#
__'_,__.
___'_._.
__———..——
-—_,_-—;_-—=—"
#
_____——'_;:—‘
_______..
:—
f,”
-___——_=—:—"
#
_______._.
fl
5-."
f___._
#
__._.—-
—____.__."'-=
p;
—-———_
’
___-———____.
__——-__=."'J
’
—-__:_-——'—"
’____
_____.__.
a
E

rHEF‘M

  
   
   

  

’ LIBRARY ‘
MichiganStaDc
University

This is to certify that the

thesis entitled

FACTORS AFFECTING JACK PINE REGENERATION

ON THE SAND PLAINS OF NORTHERN LOWER MICHIGAN
presented by

John David Marshall

has been accepted towards fulfillment I
of the requirements for ’

M.S. degree in Foregtry

/ Major professor E

Date ”4/? '82

0-7639

 

 

 

OVERDUE FINES.

25¢ per day per item

RETURNING LIBRARY MATERIALS:
Place in book return to runove
charge from circulation records

  
 
     

 
 

e ~qu A

"STATE Wlmfln
'L\ 7 .41"

MAGIC 2

Nfi‘Z‘nggg

xvr-swr‘aw'

I :(‘fi

; Q “'92 19:95“ ‘ 61' 0
AU ‘2}:32'?‘ Efl‘flfiaalflfl

 

NOV 1- 4 '99“.
V
Mammal

WW mafia

 

 

 

FACTORS AFFECTING JACK PINE REGENERATION ON THE
SAND PLAINS OF NORTHERN LOWER MICHIGAN

By

John David Marshall

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Forestry

1980

ABSTRACT

FACTORS AFFECTING JACK PINE REGENERATION ON THE
SAND PLAINS OF NORTHERN LOWER MICHIGAN

By
John David Marshall

Factors exercising greatest control over success of planting
and natural regeneration techniques were identified. Seedling stocking
was related to physical and chemical soil properties and to understory
and overstory vegetation.

Partial cut areas studied included shelterwood, seed tree,
and strip cut areas. Seedling stocking averaged 1346 trees/hectare.
Seedling numbers were positively related to bearberry and sweet-fern
cover, and inversely related to overstory basal area and several
measures of B horizon fertility.

Burned areas studied included prescribed burns and wildfires
through standing trees. Seedling stocking averaged 7123 trees/hectare.
Regeneration was inversely related to sedge cover and depth to mottling
and positively related to bearberry cover, pre-fire stand density, and
soil profile thickness. Post-fire seedling establishment was found to
reach a maximum two to five years after burning.

An average of 75.4% survival was found on the plantations, with

no significant differences among the site types studied.

ACKNOWLEDGMENTS

I would like to express my gratitude to my major professor,

Dr. J. B. Hart, for his invaluable advice and help in the design,
administration, and writing of this study. I would also like to thank
the other members of my graduate committee, Dr. D. I. Dickmann and

Dr. G. Lemme, for their advice and encouragement. Thanks are also due
to Drs. D. Mokma and C. Ramm for their assistance in the design and
analysis of this study.

I am thankful to the Forest Management Division of the Michigan
Department of Natural Resources for funding this research and for their
assistance in the location of study areas. The cooperation of the Wild-
life Division is also appreciated. Thanks also go to the Department of
Forestry and to the U.S. Forest Service for the use of their laboratory
space and equipment.

I am grateful to Jackie Myrold, for the care with which she
supervised the soil analysis. My appreciation also goes to Don Dekker,
Paul Doescher, Bart Hoag, John LaLonde, Pat McMullin, Dennis Pawl,

Dan Stouffer, and Tom Nutt, who at various times were responsible for
much of the data collection and analysis.

Finally, I owe great thanks to my wife, Beth, for her help in
preparing this manuscript and especially for her uncomplaining encour-

agement during the last harried month of its preparation.

ii

TABLE OF CONTENTS

Page
LIST OF TABLES ......................... v
LIST OF FIGURES ........................ vi
INTRODUCTION .......................... l
LITERATURE REVIEW ....................... 3
Jack Pine Ecology ...................... 3
Distribution and History .................. 3
Genetic Variability .................... 5
Seed Dispersal ....................... 6
Germination ........................ 8
Seedling Development .................... ll
Successional Role and Associated Species .......... l2
Productivity ........................ l6
Jack Pine Silviculture .................... 18
Site Preparation ...................... l8
Seeding Techniques ..................... 20
Planting Techniques .................... 24
Soil Properties of Burned Jack Pine Sites .......... 25
METHODS AND MATERIALS ..................... 31
Plot-Selection Criteria ................... 3l
Plantation Data Collected .................. 35
Fire and Partial Cut Data Collected ............. 36
Common Data-Collection Techniques .............. 37
Laboratory Analyses ..................... 39
Data Analyses ........................ 4l
Description of Study Locations ................ 4l
RESULTS AND DISCUSSION ..................... 47
Site Factors Controlling Regeneration Success on Partial
Cut Areas ......................... 47
Partial Cut Regeneration Patterns .............. 55
Site Factors Controlling Regeneration Success on Burns . . . . 55

iii

Page

Burn Regeneration Patterns .................. 66

Plantation Regeneration Success ............... 69
SUMMARY AND RECOMMENDATIONS .................. 73
LITERATURE CITED ........................ 80
APPENDICES

A. SCIENTIFIC NAMES 0F SPECIES ENCOUNTERED IN THIS STUDY . 89
8. UNIT CONVERSIONS .................... 93

iv

Table

TO.

