Tải bản đầy đủ - 0trang
IV. Soil Factors Restricting Growth
In Argentina, Vidal (1961) has reported poor results on soils less
than 18 inches deep, and in South Australia and Argentina a comparison
of soils defined by soil surveys has shown that the growth of P. radiuta
is poorer where the soils are shallow or have impenetrable horizons of
hardpan or concretionary material (Beckmann, 1964; Barrett and Garbosky, 19.60). However, in California trees as high as 60 to 70 feet are
found with as little as 6 inches of surface soil on weathering granite
rubble (Scott, 1960), but roots would penetrate into the parent material
and growth would be very slow.
Jackson (1965) has presented a table for the allocation of different
species to sites at Hawke’s Bay, New Zealand, taking into account rainfall, soil type, and profile depth. On red loams, a rainfall of 35 inches
per annum and a soil depth of 2% feet are required for P. rudiatu. It
would be desirable to examine the relationship between mean annual
increment, depth to the least permeable horizon, and rainfall for P.
radiata in a similar way to Jackson’s study for slash pine ( P . elliottii
Little is available on the rooting habits of P. radiuta. In California
roots penetrate only to depths of about 2 feet even in good soil, but far
deeper penetrations in clay soils (Bowen, 1964; Raupach, 1967) and in
sands have been found in South Australia. Windthrow is common for
this species on poorly structured shallow soils.
The volume of exploitable soil is sometimes limited severely by stones.
For example as much as 70 percent by weight of some lateritic podzolic
soils supporting P. radiata in South and Western Australia are retained
by a 2-mm. sieve. This is considered further in Section V, D below,
2. Chemical Barriers
The soil water regime will be discussed under Section IV, B, but
reduced conditions and high salt content in localized zones of soil give
rise to barriers which are more chemical than physical. Depressed growth
has been reported on sites with poorly drained heavy clay (e.g., Laughton, 1937; Weston, 1958), root development probably being restricted to
the non-clay layer. Solonetz soils have also been observed to be uncertain sites (Woods, 1955); here the salt content of the clay layer below
the sandy A horizon may be toxic under particular seasonal conditions.
Although clay layers have been said to restrict growth, high quality
trees are commonly observed in parts of Australia where the clay content
is 60 to 90 percent, whereas in Argentina Barrett and Garbosky (1960)
reported an upper limit of about 40 percent in the B-horizon. Clearly a
number of other soil properties should be considered along with clay
content in defining growth limits.
The soil profile features which contribute to a soil water regime
unsatisfactory for tree growth may be thought of as depending on both
the total quantity of available water present in the soil and the rate of
water supply at periods of peak demand. The upper and lower limits of
the total present and of the rate of supply should be considered with
respect to the whole profile as well as particular horizons. Thus conditions in the portion of the soil profile important to the tree may be excessively wet or dry or else the hydraulic conductivity may be too low
or, less importantly, too high.
While the exact details are not known in the majority of forest situations, literature on P . radiata allows the soil features associated with
moisture stress to be indicated.
1 . Internal Profile Features
In discussing the influence of physical properties of the soil on forest
site quality, Coile (1952) concluded that significant features for satisfactory growth were the volume available for root exploration, the
storage capacity of this volume, and the availability of the moisture in it.
One measure of this was the thickness of the A horizon up to an optimum
of 10 to 12 inches and the imbibitional water value of the B horizon.
FIG. 6. Heights adjusted to age 30 years of trees on various soil types from the
southeast of South Australia in relation to the total depth of the soil profile or the
depth to clay for the very deep profiles. Abbreviations: HSL, Hindmarsh sandy loam;
TFS, Tantanoola flinty sand; MMS, Mount Muir sand; MBS, Mount Burr sand; DYS,
Dry Young sand. (After Ruiter, 1964.)
This latter was correlated with the consistence and texture of the B horizon. Figure 6 expresses the growth of P. radiata as a function of depth for
different soil types in the southeast of South Australia (Ruiter, 1964).
The optimum depth for P . radiata is about 2% feet, in agreement with
Jackson ( 1965) (see Section IV, A, 1 ) . These soil types range from
shallow terra rossas to yellow podzolic soils and then to an extremely
deep pale sand, nutritionally very poor and with moisture stored at
considerable depth. At Mount Crawford, South Australia, Beckmann
(1964) concluded that the features enumerated by Coile were operative
for P. radiuta and that mid-slope podzolic soils gave a surprising degree
of water movement over the surface of the rather impermeable B
horizon, with resultant poor growth. In this way Beckmann has extended
the earlier observations of Boomsma (1949), who showed that internal
profile features gave comparatively dry B and C horizons in some soils,
even during wet winters.
