Container gardens

Gardening tools

Soil Compaction

Propagation through cuttings

Community tree plantings

Magnolia grandiflora

Horticultural tips

Transplanting tips

Aftercare tips

Mulch and restoration factsheet

Root inspection

Shoot and crown inspection

Sequoiadendron giganteum

Pacific madrone

The Myth of Soil Amendments

Compost tea


Sustainable Horticulture

The role of soil environment in Arbutus menziesii (Pacific madrone) seedling success


Arbutus menziesii (Pursh.), commonly known as the Pacific madrone, madrona, madrone, madronia, or madrono, is a member of the Ericaceae. This species is native to a region stretching from Vancouver Island and western mainland British Colombia south to San Diego, California. Pacific madrones are typically found in areas with mild temperatures and limited climatic extremes. In its native habitat, the Pacific madrone is an early successional species that prefers full to moderate sunlight. This evergreen tree has thick, glossy, alternately arranged leaves and a broadly spreading crown with curved limbs. Young trunks are often orange, while young stems are greenish. The bark matures to a deep red that peels off in large, thin sheets to expose a smooth reddish trunk. A. menziesii also has terminal clusters of white-to-pink urn-shaped flowers and orange or red fruit.

Pacific madrones are declining in health and numbers, but no one fully understands why. A number of theories have been proposed, and a number of potential causes are known. While much research has focused on biotic factors, relatively little has studied the effects of abiotic stresses on madrone health. Since most madrone pathogens are said to attack only weakened trees, we decided to study a predisposing factor. By investigating the effects of soil environment on the health and survival of madrone seedlings, we hoped to learn how trees respond to initial stresses. Our study involved placing seedlings into six treatments: sandy loam or clay loam soil, with loose, moderate, or heavy compaction. We hypothesized that the plants in the heavily compacted clay loam would fare the worst, while those in the loosely compacted sandy loam would perform the best.

The Pacific madrone typically grows in well-drained, nutrient poor soils and thrives in areas with disturbed soils and reduced litterfall. Madrones tend to inhabit exposed sites and are good indicators of moderately or very dry soils. What constitutes an optimal soil for madrones is still in question. Some authors state that silty or clay loams lead to poor performance, while some claim that clay loams are suitable for madrones. Others note that madrones grow in a wide range of textures, from sandy loam to clay or gravelly clay loam. Shoffner (1996) found that madrones grown in silty or clay loams had higher mortality than those grown in sandy loams. The difference was ostensibly due to poor drainage and anoxic conditions in the finer textured soils.

Soil compaction is often alluded to when discussing madrone decline or failure. Compaction frequently occurs when an area is developed, leading a variety of problems. Shallow-rooted madrones often don't recover well from root disturbance or loss, which suggests that compaction would be particularly hard on them.

Compaction and the anoxic conditions it induces can damage madrones and other plants in a number of ways. For instance, compaction compresses soil pores that hold air during drier periods, leaving soils less aerated and porous. Low redox potentials allow iron and manganese to accumulate to toxic levels, while also killing roots and microbes. Stressed plants often close their stomates, lowering stomatal conductance, photosynthesis, and respiration rates in the process. Anoxic conditions prevent root expansion and eventually kill entire roots and associated mycorrhizae, while also inhibiting shoot and leaf expansion. Ethylene, a plant hormone, accumulates in response to stress, further inhibiting growth and contributing to leaf chlorosis and senescence. The temperature of compacted soils is often lower than that in non-compacted soils. Finally, compaction increases susceptibility to and damage from root rots while also limiting regeneration and replacement.

Soil texture can further contribute to the poor health of plants in compacted soils. Clay loams, with their smaller pores and finer particles, are less well aerated and can be more easily compacted than are sandy loams. Clay loams can be quite detrimental to the health of plants that cannot tolerate low oxygen conditions, including madrones. To further complicate matters, clay soils bind nutrients more tightly, making them less available to plants, especially stressed plants with small, unhealthy root systems. The net result is that both clay textures and heavily compacted soils can injure or kill plants that require aerated, well-drained soil environments.

Recently, people have begun to appreciate Pacific madrones for their aesthetic and other values. Concerned researchers are looking more closely into the decline of the madrone, trying to stop the downward spiral in the trees' health.