IT.

LIST OF TABLES

Understory Species Characteristic of Various Soil

Moisture Regimes ....................

Legal Descriptions of Plantation, Fire, and Partial
Cut Locations Sampled in Jack Pine Regeneration

Study .........................

Sunmary of Partial Cut Area Seedling Numbers and Selected

Variables by Location .................

Significant Correlations Between Site Factors and

Regeneration Success ..................

Multiple Regression Equations Developed for Partial

Cut Plots .......................

Significant Differences Between Partial Cut Plots

and Burned Plots ....................

Summary of Fire Area Seedling Numbers and Selected

Variables by Location .................

Significant Correlations Between Site Factors and

Regeneration Success on Burned Plots ..........

Multiple Regression Equations Developed for Burned Plots .

Variables Significantly Different Among Texture and

Drainage Strata in Plantation Study ..........

Plantation Data Summarized by Location and by Stratum

Page

15

34

48

50

51

53

60

61
62

70
71

LIST OF FIGURES

Figure . Page

1. Locations of Plantations, Burned Areas, and Partial
Cut Areas Studied .......... . ........ 33

2. Extent of Grayling-Rubicon-Roselawn Soil Association
in Study Area ..................... 42

3. Age Distributions of Surviving Seedlings on Partial
Cut Areas ....................... 56

4. Average Age Distributions of Surviving Seedlings on
Burned Areas and Partial Cut Areas .......... 57

5. Age Distributions of Surviving Seedlings on
Burned Areas ..................... 67

vi

INTRODUCTION

The importance of jack pine.l has steadily increased in the
Lake States since the turn of the century. Once regarded as a weed
tree, the introduction of the sulphate process has made it one of the
most desired pulp species. Stumpage prices have recently risen
sharply due to the short supply of softwood fiber. This softwood
shortage is expected to continue into the near future, due to the dif-
ficulty of establishing conifer stands in this region.

Jack pine is also highly valued by wildlife managers because
of its essential role in the life cycle of the Kirtland's Warbler.
This endangered species nests only in stands of dense young jack pine.
The warbler's numbers have decreased since the advent of fire control.

Regeneration of jack pine has long been a problem. Clearcuts
nearly always fail to regenerate without additional treatment, and
what little regeneration does appear is usually of low-value hardwoods.
During the l930s, large acreages of jack pine were planted by federal
work programs. Once these programs were dismantled, however, the high
cost of planting forced foresters to find less expensive means of
regenerating this species, including various kinds of partial cuts,

prescribed burns, broadcast seeding, and direct seeding.

 

1Scientific names of species encountered in this study are
listed in Appendix A.

This study was designed to identify the site factors controlling
the regeneration success of burns, partial cuts, and plantations on the
outwash sand plains of the Grayling-Roscommon area in northern lower
Michigan. These sites, which are extensive in the area, had been iden-
tified as being particularly difficult to regenerate. It was believed
that once the controlling factors on these sites were identified, site
preparation techniques might be adjusted or changed to deal with them.

A secondary objective of this study was the development of
criteria to be used to decide which of the various regeneration tech-
niques to use on a given harvested area. These criteria were to be

developed as guidelines to be used by field foresters.

LITERATURE REVIEW

The literature on jack pine is very extensive. To make it
more understandable, this review is divided into ecology and silvi-
culture. The jack pine ecology section considers the characteristics
of the species, more or less independent of management. The jack pine
Silviculture section considers the various techniques used in the
management of this species. Both these sections emphasize regenera-
tion. A final section, soil properties on burned jack pine sites,
deals with the effects of burning on the dry, infertile, sandy soils

typically occupied by jack pine.

Jack Pine Ecology

 

The ecology of jack pine is an extremely interesting subject,
due in large part to its dependence on fire and its ability to occupy
the driest, least fertile sites available. This section of the lit-
erature review concerns itself with the characteristics of this species,
including the role it plays in the ecosystem and the adaptations allow-
ing it to fulfill that role. This includes its distribution, history,
genetic variability, seed dispersal, germination, seedling development,

successional role, associated species, and productivity.

Distribution and History
Jack pine is a short-lived, small to medium tree with the most

northerly and one of the most extensive ranges of any of the North

3

American pines (35). Found throughout the northern Lake States, it
is common or abundant in central Michigan, central Wisconsin, along
sand dunes bordering Lake Michigan, and north and west of Lake
Superior (78). Commercial stands in the U.S. are restricted to the
Lake States (47).

Jack pine grows in areas with cool to warm summers, very cold
winters, and low rainfall. Across the majority of its range, annual

2 and the annual frost-free

precipitation varies from l5 to 35 inches,
period lasts from 80 to 120 days (34, 35, 78). Summer droughts commonly
occur in the Lake States and through the western part of the range (78).
One of the unique features of this species is its ability to
grow on very dry and infertile soils incapable of supporting almost
any other tree species in this region (35, 36). These soils are often
coarse or medium sands, gravelly soils, or shallow soils underlain by
bedrock (35, 36). Fine sands, sandy loams, loams, and clay loams may
also support jack pine stands, although severe disturbance is usually
required for establishment on these finer-textured soils (35, 36, 79).
It is found only rarely on poorly drained soils (35, 79).
During the last glaciation, jack pine was apparently found in
an extensive glacial refugium in the central Appalachians (35, Bl).
The species moved north and west behind the glaciers as they receded

(l3). Pollen analysis in northern Minnesota bogs has shown that jack

pine became abundant in that area about 9,000-l0,000 years ago (13).