2. Nonwetting Characteristics of Sandy Horizons
In observations on the water-repellent properties of sandy surface
horizons containing less than about 8 percent clay, Bond (1964) showed
that water penetrated into the soil through narrow channels, the intervening soil remaining dry. Such dry zones persisted through the winter
months after the autumn rains and produced bare patches in grassed
fields. This resistance to water penetration was not as great in forests
of P. radiuta as it was in areas of pasture or native vegetation, although
it was still present in the forest. Indications are that these areas of local
drought are of particular importance in young plantations before pine
root systems have developed sufficiently to dominate the site. The water
repellence was associated with fungal mycelia which also promoted
aggregate formation in sands. The maximum aggregation observed on
a solodized solonetz with P. radiata was 2 to 5 inches below the litter
layer (Bond and Harris, 1964).
3. Tree Stress and Summer Drought
Two types of injury have been attributed by Pryor (1947) to drought:
needle cast and death of the tops of trees. Among the numerous reports
of drought deaths are those on shallow stony soils on rises, where
Boomsma (1949) pointed out that major roots often outlived the crown
of the affected trees, but that many fine rootlets died before the crown.
On soils of the southeast of South Australia, during the drought conditions of late summer and autumn in stands where canopy had closed,
the leader has been observed to die back to the next whorl down, and
in severe cases to a lower whorl or even to the whole tree (Lewis, 1957).
This condition has been called autumn brown top. Working on 2-yearold seedlings in drums, Millikan and Anderson (1957) showed that the
disorder was associated with moisture stress on shallow soils during hot
dry periods where there was a high tree density. Lewis and Harding
(1963) associated autumn brown top with summer drought on calcimorphic soils in South Australia and also in part with a variation in tolerance due to genetic factors. Ruiter (1964) found that growth on shallow
terra rossa soils from this region was negatively correlated with soil moisture tension, there being shrinkage of the stem and death of tops or
even of whole trees at tensions above 15 atmospheres. He also found that
moisture tension and growth were not related on deeper soils and that the
root systems on deep sands appeared to absorb water from depths of at
least 20 feet (see Fig. 6). Millikan and Anderson (1957) demonstrated
that moisture stress during winter produced symptoms of both general
wilting and death which were quite different from those found during
summer. Simpfendorfer (1959) concluded that death of the tops of trees
was most likely to occur (1) in years of average or greater than average
rainfall, ( 2 ) in years of severe drought, or ( 3 ) when the rainfall was
slightly below average for the last 1%years. Point (1) involves excess
water and will be considered below.
Johnston (1964) using trenched plots induced drought in 4-year-old
P. radiata trees and showed that moderate turgidity was maintained in
the needles even when soil conditions were very dry, provided that there
was frequent precipitation in amounts sufficient to wet the foliage. This
demonstrated that a supply of moisture as fog, dew, or rain was
necessary during dry summer periods in order to avoid damage on
shallow soils with limited water reserves.
4. Damage Due t o Excess Water
In considering the loss of trees due to excessive soil water, Poutsma
and Simpfendorfer ( 1962) have distinguished between seasonal surface
waterlogging and the presence of free groundwater. The vigor of both
P. pinaster and P. radiata varied significantly with the degree of surface
waterlogging. Free water, moving laterally through the surface horizons
appeared to affect the trees less than when it remained stagnant. Nutrient
deficiencies are often present on wet sites (Fig. 7 ) . An examination of
1,570 sites north of Auckland, New Zealand, by Sutherland et al. (1959)
showed that during a very wet winter when even the better drained
soils were saturated from April to August, partial defoliation or death
by wilting couId not be associated with soil type, relatively wet or Iowlying portions of the landscape, or with other soil physical factors.
Except in a single case of death by “drowning,” the presence of a path-
FIG. 7. The poor growth often found on wet sites may sometimes be improved by addition of
fertilizer; the faster growth rates obtained by this means can, over a period of years, assist water removal from the soil, with a resultant yet further improvement of growth. The photograph shows 8-yearold trees on a wet site with less than 18 inches of light textured soil above the clay horizon. The trees
on the right received 0.75 pound of superphosphate per tree whereas those on the left were untreated.
