The buzz of the horticultural world lately has been the release of the new USDA Plant-Hardiness Zone Map. So, since I am an avid amateur dabbler in all things weather and climate, I thought I might as well chime in. This actually happened a few weeks ago so I am a little late getting into the game, but hey, better late than never!
The USDA zone map was first developed a long, long time ago when dinosaurs roamed the earth. Based on a long-term average of the coldest temperature to occur each winter at a given location, it was intended to serve as a general guide to discern in which parts of the country certain plants could succeed based on their cold-hardiness. It has since taken on many different forms and grown into an industry-dominating monster with its own will far beyond that of its creators. Think I’m exaggerating? Read on.
Now the new map, it must be said, is far more accurate than any of the preceding maps (which include a couple of “botched” versions I will not discuss, to keep this from getting any longer). A major complaint about the 1990 map (the previous major USDA map produced) (warning: really big download) was that it was based on an anomalously cold period in history; thus, it made everyone look colder than they really are. It was also based on a very short time span of just 13 years. Why? Well, that’s the big question, isn’t it? No serious, professional climatologist or meteorologist uses a period that short to determine averages or “normals” for any meteorological purpose. When you look at the “average” high or low for a given date on the TV newscast or online, it is based on a much longer period: 30 years is the standard for the National Weather Service. It’s hard to dispute that this use of a 13 year sample was a deliberate and foolish decision, although I’m not quite sure whose or why. In any case, once people realized winters weren’t going to continue to be as cold as they were from 1974 – 1986 (the period used), the need for a new map was urgent. So, faster than the speed of government, we have our new 2011 map, which happens to be based on data from 1975 – 2005 since we have to allow them a few years to figure out what to do with all those data (for more specific details on the history of the maps, see “Further Reading” below).
But – lest I sound too cynical here – the new map is accurate. How do I know? Well, it’s because I’m one of those nerdy people who has independently researched the matter in great detail over the years. When the new map came out, I was impressed: it broadly agrees with my calculations for USDA zones at dozens of locations in the Pacific Northwest. Many of these I provide for you below. I should mention that I am not the only one who thinks it is quite accurate: many established industry professionals and avid hobbyists alike – at least, those who are closely tracking the weather – would agree with me. While new map is generally accurate, it is not perfect. But it is as good as it’s going to get considering the amount of data that are available.
What do I mean by “accurate” but “not perfect?” Here’s one example of why not having enough data compromises accuracy. In the Puget Sound area of western Washington, nearly all of the weather stations with decent periods of record are in average to cold microclimates and mostly not that close to the moderating influence of salt water, despite that the majority of people with gardens tend to live in relatively close proximity to the water. Anecdotal evidence clearly tells us that a lot of sheltered microclimates close to the water are well within zone 9, which on the map just barely shows in just a couple tiny places (mostly over water). This assertion is also supported by a comparison of temperature data from personal weather stations, which have proliferated exponentially in just the last few years, with data from the official stations with longer periods of record. Rarely, we see this work the other way: in Port Townsend, the weather station is close to downtown and the water, and very mild; while Kelly and Sue at Far Reaches Farm, which is a couple miles inland, consider themselves to be about a full zone colder, supported (very importantly) with their observational data from the last few winters. With the weather station being in the mild spot, the colder microclimates just outside of Port Townsend don’t show up so well.
But all this is small beans: the map is still pretty darn accurate, and as accurate as we can ever expect it to get. Enough said, right? Oh, just wait…
Now I’m seeing a variety of responses to the new map, but I will just offer my thoughts on a couple of the most common ones I see. Some, upon seeing the new map, put it forth as evidence that our climate is getting warmer, since many of us have moved up a half zone or so from the 1990 map. Frequently, these people are not aware that the 1990 map was based on an anomalously cold period in history, as I have described above. In fact if you go back even farther and look at the really old 1960 zone map, it’s pretty darn close overall to the 2011 map (except for being colder in parts of the Intermountain West):
(Incidentally, I know that out there somewhere is a detailed, colored, 1960 zone map; with a and b zones. There used to be a poster of it on the wall in the WSU halls of the horticulture department. I wish I had taken a photo of it because I have never found it online. Has anyone seen this? Let me know!)
Now I’m not a “global warming denier,” but I get a little irritated when the people with the most passion about this subject tend to lack the science to back it up. Sure, our climate’s getting warmer, but it has little to do with this map. I also find the number of news releases citing the new map as direct evidence of global warming to be pretty ridiculous.