 

2All data originally reported in English units of measurement
will be similarly reported in this review. See Appendix B for English-
metric conversions.

Charcoal is also present in these peat layers, indicating that fires
were occasionally sweeping through this area at about the same time
(l3, 92). Soil development and vegetative succession in the ensuing
time have apparently restricted the species to those areas too dry or
too frequently burned for the more demanding species belonging to
later successional stages. That fire frequency and poor soils control
jack pine distribution is also suggested by reports from a botanical
expedition that crossed the northern lower peninsula of Michigan in
1888. They found extensive stands of even-aged jack pine on the sand
plains. Evidence of fire was always present in these stands (98).
Utilization of the species was rather poor during the logging
of the old-growth red and white pine stands in the late nineteenth
century because of the high merchantability standards of the time.
The intense slash fires that followed this logging killed the few
remaining red and white pine seed trees and regeneration and created
nearly ideal conditions for dispersal and germination of jack pine
,seed (36, 48). Following these fires, jack pine became dominant on
many sites on which it had formerly been only a minor species (36).
As the second-growth stands matured, the value of the species increased,
better utilization began to occur, and regeneration of the species

became a problem (45, 46, 47, 48, 58, 94).

Genetic Variability

 

Genetic variability within this species tends to be clinal
through the western part of the range because stands are often con-

tiguous and selection has been primarily in response to climate. In

the extreme eastern range the disjunct populations show more random
variability (35). Populations from the Lake States (including all of
Michigan except the eastern upper peninsula) tend to be taller,
broader-crowned, more frequently serotinous, and lower in cone produc-
tion than the average for the species (51). In general, provenances
from the northern lower peninsula seem to be the fastest growing,
while the more northern provenances (including those from the upper
peninsula) tend to be slower growing and more’frost hardy (35, 5l, 80).
Arend et al. described four ecotypes in the Lake States. The ecotype
from the western lower peninsula of Michigan was the fastest growing,
that from the eastern lower peninsula was second fastest, the Minnesota
and Wisconsin ecotypes were variable, but generally below those from
Michigan's lower peninsula, and the upper peninsula ecotype was the
slowest growing (20). Genetic variability has been shown to exist in
susceptibility to the sawfly, the jack pine budworm, and pine-oak rust
(20, 79, 80). Lodgepole pine is a closely related serotinous species
found in the Rockies and known to hybridize with jack pine where their

ranges overlap (6l).

Seed Dispersal

 

Cameron (29) describes the serotinous nature of the cones of
jack pine. Serotiny refers to the cones' ability to hold the seed
beyond ripening. The cones are sealed by avapor-resistant resinous
material at the tip of the cone scales. A temperature of at least
50°C (l22°F) will soften this resinous material, allowing the cones

to open. Higher temperatures will open the cones more rapidly. Once

the resin has softened, exposure to low relative humidity will reflex
the cone scales, releasing the seed. The cones do not reclose upon
being returned to higher humidities.

The closed cones have very low heat conductivity, which pro-
tects the seed held inside when the cone is exposed to heat (29).

Thus in a fire, the seed survives even though the parent tree is killed,
unless temperatures are high enough for a sufficient time to cause cone
ignition and destruction of the seed (36, 48). The ungerminated
exposed seed is rather heat tolerant also, as it will not be killed
until it is exposed to a temperature high enough to ash the wing and
crack the seed coat (29).

Cone serotiny varies genetically (48, Bl). Nonserotinous cones
are more common from seed sources from the southern portion of jack
pine's range (48). Serotiny is also dependent on age and on stand
density (35). In mature stands in Saskatchewan, 10% of the cones were
open, while almost all ripe cones on 7- to lO-year-old, open-grown
trees were open (35). Laboratory tests comparing cones from trees
less than 10 feet in height to trees greater than 10 feet in height
under identical temperature and humidity conditions showed the cones
from the smaller and younger trees were more easily opened (54).
Another study in Manitoba showed that 20-year-old stands annually
dispersed approximately 36,750 seeds per acre, over four times the
amount dispersed in 40- and 60-year-old stands (35). Seedfall in open
stands often exceeds that in dense stands (35, 48).

Serotinous jack pine stands are able to store many years'

production of seed. Seeds may be locked inside the cones for up to

25 years(24). Stored seed crops have been reported from "several
pounds" (36), to 12.9 and l3.5 pounds per acre (48). Other studies
have reported l,540,000 seeds per acre in southeastern Manitoba and
from 6,000,000 to 8,000,000 per acre in a mature stand in Ontario
(35), and up to 9,000,000 seeds per acre in Michigan (46). Some
studies indicate that the viability of this older, stored seed tends
to be somewhat lower than recently ripened seed (25, 46), while other
studies have shown no effect (48, 71). Older seed may also be some-
what slower to germinate (35). Most researchers have concluded that
viability is high enough to make a significant contribution (25, 46,
48, 77).

Seedfall may begin almost immediately after a fire and may be
completed in as little as 3 to 15 days, depending on fire intensity,
air temperature, and humidity (26, 46, 48). Seedfall from logging
slash may take two or more years, usually being highest during the hot,
dry' periods of mid-summer (35, 47). Rodents, grackles, blackbirds,
robins, and other birds have been reported to eat large amounts of

seed (48).