(From Department of Woods and Forests, South Australia.)
ogen was indicated and the deaths were ascribed to Phytophthma attack,
the amount and severity of the disease being influenced by soil drainage.
By this means rootlets were destroyed at a greater rate than they could
regenerate because of the wet conditions (also see Newhook, 1959).
Observations on the growth of P. radiata at Second Valley, South
Australia, were made from ring widths of various tree sections and
covered a twenty-year period. Some of the sites on lateritic podzolic
soils normally had a perched water table within 12 to 18 inches of the
surface for several months of an average winter. Other sites, on podzolic
soils did not have perched water tables. Decreases in the growth rate
were not observed in excessively wet years on any of the sites, but lower
growth rates were found when the rainfall was less than average (Raupach, 1967), thus demonstrating that temporary waterlogging, but not
drought, can be tolerated.
Highly saline conditions are responsible for tree failure. At Mount
Crawford, South Australia, Woods (1955) found an upper limit of 0.5
percent chloride in oven dry needles. Above this value, foliage was
affected and death could result. Some afflicted trees had chloride contents
as high as 5.0 percent. Carleton (1962) has shown that P . radiata has
a lower salt tolerance than P. thunbergii or P. taeda.
Often sites on poorer soils are unable to support a stand of P. radiata
unless fertilized; this makes the addition of nutrients commonplace for
this species. Apart from soil analysis used as a systematic assessment
tool, some general conclusions can often be drawn from an examination
of the soils and a knowledge of the parent materials from which they
are formed, for planting unknown areas.
In southern Australia for example, the principal soils on which plantations of P. radiata are grown comprise: (1) groundwater podzols; ( 2 )
podzols; ( 3 ) yellow podzolic soils; ( 4 ) red podzolic soils; (5) lateritic
podzolic soils; ( 6 ) meadow podzolic soils; ( 7 ) krasnozems; ( 8 ) terra
rossa. The first six are formed on a wide range of parent materials, the
krasnozems are practically restricted to basalt, and the terra rossa to
limestones of both dune and sedimentary origin. The parent materials
of the first three include large areas of dune sands of coastal origin now
stranded varying distances inland and leached to different extents, with
depletion of their original content of calcareous and other minerals.
These leached sands are, with the soils formed on them, necessarily low
in nutrient status.
The groundwater podzols and the podzols have failed as plantation
soils where they are extremely acid, as in parts of Tasmania and Victoria;
but even where they are above pH 4.5 in the surface horizon, they still
grow inferior stands unless phosphorus and zinc are added. The yellow
and red podzolic soils generally produce acceptable stands but also give
varying responses to the above nutrients. Lateritic podzolic soils generally
have produced inferior trees, but this may be overcome by heavy applications of fertilizer. The results on meadow podzolic soils are variable,
largely reflecting the moisture status of the site. Krasnozems usually
produce vigorous trees whereas plantations on terra rossa soils have variations in growth and irregular stocking which are usually associated with
their great variation in depth and stoniness.
Deficiencies are thus apt to develop on soils that are heavily leached
or have parent materials of extraordinary deficiency in one or more of
the essential nutrients. Such is the situation for the nutrients phosphorus,
calcium, magnesium, potassium, zinc, and boron in P . radiata, and
responses have been obtained by addition of fertilizer. Factors limiting
root exploration, described under Section IV, B, may also give a poor
nutrient status (Fig. 7 ) . Of the above soils, lateritic podzolic soils and
the deeper sands associated with podzols and yellow podzolic soils have
received most attention from investigators because these represent contrasting situations in which intense fixation and leaching of nutrients
On lateritic podzolic soils in Western Australia (Kessell and Stoate,
1938; Anonymous, 1958) and South Australia (Boomsma, 1949; Beckmann, 1964; Raupach, 1967), phosphorus deficiency has invariably been
reported together with indications of the importance of nitrogen and
trace elements. Field responses have been obtained to zinc, nitrogen,
and possibly copper and nickel (for the latter three, see Raupach, 1967).
On the deep siliceous sands, plants have brought nutrients from
deeper horizons to the surface, where concentrations of nutrients have
been built up as residues insofar as has been permitted by continued
leaching. These may still be insufficient to allow tree growth (e.g.,
Humphreys, 1964) since the tree may not be able to explore the deeper
horizons which have been thus depleted, (e.g., Gentle et al. 1965).