What's Killing Madrones?

While madrone seedling germination rates in the wild tend to be very high (close to 90 percent), mortality is also high, often 90-100 percent. Seedlings are killed by a variety of factors, most commonly damping-off fungi, root rot, desiccation, predation by slugs and other invertebrates, smothering by litterfall, and frost heave (Tappeiner et al. 1986).

Adult madrones can have a number of health problems, some of which are fatal, and others that weaken a tree's defenses, providing an entryway for more damaging pathogens. Most pathogens attack only weakened or stressed trees. Many fungi that attack madrones are foliar pathogens that pose no serious threat unless repeated defoliation occurs. Though a number of fungi and other pathogens are associated with madrones, three pathogens are most likely to cause madrone death: Phytophthora cactorum, Nattrassia mangiferae, and Fusicoccum aesculi.

Phytophthora cactorum is a pathogen that causes a root rot, the most common root disease in mature madrones; it can also cause trunk cankers. The infected bark turns dark brown to black and the sapwood may also be discolored to a depth of 1-2 mm (Elliott 1999). Early symptoms of infection include a loss of upper crown foliage, browning and death of new leaves, and unusually small leaves with curled margins. Trees in poorly drained soil are the most susceptible to P. cactorum.

Nattrassia mangiferae (Arbutus canker) is another fungus which causes cankers in Arbutus menziesii. These cankers first emerge as areas of bark discoloration. Eventually, the discolored bark peels off, revealing masses of dark fungal spores and longitudinally cracked wood. In most cases, a callused ridge forms around the canker's margin; however, rapidly spreading cankers have smooth margins and no callus tissue (Elliott 1999). Arbutus cankers are most commonly found on trees injured by pruning or mechanical causes.

Fusicoccum aesculi (called Botryosphaeria dothidea in its sexual stage) is another canker-causing fungus in madrones, and one that works in conjunction with branch dieback, attacking branch tips, then gradually moving inward. This fungus does not attack vigorous trees, preferring trees weakened by other canker fungi or water stress. In diseased trees, the bark is dark red, turning a deep black that looks like a burn wound after the infected branch dies. Removing dead or diseased branches will prevent attack by F. aesculi.

Foliar diseases are also common among madrones, though they usually cause only cosmetic damage. Older leaves are attacked by pathogenic fungi, usually between late fall and early spring, and are particularly susceptible during warm, wet weather. Infected leaves are dropped in the spring, then replaced by new leaves in the summer. At least nineteen different fungi cause leaf spots in Pacific madrones. The most common genera include Didymosporium, Diplodia, Coccomyces, Rhytisma, and Mycosphaerella.

Numerous other stresses can weaken madrones, sometimes leading to pathogen entry and tree death. In general, A. menziesii can establish itself in environments that are already disturbed. However, if a disturbance affects an established madrone the tree is usually weakened or killed. Poorly drained or heavily irrigated soils tend to stress madrones. Pruning or mechanical wounds weaken trees and provide an entry point for pathogens. According to Shoffner (1996), trees with thin bark or small canopies are more susceptible to injuries such as sunscald or cankers. Any damage to or loss of the rooting system may also lead to root rot or the tree's death. Finally, turf can create problems for madrones by utilizing vast quantities of water and nutrients, making them unavailable to the madrones.

Our Study

We began our experiment by gathering seeds from several healthy trees (those with no cankers, galls, significant canopy dieback, or other signs of damage or stress) in the Seattle area. The seeds were stratified, then sown and germinated. At 6 weeks of age, 120 seedlings were transplanted into sandy loam and clay loam. After transplanting, the 60 seedlings in each soil texture were randomly placed in one of six soil treatments (20 per treatment): sandy loam or clay loam with low, moderate, or heavy compaction. For moderate compaction, the soil was compacted down 0.5 cm, while for heavy compaction the soil was compacted 1 cm, using the plywood device shown in Figure 1. The plants were then moved outside to a protective cage covered with shade cloth.

