season's bloom cycle. Rather, as Gary says, modern roses, as remontant, repeat-blooming  plants, "by their very nature, fail to go dormant in winter. So they have a much-reduced ability to cold harden."  In other words, rose canes die back in winter because they do not sufficiently "cold harden" and this die-back can only be controlled at the crown or bud union levels by proper winter-protection, as described above.  This affirms something that I have advocated for many years, i.e., repeated applications of potassium in the fall to "cold-harden" rose canes before the first hard freeze. My experience, over more than 20 years, is that hardening rose canes off with a potassium feast has the effect of significantly reducing cane die back in the winter.  Please see my several articles about the "potassium feast":

But something else seems to be happening here

Over the last several years, as the Twin Cities metro has moved solidly into zone 5, my observations indicate that modern roses may exhibit a characteristic, which may be related to the chilling requirement inherent in plants that experience dormancy in winter.

Even though our winters are warmer, in terms of extreme minimum temperatures (EMT), they seem to be just as long, or perhaps even longer in certain years, thereby keeping our roses frozen for a longer period of time. For example, our ground (and therefore our roses) stayed frozen into late April or early May in 2013, and we had snow on the ground into early May.  This is 2-4 weeks later than normal. What happened in May, once the ground thawed out, was that the roses had a very hard time getting started and there seemed to be more die-back than usual, even with shrubs that are zone 3 and 4 hardy. One of our husband-wife TCRC members,who have had good success over the years planting their hybrid tea and shrub roses with bud-unions and root crowns six inches below ground level, and using minimal winter protection above ground, lost a number of roses in 2013, even though the same method had worked perfectly in colder EMT winters.

In other words, with a 2013 EMT of -13 (well above the median for zone 5), our roses actually seemed like they had been through a much harder winter.  So it would appear that the length of time roses are frozen, not just the low temperature in a given year, impacts survivability.  After all, if you think about it, frozen is frozen; the only thing that happens with a lower temperature is that the ground freezes deeper and the roses take longer to thaw out and start growing in the new season.  But what happens to them when the winter is so long that they can't start growing again in a timely way?  To my knowledge there is no scientific reasoning for this phenomenon.  However, I found a clue in Part III of Gary Ritchie's series, where he speaks of cold weather breaking dormancy in plants.  Speaking of dormant plants in the first person, he says:

 "...One way would be somehow to keep track of the amount of cold weather to which you had been exposed during winter.  After a certain number of hours or days of cold exposure had occurred you would have a clear indication that winter was finally over and it was safe to resume growth.  This is exactly what plants do...."

What he is saying is that dormant plants apparently have an internal clock mechanism buried deep in their DNA that tells them it's time to start growing again, after they have been exposed to a certain number of hours or days of cold weather.  However, what happens if that internal clock tells them it's time to grow and they're still frozen solid?

Now, this is pure conjecture on my part but, based on my observations in the past year, I would theorize that (1) modern roses, although they do not experience dormancy, might share a similar DNA clock mechanism with plants that do, such as their first-cousins, the species roses; and (2) the growth signal coming from within the plant might be distorted by longer than historically normal periods of remaining frozen, such that the plant's internal growth pattern is interrupted, or even curtailed altogether, thereby causing much slower growth or even plant death.  This could account for what I and a number of Minnesota friends experienced in our warmer, but longer than normal, winter of 2013.  This was truly something I had never seen in my near-lifetime of growing roses in zones 4, 5 and 6.

I had been thinking about this since last spring and Gary Ritchie's four-part series in the American Rose was such an “a-ha” moment for me, that I couldn't wait for the next installment to come.  Gary might not agree, but it seems logical to me that, while modern (non-species) roses do not experience dormancy, per se, they might share some form of the so called "chilling requirement" of species roses.  There is much we don't know about the effects of winter on roses but, by observing the effects of the weather anomalies we are currently experiencing, we can learn a lot about what makes our roses tick and how we can better protect them in winter.  Unfortunately, we can't do much about the undue length of some winters, except to realize that not all winter effects on roses are related to extreme low temperatures.

I would be very interested in the reactions of readers to the theories I have set forth in this article.  My findings are 100% empirical and can be enhanced by the observations of others growing roses in cold zones. As always, please let me know what you think.

Jack Falker
September 17, 2015

Note:  Dr. Gary Ritchie's four articles on Dormancy appeared in the May/June, July/August, and September/October, 2013, and the January/February 2014 issues of the American Rose.  By the way, articles like these, written by outstanding rose-scientists like Gary Ritchie, are one more reason that all rosarians should be members of the American Rose Society!


  1. This is thought provoking. Number of days must be one of many different ways a plant might know that it was time to break dormancy. Air temperature would be another, and soil temperature yet another. Daylight length could also be one. And various species may rely on one of these signals, or perhaps a combination of signals. There is so much we don't know!

  2. Kathy... I'm convinced that length of winter, as much as severity, is a factor in survivability. Otherwise people around here would not have lost zone 3 hardy shrubs in 2013/2014, when the Extreme Minimum Temp only reached zone 4 by one degree (-21) on one night of the winter. But there were many below zero nights in the zone 6 and zone 5 ranges and winter extended well into April and even early May, when things were "scheduled" to start growing and couldn't. This was the first time we'd noticed this because earlier winters had always been much colder in January and February. So this is a learning experience for us, if we pay attention ti what's happening. I didn't lose anything but had several very hardy zone 3/4 plants have trouble getting started in May 2014.