Then – and I’ll spend much more time on this one – we have people who are convinced beyond any doubt that the map is not accurate, as if this were some kind of vast conspiracy. I have carefully avoided getting into big arguments about this on several occasions since I think these sorts of debates are not usually helpful without the chance to present my case in its entirety (and hear theirs out in detail if they have one). For example, one of these was (or, should I say, might have been) with a lady from Boise – a very experienced gardener and plantsperson – who seemed to be not taking the map seriously when it put Boise in zone 7a. The thing is, statistics don’t lie: if you take the average of every winter’s coldest temperature, and it comes out between 0 – 5°F, you’re in zone 7a and that’s that. Did I go wrong somewhere?
Well then once that point has been made, the response tends to be (and here I’m speaking generally, not of the aforementioned person – which I’d better say just in case she reads this!), “Then the USDA zones need to be adjusted/are not really that useful/whatever.” OK, so why not just say that in the first place? Don’t go saying “There’s no way we’re in such and such zone” without statistics to back it up – pretty please. It’s not like we can deny how the weather is. When was the last time you heard someone say “It’s 45 degrees out! It has to be!” when the thermometer shows it’s 55 degrees? With USDA zones, the only difference is it’s not happening in the here and now, so you have to actually go back and look at the historical data.
I can only conclude that the reason a lot of people don’t trust the map is that it does not fit their perception of reality, or of some notion (of vague origins, but very strong nonetheless) of “What My Zone Should Be” (and sometimes other people’s zones too!). Now I know that much of this discrepancy is based on actual plant performance, which is important to discuss since, supposedly, the main function of the map is as some sort of predictor of plant performance. So how did this discrepancy arise? Let’s consider an example: Some (not me!) might suggest that, say, Arbutus unedo will grow in Boise because it’s rated hardy to USDA zone 7. What I want to know is, why, when a discrepancy arises, do so many gardeners immediately distrust the map? Could it be that perhaps Arbutus unedo should not be rated to zone 7?
And I actually think this is the major problem with our understanding of USDA zones: PLANTS ARE NOT ACCURATELY RATED FOR THE RIGHT USDA ZONE. And now that we have a good, trustworthy map, it is the only problem (considering the limitations of the system in general, as I will discuss below). Prior to the 1990 map, it seemed that plants were often rated a bit too optimistically, but things still jived well enough that no one made a big deal over it. Now that we have the new map, the discrepancy is even greater!
So who’s behind all this? Gardeners get plant-hardiness ratings from many sources: books, plant and seed catalogs, growers’ tags, and the internet, to name a few. Ultimately the problem is that, somewhere in the pipeline, someone draws assumptions about plants and climate zones without sensible supporting data. Based on my observations, I would say growers’ tags are especially problematic: many companies/tags/etc. will rate a plant’s hardiness according to its ability to tolerate an average winter in the zone number it is assigned, seeming to forget that weather deviates significantly from average from one year to the next. For example, I often see Cordyline australis rated as “zone 8.” What is up with that? It is certainly no hardier than 15°F. Even at the top end of zone 8, you’re likely to drop below 15°F at least once or twice in 10 years – enough to freeze your plant to the ground — and at the bottom end, forget about it; you’ll be lucky if you ever get it through two winters in a row.
And this isn’t just a problem with a few certain plants. If you start looking carefully at books, and plant tags, and comparing these ratings with actual results with the plants (in their respective zones as they truly are); you will notice that most of the time plants are rated with the same degree (pun intended!) of inaccuracy; and, I think it’s fair to say, based on the same faulty assumptions.
Now we begin to understand the “problem” with Boise. (And anywhere else, too, but I’m using Boise, not because I like to pick on them, but because it is a nice extreme example of how deviation from average messes with zone ratings and the assumptions surrounding them.) So it’s zone 7a: but during the time period used in the data sample, the coldest temperature recorded was -25°F! Now how many plants rated to zone 7a can live through that? A couple other winters were close to that as well: the climate of Boise exhibits extreme deviation from their average coldest temperature every winter. Boise has also had a lot of mild winters (especially recently) with no temperatures anywhere near 0°F; and — wouldn’t you know it — these have to be averaged in as well to compute their USDA zone!