Soil environment

To obtain a more detailed picture of the soil environment, pH, bulk density, and redox (reduction-oxidation potential) measurements were taken in the soils. Redox measurements, which indicate how aerobic a sample is, were taken for every plant. Redox potentials vary with porosity, microbial activity, and nutrient, oxygen, and water levels in a given section of soil. Plants that aren't adapted to flooding, including madrones, suffer at low redox potentials. Without sufficient oxygen, roots will stop growing and functioning and may die, leading to severe water and nutrient stress.

Redox measurements were fairly consistent between treatments, with the largest range in the most heavily compacted soils. Most of our measurements were right on the edge between aerobic and anaerobic conditions, but a few of the plants were clearly experiencing anoxic conditions, and these plants were smaller and less healthy than plants in more aerobic environments.

Along with the redox measurements, several observations suggested that the containerized environments were anoxic at least part of the time. Standing water was regularly observed in clay loam soil, particularly in the moderate and heavily compacted treatments, and lasted anywhere from one day to two weeks before draining (Figure 2). Standing water in the sandy loam occurred far less frequently and rarely lasted more than several hours. At our study's end, we noted soil mottling, with both gleyed and reddish sections of soil indicative of iron deposits and a fluctuating water table. In addition, a strong sulfur smell was present in many containers of clay loam soil, indicating anaerobic microbial activity.

To obtain pH data, five plants per treatment were randomly sampled. We found soil pH to be quite dependent on soil texture. In the clay loam, more compacted soil was less acidic, but increasing compaction in the sandy loam led to more acidic soil. Interestingly, higher pH's tended to correspond to lighter and smaller plants, perhaps because the plants had smaller root systems or fewer available nutrients.

Bulk density was also measured on five plants per treatment. As expected, we found that bulk densities were lower in the sandy soils and generally increased with compaction level.

Gas Exchange Analysis

We used an infrared gas analysis system to measure CO 2 concentrations, photosynthetically active radiation, transpiration rates, stomatal conductance, and several other factors affecting gas exchange in the madrones. The most negative carbon assimilation rates meant that a plant was taking in more CO 2 for photosynthesis. A few plants in clay loam were giving off more CO 2 than they were taking in, giving them positive rates. Since increased disease susceptibility is associated with reduced capacity for photosynthesis, this could create problems for those plants which had low (or nonexistent) rates of CO 2 incorporation as well as low stomatal conductance, greatly lowering their photosynthetic capacities.

Transpiration rates, unlike our other variables, showed no clear trends. None of our environmental variables were very highly correlated with transpiration rates, though we found that transpiration and size had a general tendency to increase together. Perhaps healthier, more actively growing plants could afford to lose more water, taking up additional CO 2 to use in growth and compensating for any loss through having root systems that took up sufficient water.

Stomatal conductance is a measure of how well a plant can control its stomata, ideally opening them only when doing so benefits the plant. In general, higher conductance rates are better, as they indicate that more stomata are open, allowing the plant a greater opportunity to photosynthesize. However, if the stomata don't close in times of high temperature and/or low water levels, the plant can suffer drought symptoms, including anthocyanin production and smaller leaves and stems. We noted such symptoms more frequently among plants in clay loam and in the most heavily compacted sandy loam. We also found that higher bulk densities and clay texture both led to decreased stomatal conductance, likely due to the stress they induced on roots and plants.

Leaf Color

Leaf color was analyzed as a measure of the presence or absence of anthocyanins, which are water-soluble pigments that give plants red, blue, or purple coloration and can act to limit the harmful effects of UVB, drought, freeze, and other environmental stresses (Chalker-Scott 1999). We determined the predominant color(s) for each of the plants, defined as colors that covered a majority of at least two leaves on a given plant.

Several of our plants, particularly those in the clay or more compacted soils, turned shades of red, purple, and orange. Many plants began to redden, but reverted to green coloration following fertilization, suggesting nutrient deficiencies accounted for some of the color change. Other plants, especially in the clay soil, remained red or grew redder with time, particularly after they were moved outside. Simply being green did not mean that a leaf was healthy, as paler shades of green often were indicative of nutrient deficiency.

Growth Variables

At our study's end, ten plants from each experimental group were randomly selected for harvest. Fresh weights, root and shoot lengths, and leaf areas were measured, then dry weights were obtained. The root, shoot, and leaf data supported our hypothesis. In every category there were basic gradients of response, with plants in loosely compacted sandy soil performing the best and those in heavily compacted clay soil faring the worst. Texture played a slightly bigger role than bulk density in determining plant size, which was unexpected, but not too surprising given the heavy clay soil we used.