Germination

 

Dispersal of large amounts of viable seed is the first require-
ment for regeneration. The seed must also fall on a seedbed able to
provide the conditions necessary for germination (22, 58). Seedbed
moisture levels are probably the most common factor limiting germina-
tion (2l). The seed, released from the cone at a moisture content of

l2-l5%, must absorb enough water to reach at least 53-54% moisture

content before it will begin to germinate (29). Seedbed moisture
levels are dependent upon the amount and distribution of rainfall,
shading, evaporation rate, depth to the water table, soil texture,
seed covering, season of dispersal, seedbed type, and competing vege-
tation (5, 21, 22, 35, 36, 38, 39, 50, 54, 57, 58, 87, 95).

The factor most important in controlling seedbed moisture is
seedbed type. Mineral soil and humus serve as the most favorable
seedbed types for jack pine, and deep undisturbed litter serves as the
poorest (5, 35, 36, 48, 57). This is partially due to the low heat
conductivity of dry litter, which results in higher maximum surface
temperatures than on mineral soil. Litter is also less able to provide
the moisture needed by the germinating seed because much of the water
it absorbs is held within the cellulose fibers rather than on the sur-
face of the particles, as in mineral soil and humus. Also, the textures
of mineral soil and humus tend to be finer, allowing closer packing
around the seed, so that capillarity may keep up a steady water supply
to the seed during imbibition (21, 48, 57). Shallower litter layers
may improve regeneration by allowing the seedling easier access to the
mineral soil below (39). A thin (less than 0.5 inch) soil covering
may also improve germination (50).

Temperature conditions can control germination. One study
(19) has shown that field germination begins whenever daily maximum
air temperatures exceed 64°F (18°C). Eyre and Lebarron (48) found
that field germination occurred when the 10-day mean maximum tempera-
ture was greater than 65°F (18°C). Germination under controlled con-

ditions was found by Ackerman and Farrar (15) to be three days slower

10

at constant temperatures of 60°F (16°C) than at various combinations of
60°F (16°C), 70°F (21°C), and 80°F (27°C) over 24-hour periods.

Light exposure also controls germination since jack pine seed
has a minimum light exposure requirement of two to four minutes to
break seed dormancy. This light exposure will not break dormancy
unless the moisture content of the seed is greater than 10 to 20% by
weight. This breaking of dormancy is not reliable at temperatures
below 60°F (16°C) (2).

Jack pine seed viability is usually above 70% (5, 35, 67, 71).
with lower values due to inferior cleaning techniques, to seed not
given its light requirement, or to slower germination of older seed
(2, 25, 35). Seeds exposed to fire have been shown to germinate bet-
ter than unburned seed (5, 71).

In a normal growing season in Manitoba, germination begins as
early as mid-May and ends, in most cases, by the end of June, with
sporadic germination through the rest of the summer (35, 71). However,
this generalized germination pattern is seldom described in case
studies, because while the seed of most conifer species germinates the
spring following ripening, jack pine seed germinates whenever condi-
tions are suitable (3). For this reason and because of the slower or
less complete germination of older seed, reports of 99% germination
(32) are interspersed with findings of much lower first-year germina-
tion percentages (25, 35). Beaufait (26), studying post-fire jack pine
regeneration, describes "apparent dormancy" of seed resulting in large

numbers of germinants in the second and third growing seasons.

ll

Seedling Development

Jack pine seedlings are particularly vulnerable to mortality
during the first growing season following germination (5, 57). One
study found that, of the seedlings germinating the spring after a
fire, 75% survived June, 56% survived July, and 42% survived August,
with 38% surviving through the end of the second growing season.

The succulent first-year seedlings are especially susceptible
to heat and drought injury (16, 21, 32, 35, 48, 57, 63, 87). A tem-
perature of 49°C (120°F) is considered lethal for first-year seedlings
(87). Another study reports that temperatures in excess of 122°F
(50°C) for more than two hours will cause injury (48). Heat and drought
injury are more likely to occur on dry, sandy sites and near mid-
summer (18, 32, 35, 87). Dry sites may be higher risks because of the
lack of shade cover (48) and because surface soil moisture is not
available to evaporate and cool the site (21). Removal of litter,
shading, and early spring germination reduce the temperatures to which
newly germinated seedlings are exposed (6, 21, 39, 48, 59, 67, 79).
Other major causes of mortality include herbivores (35, 48, 57),
damping-off fungi (especially under deep slash piles) (54), and high
salt concentrations from ash and from charcoal (5, 93).

Jack pine grows relatively slowly during its first three to
four years (5, 35). During this period a taproot develops to about
50 cm in length and penetrates below sod competition and possibly into
a good source of water and nutrients (35), often a spodic horizon
(85). After approximately four years, lateral root growth in the

A horizons becomes dominant over taproot growth. At this time, the

12

seedling begins to grow very quickly and the topzroot ratio rapidly
increases (35). Interestingly, jack pine growth is greatest with
less than full sunlight for the first four years, after which time
100% sunlight becomes optimal (59).

Early growth rates are much higher on mineral soil and burned
seedbeds than on undisturbed litter (54). Competition and water
availability influence growth rates at this stage, with best growth
on a site dry enough to support less vigorous competition and yet
moist enough for rapid jack pine growth (35, 88). However, moister
sites have better growth rates after crown closure (37).

According to Rudolf (79), the most common causes of mortality
in established seedlings older than age one are heat, drought, and
competition. Herbivores (35, 48, 77) and deformed roots in planted
trees (77) are also important. Heat is most damaging to smaller trees
(77), with little mortality in trees greater than 1/2 inch in diameter
(16). Heat damage on small trees can be identified by a layer of
discolored cambium just above the soil surface (16). Heat often pre-
disposes seedlings to drought damage and vice versa (16, 87).