So it’s easy to see what motivates a perception that the map must be lying to us – or even a desire to “keep” Boise in zone 6 (or even 5?) if such a thing were possible. Nurseries want gardeners to succeed with their plants. When customers ask nursery staff in Boise why they can’t recommend plants that are rated to zone 7, they need to have a succinct answer that doesn’t call their grower’s tags (and hence, the integrity of their product) and reference books (potentially associated with the integrity of their expertise) into question. Maybe it’s just me but I would call this a mighty big problem.
And if we look just a little further at “who’s behind all this,” one more interesting factoid comes to light. When the USDA published the 1990 zone map, they also published statements about each zone with a list of plants that can be “expected” to be found within that zone. For example, within zone 9, one usually “expects” to find Phoenix canariensis and Grevillea robusta, among others.
I believe this to have been huge tactical error: not that it’s terribly wrong in general, but because it sets forth a backwards way of thinking about zones. Many gardeners, including the very experienced and professional, have become so firmly convinced that certain plants are “zone X plants” that they think (usually subconsciously, and without admitting it) that plants can tell us better than historical weather data what zone we are in. Sure, plants can tell us a lot: it’s where I get some of my anecdotal evidence for zones locally in a region I am very familiar with. But to draw conclusions about zones based on plant performance observations always requires support from hard weather data first and foremost. Frequently, our firmly entrenched “zone X plant” categories often tend to be based on incorrect information, as described above. What we need is for zone ratings for plants to be based on actual evidence based upon the zones, not the plants. For anyone to draw conclusions about their zone based on whether a “zone X plant” will thrive in their area, is to look at this all backwards.
What we also need is to get back to the original intent of the map, which is as a general guide. I hear a lot of gardeners who claim this mentality, yet still look with extreme suspicion upon that Grevillea which is in fact hardy to zone 8b, when considering it for use in zone 8b; or who refuse to let go of their “zone identity” because of their plant-zone-based perceptions, regardless of what maps and statistics tell them. In both cases we see how firmly conclusions about the USDA zones in which we garden are rooted in the assumed accuracy of zone ratings assigned to plants, which are assigned by people who usually remain (for most purposes) completely anonymous to the end user, and are called into question all too infrequently.
So, what to do? I almost think a rigorous national campaign to re-assign zone ratings to garden plants is in order, but if the USDA takes charge of this, our climate may have completely changed again in the time it takes them to produce any results. But in any case, the line of thinking would go like this: Will Arbutus unedo grow in Boise? No. Therefore, it’s not a zone 7 plant. Will Cordyline australis grow in zone 8 in western Washington? No – therefore, it’s not a zone 8 plant. Will Grevillea robusta grow in zone 9 in downtown Port Townsend? No, so it’s not a zone 9 plant. It’s just a plant. It’s not a set of historical climate data used to determine a USDA zone. Now (perhaps slightly tangentially) this way of looking at plants will also result in some additional discrepancies highlighting the need to think of the map as no more than a general guide. For example, Carpobrotus edulis grows happily in zone 8a in New Mexico, but can’t handle zone 8b in Washington. This is because of other climate factors such as summer heat, and light and precipitation patterns throughout the year. Climate is ultimately regional, and plant performance is ultimately based on empirical evidence: hence the limitations of the USDA map are significant. To their credit, the USDA is pretty clear about this stuff on their new web site.
And in case reading this hasn’t made you completely confused yet, I’ll just add this: In my conversations with customers, I notice that gardeners in western Washington often experience hardiness let-downs based on cultural conditions. For example, if you plant that Grevillea in rich, heavily amended soil and water it all the time, so that it keeps growing into the fall and fails to harden off; it should hardly be surprising that it takes a hit when that “Arctic blast” comes along. When we pamper our plants more than they get pampered in nature, hardiness often suffers. Of course there is a time and a place for soil amendment and watering – for example, I would pretty much always recommend amending severely degraded or compacted soils – but balance is needed. Back to the point, though; this is just one more factor among many that often skews our perceptions of plant hardiness.