Madrones have several adaptations that allow them to live in soils with low moisture content. Zwieniecki and Newton (1995) found that roots growing in compact, clayey soils tend to be thick and short, but roots growing in rock fissures are thinner, longer, and less well branched than normal, allowing madrones to find and take up any available water. We, however, found that roots in clay loam and the more compacted sandy loam tended to be thinner and shorter than those in less compacted sandy loam, perhaps because they found small spaces in which to grow. These root systems also exhibited less branching, fewer fine roots, and more dead roots than did those in less-compacted and sandy loam soils.

Before inducing compaction, 5 plants per soil texture were randomly chosen to obtain baseline data on root and shoot length. At that time, both shoot and root lengths were longer in the sandy loam, and this remained true throughout the experiment. By the study's end, between-texture differences were more pronounced and between-treatment comparisons could be made. The sandy loam plants made bigger gains in root and shoot length than did the clay loam plants, with one exception: clay-low roots grew more than sandy-high compaction roots. Within both texture classes, plants in less compacted soils made larger gains in root and shoot length than did those in more compacted soils. In both textures, ending root lengths varied considerably between low and moderate compaction, with far smaller differences between moderate and high compaction. In each treatment, root lengths varied more than the shoot lengths did, with both a wider range and a larger standard deviation. Roots were forced to find aerated areas in which no rocks or other obstacles to growth existed. Shoots, meanwhile, were only limited by the relative amount of energy the plant put into producing and lengthening its aboveground portion. Shoot lengths were quite consistent among the six treatments, always averaging 3 inches, with slightly longer shoots in sandy loam.

Fresh and dry weights showed similar trends to shoot and root lengths. Dry weights varied less than fresh weights, with more-similar means throughout the treatments as well as smaller standard deviations. We found that texture has a greater influence on weight than does compaction level, but the factors overlap in their influences. For fresh weight, averages varied by about 2 grams within a texture, and about 4 grams between the textures. In dry weight, meanwhile, bigger differences occurred between compaction levels within the sandy loam than within the clay loam.

One interesting note is that differences between the fresh and dry weights of those plants in sandy soils were about twice that for plants in clay soils (7 and 3.5 grams, respectively). This could simply be due to the higher fresh weights found in the sandy soils or it may mean that the sandy loam plants utilized their larger, healthier root systems to take up and retain more water.

Finally, leaf area behaved much like fresh and dry weight. Leaf area means varied by about 20 cm 2 between high and low compaction in each texture. Large differences existed between the textures, with leaf area in sandy loam averaging twice that in clay loam. Compaction at either level was enough to drop leaf area by 15 cm 2 or more, though differences between moderate and low compaction were relatively small.


If the Pacific madrone is going to be saved, we believe that two things must happen:

  • Treatments for existing diseases must be found.
  • Currently healthy plants must be kept healthy through preventing the cultural stresses that leave them susceptible to pathogens
Though little has been studied concerning the impacts of soil environment on Arbutus menziesii success, others have noted that this species has a fragile root system, likes well-drained soil, and doesn't tolerate disturbance. Our study supports those findings and suggests that texture and bulk density both influence plant success, with both clay loam and high bulk densities causing the same stunted growth, decreased gas exchange capability, and general poor plant performance. Texture has a bigger impact on madrone health and survival than does compaction level, though most species would have had difficulty thriving in the heavy clay loam. Finally, we found that any level of induced compaction causes a significant decline in plant health, but differences in plant response between moderate and heavy compaction levels tend to be rather small.

A stressful or inappropriate soil environment may well be one of the key stresses that trigger the downward spiral of A. menziesii. Plants that are suffering from damaged or dead roots, lowered photosynthetic ability, and saturated soil conditions are bound to be weakened and lack disease-resistance. These are precisely the sorts of trees that Nattrassia mangiferae,Phytophthora, and other fungi like to attack. In addition, many madrones are placed in urban areas surrounded by turf or other plants that require far more irrigation than do madrones. Heavy watering will flood the madrone root zone, potentially spreading disease or creating an oxygen-poor environment that limits root growth. Protecting madrones from compaction, poor drainage, and other soil-related stresses could prevent them from becoming weakened and disease-prone.