Successional Role and
Associated‘Species

 

 

Jack pine is commonly recognized as a fire species because its
silvical characteristics make it "ideally suited to regenerating after
wildfire" (36). The advantages of cone serotiny on frequently burned
sites, discussed earlier, and the beneficial effects of burning on

seedbed conditions explain the common occurrence of this species in

13

pure or nearly pure even-aged stands throughout most of its range
(35, 36, 48).

Jack pine is able to function as a pioneer species most effec-
tively on dry, infertile sites. Wilde (102) found it to have the
lowest nutrient requirements of any of the Lakes States conifers. Its
drought resistance was demonstrated by Pereira and Kozlowski (74), who
found that, relative to red pine, jack pine stomates close rapidly in
response to soil water depletion. Its internal resistance to water
flow is also relatively high (74). Because of these and possibly
other adaptations, jack pine is able to survive on coarse sands and
shallow outcrop soils, where conservation of moisture is a necessity,
and where few other tree species can survive, except in the under-
story.

The individuals of species commonly occurring in association
with jack pine must have means of surviving frequent fires. Ahlgren
(10) found that the majority of plants reproducing after a fire in
northeastern Minnesota were of vegetative origin. Darlington (42)
found that 95% of the common plants of the jack pine plains of Michi-
gan were perennials with deep roots or rootstocks adapted to severe
conditions of drought or of surface burning. For example, bracken
fern, the blueberries, and aspen survive a fire predominantly by
underground structures which sprout following a fire (10). Ahlgren
(8, 9) found that the grasses and sedges, and many other post-
disturbance pioneer species, become much more common and produce a

great deal of seed after the removal of an understory. This seed is

l4

stored in the forest floor and germinates at the time of the next
disturbance.

In the boreal forest region, which is north of the study area,
jack pine is commonly associated with trembling aspen, paper birch,
white spruce, black spruce, and balsam fir (36, 48). It is succeeded
by a mixture of spruce and fir on all but the driest of sites. Pure
or almost pure jack pine stands can be found on thin soils overlying
rock outcrops; on glaciofluvial plains, eskers, and kames with sandy
and gravelly soils; on fluvial sand terrances along rivers; on sand
dunes; and on upland sandy and loamy tills. It occasionally occurs
on lowlands of all textures (37). On better sites with silty sands,
loamy sands, and loams, jack pine is often mixed with varying propor-
tions of trembling aspen and white birch (37). Black spruce frequently
forms an understory on rocky upland tills and in depressions on low-
lying flats (37). White spruce and balsam fir are commonly found in
the understory on all soils (37).

In the Lake States, common associates include those listed
above as well as northern pin oak, bur oak, red pine, and eastern
white pine. In this region jack pine is found primarily on sandy sites,
and may be replaced by red and white pine, followed by a mixture of
hardwoods dominated by sugar maple, basswood, and northern red oak
(48). Cayford et a1. (35) suggest that the red and white pine stage
is succeeded by a spruce-fir mixture in the Lake States. In north-
eastern Minnesota, where jack pine is found on finer-textured soils

as well as sandy soils, jack pine may also be succeeded by a dense

l5

undergrowth of alder, hazel, and beaked hazel, which is gradually
replaced by a spruce-fir mixture (87).

The Canadians have classified the moisture regimes of differ-
ent jack pine sites based on understory vegetation. Table 1 gives a
general summary of this system (adapted from 35). Chrosciewicz (37)
found that the water table remained below seven feet throughout the
year on the dry and very dry sandy sites, but in the fresh and moister
than fresh sands, the depth to ground water varied between three and

six feet.

Table l. Understory species characteristic of various soil
moisture regimes

 

 

Moisture Regime Characteristic Species
Very dry Reindeer lichen
Dry Bearberry, Blueberry spp., and Wintergreen
Fresh Twinflower, False lily-of-the-valley,
Bunchberry, Bracken fern, and scattered
shrubs
Moist Labrador-tea, Leatherleaf, and Raspberry

 

Using the Canadian system as a starting point, it is apparent
that a majority of the sites studied by Ahlgren (5, 6) in northeastern
Minnesota are fresh or moist, as evidenced by the predominance of
bunchberry, labrador-tea, bracken fern, raspberry, false lily-of-the-
valley, twinflower, and many shrubs. In contrast, Beaufait (24) found

a dry portion of an outwash plain in northern Michigan to be dominated

16

by bearberry, blueberries, bracken fern, and lichens, among other
species indicative of drier conditions. Thus, where site descriptions
in a study are incomplete, one can predict the character of the site
by examining the understory vegetation. Also, this points out the
fact that all jack pine sites are not alike and caution must be used

in extrapolating research results from other regions.

Productivity

 

Pawluk and Arneman (73), working in northeastern Minnesota and
northwestern Wisconsin, concluded that fertility and especially water-
holding capacity were important determinants of site quality as
measured by site index. They used regression analyses to relate jack
pine site index to various site factors in eight soil series with
textures ranging from sandy loams to sands. They found highly sig-
nificant curvilinear relationships between site index at 50 years and
texture, with site index increasing as the sum of the percentages of
very fine sand, silt, and clay of the A2 and B horizons increased.
They also found a curvilinear relationship between site index and the
available moisture capacity of the A00, A2, 82, and B3 horizons, with
higher moisture capacities increasing site productivity. Significant
linear relationships were found between site index and the total
exchange capacity of the A2,82, and B3 horizons, exchangeable potas-
sium of the A2, 82, and B3 horizons, and the percent base saturation
of the A00, A0, A2, and 82 horizons with increases in any of these
variables causing an increase in site index.