So where does all this leave us at the Desert Northwest? Well, I haven’t been rating hardiness for the plants we sell according to the USDA zone system – not yet anyway. Basically it’s because I don’t want to mislead anyone, since it’s hard to predict how readers of our plant descriptions will understand a zone rating. I prefer to assign actual temperatures at which a certain plant will be damaged, killed, etc. with some wiggle room for other factors as described above (duration of the freeze, etc.). These are generally based on actual experience growing them here in western Washington, or our best educated guess. Also, not infrequently, we make the comment in our plant descriptions that such-and-such is likely to be hardier in a climate with hot summers. We figure a lot of plants (though not all) that just make it in the Pacific Northwest can probably handle temperatures a full zone colder in the Southeast, or nearly; and we make note of it. Finally, we wish to emphasize that plants don’t look at maps or weather data, and sometimes the only way to determine whether something is hardy enough for your garden is to try it and find out, and sometimes more than once. We do plenty of that around here!
Do I sound mad? I’m not really. At least, not in a bad way. But I’m definitely crazy. And to prove it, just for fun, here are some USDA zones for a number of randomly selected Northwest locations, which I calculated based on official historical climate data. The middle column is the average of the coldest temperature every winter between 1969 – 99. The source for nearly all this data was accessed via this page on the ESRL web site. Through this page you can do your own independent research if you like, and I think you will be surprised with the accuracy of the map! (Note: for the most part, my results agree with the map pretty well, or are a little lower than what the map suggests. This discrepancy is because the data below are based on 1969 – 99, not 1975 – 2005 as in the map. The period of 2000 – 2005 had mostly very mild winters in the Pacific Northwest. Thus, Boise, for example, shows as 6b here rather than 7a, since the milder winters in the early 2000’s are excluded.)
Aberdeen | 19.97 | zone 8b
Anacortes | 17.3 | zone 8b
Battle Ground | 12.93 | zone 8a
Blaine 1NNE | 11.37 | zone 8a
Buckley 1NE | 13.8 | zone 8a
Centralia | 13.03 | zone 8a
Chelan | 4.47 | zone 7a
Clearwater | 16.2 | zone 8b
Coupeville 1S | 15.2 | zone 8b
Dallesport Airport | 7.93 | zone 7b
Darrington | 7.53 | zone 7b
Ellensburg | -6.83 | zone 6a
Elma | 13.6 | zone 8a
Elwha Ranger Stn | 17.3 | zone 8b
Everett | 15.37 | zone 8a
Forks 1E | 17.03 | zone 8b
Goldendale | 1.23 | zone 7a
Kennewick | 3.97 | zone 7a
Longview | 16.33 | zone 8b
Olympia Airport | 8.57 | zone 7b
Omak 4N | -6.2 | zone 6a
Othello | -2.23 | zone 6b
Port Angeles | 18.97 | zone 8b
Port Townsend | 20.9 | zone 9a
Pullman 2NW | -5.6 | zone 6a
Quincy 3S | -5.17 | zone 6a
Richland | 4.03 | zone 7a
Sea-Tac Airport | 18.57 | zone 8b
Spokane Airport | -7 | zone 6a
Stampede Pass | 1.53 | zone 7a
Stehekin 4NW | 3.9 | zone 7a
Wenatchee | 4.37 | zone 7a
Yakima Airport | -3.27 | zone 6b
Ashland 1NW | 14.0 | zone 8a
Bandon 2NNE | 22.7 | zone 9a
Bonneville Dam | 17.17 | zone 8b
Brookings 2SE | 29.4 | zone 9b
Elkton 3SW | 19.3 | zone 8b
Eugene | 15.33 | zone 8b
Gold Beach | 27.3 | zone 9b
Grants Pass | 16.2 | zone 8b
Hood River Exp Stn | 6.43 | zone 7b
Klamath Falls 2SSW | 0.23 | zone 7a
McMinnville | 15.63 | zone 8b
Medford Airport | 14.8 | zone 8a
Newport | 21.2 | zone 9a
North Bend | 24.87 | zone 9a
Oakridge | 14.77 | zone 8a
Ontario | -4.93 | zone 6b
Portland Airport | 17.93 | zone 8b
Salem Airport | 13.43 | zone 8a
The Dalles | 9.73 | zone 7b
Tillamook | 17.37 | zone 8b
Boise | -2.7 | zone 6b
Lewiston Airport | 3.07 | zone 7a
Moscow | -8.13 | zone 6a
Riggins | 6.6 | zone 7b
Further reading about USDA Plant Hardiness Zones and the Map:
Plant Hardiness Zone Maps: The Rest of the Story by Tony Avent
Plant Delights’ January Newsletter, discussing the new map and what went into it by Tony Avent
Climate Zones Gone Wild! A basic introduction, by me, 2008.