Since compaction tends to have the same general impacts on all non-tolerant plants, our findings could also be applied to other species. Urbanization in general--and soil compaction in particular--contributes to the poor health and failure of many plants. Once compaction has been induced it can be quite difficult to reverse. Fortunately, soil compaction can be avoided. Before a disturbance occurs, existing plants should be protected with barriers at least twice as wide as their driplines or circles the same circumference as the plant's height. To preserve fragile roots, construction equipment and other disturbances should not be allowed in these protected zones. In addition, a thick layer of well-drained mulch should be spread in any area that will receive heavy traffic. The mulch will spread the weight of the traffic, minimizing the compaction damage in any one area. Finally, construction should not take place on wet soils, as they are far more easily compacted than are dry soils.

We believe that additional research should be done on such factors as compaction in the field, fertilizers, mycorrhizae, excess irrigation and drainage problems, and other variables that impact root--and ultimately tree--health, as well as on urbanization impacts in general. Determining the major stresses contributing to madrone decline, then preventing them from impacting additional trees, will be the key to saving this magnificent species for future generations to love and admire.


This research was funded through a Mary Gates Research Training Grant provided by the Mary Gates Endowment for Students and the University of Washington. Our friends and colleagues at the Center for Urban Horticulture were quite helpful and supportive. In particular, we wish to thank Kern Ewing, Perry Gayaldo, Jim Scott, and Barbara Selemon for their willingness to provide any needed assistance.


Chalker-Scott, L. 1999. Environmental significance of anthocyanins in plant stress responses. Photochemistry and Photobiology 70: 1-9.

Elliott, M. 1999. Diseases of Pacific madrone. In Adams and Hamilton (eds.), The Decline of Pacific Madrone (Arbutus menziesii Pursh): Current Theory and Research Directions. Pollard Group: Tacoma, WA pp. 42-59.

McDonald, P.M and J.C. Tappeiner, II. 1990. Arbutus menziesii Pursh: Pacific Madrone. In Burns, R.M. and B.H. Honkala (Eds.) Silvics of North America (Vol. 2, pp. 124-132) Washington, D.C.: GPO.

Shoffner, T. 1996. Light, irrigation, and native soil addition effects upon the establishment and growth of outplanted Arbutus menziesii seedlings in the urban landscape. Master's Thesis, University of Washington 91 p.

Tappeiner, J.C. II, P.M. McDonald, and T.F. Hughes. 1986. Survival of the tanoak (Lithocarpus densiflorus) and Pacific madrone (Arbutus menziesii) seedlings in forests of southwestern Oregon. New Forests 1:43-55.

Zwieniecki, M.A. and M. Newton. 1995. Roots growing in rock fissures: Their morphological adaptation. Plant and Soil 172:181-187.

Table 1: Generalized impacts of texture and increasing compaction on the variables


Increased Compaction

Sandy loam (as compared to clay loam)

Clay loam (as compared to sandy loam)

Redox potential

Decreased-more so in sandy loam

» 50mV higher



Higher in sandy; lower in clay

More acidic--mean of 5.34

Less acidic--mean of 5.34

Bulk density


Lower than clay loam



More non-green (clay); slightly more green(sandy)

More green

More non-green; wider range of colors


No clear trends

No clear trends

No clear trends

Stomatal conductance

Decreases slightly

Much higher

Much lower

Differential CO 2 assimilation

Slightly lower

Much higher--mean of -13ppm

Much lower --mean of -4.5ppm

Root length



Shorter (each compaction level)

Shoot length

Slightly shorter; skinnier

Slightly longer; thicker

Slightly shorter; skinnier

Leaf area

Slightly smaller

Much larger

Much smaller

Fresh weight

Slightly lower

Much higher

Much lower

Dry weight

Slightly lower

Slightly higher

Slightly lower

Figure 3: Typical plants from the sandy, low compaction treatment (left) vs. typical plants from the clay, high compaction treatment (right)

sandy, low compaction clay, high compaction

CoContact Us 2006-2008Sustainable Horticulture