Wilde et al. (103), working in jack pine plantations through-

out the state of Wisconsin, related site quality in terms of average

17

annual growth to physical and chemical soil properties of the surface
six inches. They found that nearly all low-quality sites, which they
defined as those with less than 13 inches of average annual height
growth or site indices less than 53 feet, were deficient in at least
one nutrient. .They found the height to age ratio to be most strongly
determined by organic matter content, followed by available phosphorus,
silt plus clay content, and available potassium. Wilde (102) then
published minimum levels of pH, % silt plus clay, % organic matter,
exchange capacity, total nitrogen, available phosphorus, available
potassium, exchangeable magnesium, and exchangeable calcium for a
site with a site index of 53 feet.

Jameson (55) and Chrosciewicz (37) have related jack pine
productivity to the Canadian system of classifying moisture regimes.
Jameson developed site index curves for the various moisture regimes
on fresh tills through very dry dune sands. He found the average
site index on dry sands, the site type most similar to the Grayling
sand (typic Udipsamment, mixed, frigid) to be 48 feet.

Chrosciewicz (37) looked at pure or almost pure jack pine on
deep siliceous very fine sands, fine sands, and medium sands and
examined the effects of higher percentages of basic rock particles in
the sands. He found that site indices were highest on moist sites,
decreasing on wetter sites and on fresh sites, lower still on dry
sites, and lowest on very dry sites. He also found that site indices
decreased in going from very fine sands to fine sands, and from fine
sands to medium sands. Sands with 30 to 40% basic intrusive and

effusive rock particles also had higher site indices than sands with

18

less than 10% basic rock particles. The effects of texture and basic
particles were especially evident on the drier sites. Climatic region
was also shown to have a significant effect on site productivity. With
few exceptions, the same results were found for diameter growth.
Bensend (21) found that jack pine seedlings showed greatest
height and weight with soil nitrogen concentrations of 200 to 250 ppm.
Root weight increased as concentrations increased to 100 ppm, beyond
which there was little change. Seedlings grown under optimum condi-

tions were as drought resistant as those grown on deficient soils.

Jack Pine Silviculture

Jack pine regeneration has been a problem for foresters for
some time (35, 48). This species lacks the sprouting ability that
makes aspen regeneration easy to obtain. Yet, it does not regenerate
as poorly as red and white pine (7). In fact, it regenerates very
well just often enough to allow foresters and wildlife managers to
believe that obtaining natural regeneration should be a fairly simple
task once some reliable technique can be found. The following sections
will review the major natural and artificial regeneration methods that
have been attempted and discuss their risks, advantages, and disad-

vantages.

Site Preparation

The success of both seeding and planting is highly dependent
on site preparation (41, 57, 58, 78). Site preparation involves the
management of site factors associated with soils, vegetation, and

slash. Rudolf (78) goes as far as to say, "Generally speaking, field

19

planting without some sort of site preparation is merely so much
wasted effort." At least two to three years of freedom from severe
competition are needed for seedling establishment (78). This is
especially important on dry sites (35, 87).

Furrowing and scalping are two methods commonly used to pro-
vide the conditions needed for seedling survival (34, 76, 78, 98).
Furrowing usually provides for faster growth (22) and better survival
than scalping (54, 87). Furrowing should be just deep enough to turn
back the surface root mat (36). Machinery used includes front-
mounted equipment, such as the V-blade, and pulled equipment, including
fireline plows and disks (36). The Athens disk is often recommended
for use on coarse-textured soils (35, 41, 46).

Burning is also used for site preparation. It can be useful
for slash reduction, which reduces planting costs and fire hazards
(3, 97). In many cases, however, the primary purpose of prescribed
burning is to reduce the depth of the surface organic horizons prior
to seeding (35, 38). These seedbed improvement burns must take place
under dry conditions to bring about sufficient litter reduction (35,
38, 39, 97). The subsequent seeding is most successful on those micro-
sites having the shallowest litter depth (5, 35, 39, 54, 97). The
maximum litter depth considered a "high quality seedbed" is 0.5 inches
(39). This litter reduction must also occur when burning slash under
seed trees (26, 35).

Mechanical site preparation is presently more common than the
use of fire (36). The advantages of mechanical site preparation include

that it can be done almost independent of weather conditions, it is less

20

labor-intensive, and it is less risky than burning (36). Some of these
same advantages apply to the use of herbicides, which have been shown
to improve growth and survival of red pine on dry sandy sites in

Wisconsin (56).

Seedingglechniques

 

The least expensive methods of regenerating logged jack pine
stands involve the dispersal of seed across the area (48). The tech-
niques used to accomplish this may involve dispersal of slash containing
seed, seed produced and released from nonserotinous overstories, seed
released by slash burning under serotinous seed trees, or seed brought
in from off the site and dispersed by ground or aerial application
methods.

Clearcuts on which the slash is untreated or windrowed almost
always fail to restock sufficiently, despite the large amount of viable
seed frequently left in cones in the slash (30, 33, 35, 48). The ger-
mination failures are due largely to poor seed dispersal and poor
seedbed conditions (11, 48). Seed is released from windrowed slash
only on the outside edges of the slash piles because the serotinous
cones inside the piles are shaded from solar heating (48). Cones in
untreated slash seldom open because air temperatures above ground level
seldom reach the levels needed to open the cones (48). What seeds are
released are not likely to germinate and survive because conventional
logging leaves litter layers intact on most of the area (48).