However, sometimes relatively rich underlying strata allow substantial
nutrient supplies to be accumulated at the surface. This has been shown
by Tiller (1957a,b) for deposits of volcanic origin which sometimes
underlie soils on dune sands for the elements manganese, copper, cobalt,
zinc, nickel, and gallium; as instanced by Tiller, access to the rich strata
can be prevented by hardpan or other impenetrable layers in the soils
Ectotrophic mycorrhiza fungi have been recognized to associate with
P . radiata, and the natural occurrence of such fungi has been studied
in South Australia (Bowen, 1963) and Victoria (Marks, 1965). Difficulties are met in the isolation of mycorrhiza fungi, and this has led to a
tendency among forestry workers to look at the fungi producing fruiting
bodies. Not all the fungi involved necessarily produce fruiting bodies,
and in any case the degree to which a fungus is a symbiont of the stand
may often bear little or no relation to its fruiting capabilities. Bowen
( 1966) has given an annotated bibliography of microbial interactions
with trees, and this includes a number of references to P . radiata. In
considering mycorrhiza inoculation in forestry practice, Bowen ( 1965)
has indicated that there are possible differences in mycorrhiza characters
and that by application of the ‘‘correct” fungi, the normal characteristics
of the root may be changed with advantage. He observed that different
mycorrhiza fungi from P . radiata have been found to give large differences in the uptake of 32P in short-term laboratory experiments, the
poorest performer being the commonest type in South Australian soils.
Clode (1956), however, while finding greater uptake of 32Pin mycorrhiza plants, observed more growth in P . radiata which was not inoculated. Much remains to be done on the way in which specific mycorrhiza fungi may assist the plant by such means as nutrient uptake and
protect it against factors retarding growth, e.g., pathogen attack.
Eventually it may be possible to establish a selected advantageous suite
of fungi in forests as a regular practice.
Recently developed bioassay methods ( Harris, unpublished work)
show that the growth of P . radiata is accompanied by a greater buildup
of toxic factors in the soil than occurs under pasture or cereal crops.
Acute levels of toxicity can be developed which could cause the second
crop of pines to be unthrifty.
Biological mobilization of potassium from biotite, muscovite, and
microcline by seedlings of five tree species including P . radiata showed
that the latter did not behave any differently from the others although it
gave the highest yields (Voigt, 1965). The weathering intensity was a
function of the total root surface and was probably also related to the
activity of microorganisms associated with the surface.
Assessment of limiting Factors
While the most complete method of investigation is to establish a
balance sheet for the nutrient cycle including the return by litter as
done by Ovington (1959) for P. syluestris, efforts with P. radiata have
largely but not entirely gone into nutrient surveys of either existing
stands or unplanted sites. This has been with the practical end of providing a guide to planting and fertilizer procedures. In the short term,
it is the relationship between the abnormal or unthrifty tree and the
normal tree which has been examined to find what improvements can
be made. Details on nutrition as such and on the interaction between
the soil and the tree should be first worked out on normal trees.
Over a series of years, Will and his co-workers in New Zealand have
investigated various aspects of nutrient distribution of P . radiata on
pumice soils, and these studies are of value in establishing the status
of normal trees.
The mineral nutrient content of trees has been investigated for
nearly a century, and the use of analysis of plant tissue as a diagnostic
technique for evaluating site fertility continues to be discussed and
reviewed (e.g., Viro, 1961; Tamm, 1964).
Work on tissue analysis of P. radiata under forest conditions has been
restricted to foliar analysis except for that of Will (1962), who suggested
the use of wood for phosphorus and later showed (Will, 1965) that both
wood and bark analysis were promising for detecting deficiencies. The
total amount of work done has been rather limited, perhaps as a result
of the large variation within the species including differences in the
number of whorls from tree to tree as discussed under Section 111.
Even though a deficiency may be sufficiently obvious for it to be
recognized from its symptoms, this should be checked by analysis before
undertaking field experimentation. This checking both confirms the diagnosis and indicates whether a second nutrient is involved in deficient or
1 . Requirements for Foliar Analysis
The technique of foliar analysis requires that there should be a significant relationship between growth and the foliar nutrient content
when the supply of the nutrient to the tree is progressively restricted.
The measure of growth may be a total over the life of the tree, e.g.,
height, or better still an increment over the current or the previous
growing season. The foliar nutrient contcnt has been measured in various
ways in pines-for example, as weight of nutrient per needle in the wet
or dry state or as percentage of the oven dry material. The last of these
has been used extensively for P . radiata and will be considered here.
Further requirements are: (1) to have an estimate of errors within
and between trees and between forests; ( 2 ) to sample from material of