When I discovered that Salvia guaranitica ‘Black and Blue’ had been thriving for the last 5 years in my garden, but had been listed by Burpee and other nurseries nationwide as either an “annual” or a perennial hardy only in zones 7 to 11, I was really surprised. I remember seeing it listed as a zone 9-11. Although Fordhook Farm is a solid zone 6, ‘Black and Blue’ is a spectacular perennial here, especially with Zinnia ‘Zowie Orange and Scarlet’. Other years we’ve planted white or yellow petunias alongside it. With its moody blue florets and bold black sepals on tall stems, ‘Black and Blue’ towers authoritatively over the playful, effusive annuals. The sage green foliage is perfectly matched with its blooms, as if it knew how to perfect itself.

Also at Fordhook, plants of Phygelius aequalis ‘Sensation’ are still holding their leaves, staying green and vigorous, despite several killing frosts each of the last several years. Most P. aequalis are zone 8; this unique cultivar is hardy to 7, and here it is doing fine in our zone 6. Plus, we have reports that other varieties of P. aequalis have survived the zone 5 of Lansing, Michigan.

How reliable are zone designations?

I became a bit concerned. So last fall I asked a friend from high school, Nick Rhodehamel, the retired Managing Editor of ‘Crop Science’, the journal of the American Agronomy Society, to help me by reviewing the background of the USDA plant hardiness system. Here’s Nick’s first report:

Dear George,

The current USDA plant hardiness zone map is published as USDA misc. publ. 1475 (1990). It supercedes USDA misc. publ. 814 (1960; revised 1965). The USDA is currently revising the 1990 iteration. Print copies of these don’t seem to be available, however, even at the Natl. Arboretum where the map was devised.

The National Arbor Day Foundation has just completed an extensive update of the USDA 1990 map based on 15 years of data (see http://www.arborday.org/media/zones.cfm for the update; there is an animated comparison of the USDA map with the NADF one).

In addition, the American Horticultural Society (AHS) in 2003 created a zone map that is based on high rather than low temperatures. The AHS map uses 12 years of data. This zone map is intended to be a complement to the USDA map.

The original USDA plant hardiness zone map (1960) was developed by H.T. Skinner (1907-1984; director Natl. Arboretum 1952-1973), the second director of the Natl. Arboretum. In cooperation with AHS, he incorporated pertinent horticultural and meteorological information into the map. The basic considerations underlying the map were the following. (i) Zones-The contiguous USA and southern Canada were divided into 10 zones on the basis of a 10°F (6°C) difference in average annual minimum temperature. (ii) Winter hardiness-Survival of landscape plants over winter was selected as the most critical criterion in their adaptation to the environment. (iii) Classification-The zone ratings were intended to indicate “excellent” adaptability of the plants. Many plants may survive in warmer or colder zones. Under this scheme, survival alone does not represent satisfactory performance. (iv) Interactions with other environmental factors-Many other factors may come into play in determining satisfactory growth. Wind, soil type, soil moisture, humidity, snow, and winter sunshine may greatly affect the adaptability of plants. (v) Interactions with cultural factors-The way plants are placed in the landscape, how they are planted and their size and health can greatly influence satisfactory adaptability.

The 1990 USDA version expanded the earlier map and showed in detail the lowest temperatures that can be expected each year in the whole of North America. As in the earlier iteration, these temperatures are referred to as “average annual minimum temperatures” and are based on the lowest temperatures recorded for each of the years 1974 to 1986 in the USA and Canada and 1971 to 1984 in Mexico.

It is this concept that is used to define the zones. If, at a particular site, the lowest recorded temperature for five successive winters is 7 (-14), 10 (-12), 17.6 (-8), 3 (-16), and 30°F (-1°C), the mean coldest temperature is 14°F (-10°C). The site is then assigned to Zone 8a.

This map shows 10 different zones, each of which represents an area of winter hardiness for agricultural and natural landscape plants. It also introduced Zone 11 to represent areas that have average annual minimum temperatures above 40°F(4°C) and that are essentially frost free.

Data used for this map were obtained from 14,500 stations from USA, Canada, and Mexico; 8,000 of these stations can be identified by latitude and longitude and by a valid average annual minimum temperature. Only data from these stations were used in the map. The map was computer generated on the basis of latitude and longitude but because of the large area involved, it is not possible to draw one map that is accurate for all of North America. The part representing the USA has the least distortion. ARS proposes periodic updated maps when necessary and appropriate.

On the basis of empiricism and plant passport information, the USDA assigned particular plants to particular zones. Presumably, the data that are found in industry catalogues are also based on breeder and grower information. However, I can find very little hard information on this.

The hardiness zones are effective for the most part because cold is the primary limiting factor in whether a plant survives at a particular location. However, there are obvious drawbacks. One major one is that summer heat levels are not incorporated into zone determination. Sites that may have the same mean winter minimum but quite different summer temperatures will still be assigned to the same hardiness zone.

The 2003 ASH map (http://www.ahs.org/publications/heat_zone_map.htm#1) attempts to address this weakness by defining zones on the basis of “heat days” or temperatures over 86°F (30°C) rather than average annual minimum temperatures. The ASH map is used in the same way as the USDA hardiness zone map. However, there are 12 zones that indicate the average number of days each year that a given region will experiences heat days, and a heat day is defined as “the point at which plants begin suffering physiological damage from heat.” The zones range from Zone 1 with less than one heat day to Zone 12 with more than 210 heat days. AHS plant heat-zone ratings assume that adequate water is supplied to the roots of the plant at all times. AHS has coded thousands of garden plants for heat tolerance and suggests that their map be used in conjunction with the USDA plant hardiness zone map. Apparently, some nurseries have adopted the AHS scheme, and many garden plants are coded for both schemes at retail nurseries.

All three zone maps (USDA, 1960, 1990; NADF, 2006; AHD, 2003) use the same source of climatic data-archives of the National Climatic Data Center, though the years of data are not identical. The NADF zone map is (apparently) purely an update of the USDA scheme. The other two were created by essentially the same group using the same assumptions and methods and overseen by the same person.

Why do you see plants that are assigned to particular zones that don’t appear to be able to withstand the conditions in that zone? How do you assign a zone to an exotic cultivar? The zone maps are simply rating schemes, and as you see, they’re rather basic and rely on only a few assumptions. The interaction between plants and the environment is dynamic and complex. And even though there is an average minimum temperature for a particular zone, in some rare years, it may get considerably colder in that zone, analogous to a 25-year flood. Or there may be an unusually severe, early cold spell that occurs before plants have adapted to cold temperatures. Cold and heat tolerance are complex traits for which there are no generic laboratory tests, and there is certainly no simple way to quantitate something as non-discrete as the effect of wind on a plant. Obviously, in assigning a plant to a zone, a pretty conservative approach should be adopted, and short of passport data and/or fielding testing, I don’t see that there’s any reliable, simple way to make this assignment-particularly since some of your germplasm may be fairly exotic and niche-limited.

Hope this helps.

Best regards,

Nick