Eyre and LeBarron (48) suggested site preparation followed by
lopping and scattering of the slash. The lopping of slash puts the

21

slash in closer proximity to the soil surface, where temperatures are
more likely to be high enough to open the serotinous cones (22, 54,
71, 75). Noakes (44) found that in lopped and scattered slash, many
cones within one foot of the ground had opened, fewer had opened at a
height of three feet, and very few had opened more than three feet
above the soil surface. Another study found that nearly all seed
released was from cones within seven inches of the surface (35).
Ground scarification is absolutely essential to the success of this
treatment (22, 35, 58, 75). A minimum of 60% of the area must be
scarified (75). Jameson (41) reports that this method works in
Saskatchewan, even on dry sites, and it is recommended for nutri-
tionally poor, dry, and very dry sites in Manitoba (33). In general,
however, this method has not been widely used, due to the inevitable
failures where it was misused, but also because of the high labor
costs involved relative to the low productivity of the sites (26).

It should also be noted that the minimum temperatures necessary for
cone opening are also lethal temperatures for first-year jack pine
seedlings (25, 48, 87).

Broadcast seeding has also been used with extremely variable
results (33, 35, 41, 48). Adequate seedbed preparation and favorable
weather conditions are necessary for success (35, 41, 79). Some suc-
cesses in improving numbers of seedlings have been reported, although
a high proportion of failures have also been reported, and in many
cases those trials labeled successes were far below optimum stocking.
Cooley (41) reports on a successful seeding attempt on a Grayling

sand site in northern lower Michigan, but the study may have been

22

conducted during an unusually wet growing season (personal comm., 1979).
In general, seeding is not recommended in the Lake States except where
soils are fine-textured or where the water table is permanently within
two to five feet of the surface (48, 79).

Spot seeding involves the placement of seed into a scalp or a
furrow. Although results are extremely variable, this method may be
successful if the seed is planted in sufficiently large mineral soil
scalps or furrows, and if the weather allows for germination and
initial survival. Sowing at depths of 0.25 to 0.75 inches improves
results (35, 50). Some tentative successes have also been achieved
in Manitoba with a corn seeder planting in furrows behind a V-blade
(33, 35).

Burning of slash under seed trees has been recommended as
another method of obtaining regeneration (26, 32, 35, 36, 48). It
has been successful in Saskatchewan in a 70-year-old stand where 60
trees per acre were left (35), in Manitoba where 20 dominants and 150
unmerchantable one- to three-inch trees per acre were left (35), and
in Minnesota where 10 seed trees per acre averaging 12 inches d.b.h.
yielded 15,000 to 20,000 seedlings per acre (36). Beaufait (26),
working in the northern lower peninsula of Michigan, found that slash
burning under 12 to 50 seed trees per acre under dry conditions usually
yielded sufficient seed and litter reduction to regenerate stands.
Slash burning cannot be expected to release live seed from the slash
itself, however. The slash carrying the fire nearly always ignites
the cones it contains, killing the seed (26, 35, 4B). The seed regen-

erating the stand must come from cones in the overstory. Beaufait (26)

23

concluded that the number of seed trees to be left depended on the
number and position of the cones. This slash-burning method should
only be applied in stands with greater than 100 square feet of basal
area, yielding at least 18 inches of evenly distributed slash, on
areas greater than 10 acres in size, in areas not near other flammable
stands, and with a minimum of 12 good trees per acre left as seed
trees (26). Generally this method is more successful as more seed
trees are left (32). Failures using this method have also been
reported (35, 4B).

Shelterwood cutting methods have been used with jack pine in
the northern lower peninsula for the past 20 years. Caveney (31) found
jack pine stocking on shelterwoods to be closely related to residual
basal area of jack pine, residual basal area of open-coned trees,
residual basal area of white pine, residual basal area of aspen, and
basal area of recent mortality. The average stocking in his study
was 36% of mil-acres stocked with jack pine, or 754 jack pine seedlings
per acre. The average stocking on Grayling sand, the most heavily
sampled soil series, was 37% of mil-acres stocked. Much better stock-
ing occurred on sites with ground cover less than six inches in height.
0n those sites dominated by reindeer lichen, 76% of mil-acres were
stocked. Stocking generally improved with time since first cutting.

An advantage of this method is that it is aesthetically pleasing
throughout the regeneration phase (31). Disadvantages include the
cost, harvesting an area twice, and the creation of ideal conditions
for the jack pine budworm (31). In Minnesota, Shelterwood cuts have

not been successful in obtaining adequate regeneration (48). Neither

24

seed tree nor strip cuts regenerated well in Manitoba, presumably

because of poor seedbed conditions (35).

Planting Techniques

 

Planting is the most reliable regeneration technique, but it
is also more expensive (26, 48). Several studies suggest planting in
furrows for dry, sandy sites in Canada and in northern Michigan (26,
33, 35, 87). The advantage of planting is that it circumvents the
germination and first-year survival problems so common on dry, sandy
sites (26, 33, 35, 87). The planted seedling has a well-developed
root system placed immediately into a relatively stable source of
moisture. It also is likely to be more heat-resistant than the new
germinant (87). Studies with jack pine and red pine have shown that
larger, sturdier, and older stock nearly always outperforms smaller
seedlings (17, 4B, 69, 76). In one study, planted 2-1 jack pine
averaged 10 feet higher in site index than plantations of 2-0 stock
(69).

Fall planting is not recommended on finer textures because
of the possibility of frost heaving (48). On sandy soils, Rudolf (79)
has shown that spring plantings have higher survival. Eyre and LeBarron
(B7) concur, attributing the poorer survival of fall plantings to
insufficient hardening off of seedlings and poor root contact due to
root dormancy. However, fall planting was a common practice in the
Lake States in the past (76), and if lower survival can be tolerated
or compensated for by planting more seedlings, the extension of the

planting season may make fall planting worthwhile.

25

Partial shading has been shown to improve first-year survival,
probably because of lowered transpiration rates (18, 60, 77). However,
this overstory must later be removed for best development (18).

Jack pine has also been experimentally planted in tubes and
in Jiffy peat pots on Grayling sands. Success has been shown to depend
on shade and on watering regime (67, 68).

Two final points applying to any silvicultural prescription
should be made. First, prompt regeneration of cutover stands is
easier and more likely to be successful than allowing several years to
pass before regeneration is attempted (35, 57). Second, no silvicul-
tural treatment utilizing seed from the existing stand should be pre-
scribed without consideration of the relative numbers of serotinous

and nonserotinous cones on the trees in the stand (31, 57).

Soil Properties of Burned Jack Pine Sites

 

Generalities are hard to draw regarding the effects of fire
on soils. The effects are highly variable, being a function of such
things as region, climate, forest association, soil type, plant species,
and fire intensity (10, 99). This section will briefly review the
literature on fire effects, with emphasis on those aspects most per-
tinent to jack pine regeneration, and also emphasizing studies con-
ducted on sites similar to the sand plains on northern lower Michigan.
For additional reviews the reader is referred to Wells et al. (99) and
especially to Ahlgren and Ahlgren (10).

This section of the review is based largely on two studies.

The first is by D. W. Smith (89), in which he studied nutrient dynamics

26

over a 17-month period immediately after a severe fire. The site was
a well-drained sandy podzol on an outwash plain in Ontario dominated
by jack pine. The second study on which this section is based is a
Master's Thesis by D. G. Scholl (82), in which he compares soil prop-
erties on a prescribed burned to those on an adjacent unburned site
on a Grayling sand in Delta County in upper Michigan.

The effects of burning on soil organic matter are extremely
important because large losses of soil organic matter might often result
in lowered site productivity, especially on the coarser-textured soils
(103). The soil organic matter includes the surface layers referred to
as the O horizon, forest floor, litter or duff, and the organic material
incorporated into the mineral soil. Smith (89) found that 79-91% of
the organic matter in the L-F-H horizons was consumed in a hot fire.
Scholl (82) found that an average of 72% of the organic matter in the
O horizon (L-F-H) was consumed. Smith (89) reported that the per-
centage of organic matter between depths of 0 and 2 cm in the mineral
soil was decreased for three months, then rose to above pre-burn levels
after snowmelt the following spring. The organic matter contents in
the deeper mineral horizons were unchanged for three months, after
which they increased. Organic matter losses are usually restricted
to the organic horizons and the upper inch of mineral soil, the losses
decreasing with depth (10, 12, 82, B9, 99). Frequent burning, rather
than causing an absolute loss of organic matter, causes a redistribu-
tion from the forest floor into the upper mineral horizons (10, 12).

Fire usually increases soil pH, although the persistence,

depth, and degree of effect vary (10, 99). The amount of the increase

27

depends on the amount of ash released, the original soil pH, the chemi-
cal composition of the ash, and the wetness of the climate (99). The
increase is often closely related to the amount of calcium released in
the ash (99). Smith (89) found large pH increases in the L-F-H hori-
zons four days after a fire. He found the pH's remained high for 10
months, then began to approach pre-burn levels. Fifteen months after
burning, pH's were at or below pre-burn levels. The pH's in the min-
eral soil horizons increased slightly as those in the organic horizons
decreased.

Smith (89) reported conductivity, associated with higher salt-
electrolyte concentrations, to be sharply increased in the L-F-H
horizons four days after burning, but it decreased rapidly between
five weeks and three months after the fire. As the conductivity of
the surface horizons decreased, that of the deeper mineral horizons
increased, especially in the first spring following the fire. Within
15 months, conductivities were near or below pre-burn levels.

The calcium contained in the L-F-H horizons in Smith's study
(89) was predominantly in acid-extractable forms prior to burning.
Following the burn, there was a sharp increase in water-soluble forms,
and then a decrease to below pre-burn levels within 15 months. Cal-
cium contents increased throughout the deeper mineral horizons, with
the increase in water-soluble forms between 2 and 12 cm especially
noticeable. After 15 months, the mineral soil horizons were near
pre-burn levels. In general, calcium concentrations change as described
above, increasing due to increased mineralization and then decreasing

as some leaching and redistribution takes place (23). Scholl,

28

sampling 17 months after a burn, found much more variable calcium con-
centrations on burned plots than on unburned plots in the O and Bir
horizons. The concentration differences were not significantly dif-
ferent, however (82). The rate of leaching of calcium relative to
that of other cations on these sites has been described as high (4)
and as low (12, B9).

Magnesium concentrations have been shown to increase imme-
diately after burning and then to decrease (4, 97). Magnesium concen—
trations at one-meter depths were found to be higher two months after
fire than before it. However, because these concentrations were lower
than those at 20 cm, Adams and Boyle (4) concluded the subsoil was
retaining magnesium. Scholl detected significantly higher post-burn
magnesium concentrations only in the 0 horizon (82).

The concentrations of potassium, especially the water-soluble
forms, were shown to increase sharply in the L-F-H horizons immediately
after burning in Smith's study (89). The concentrations then decreased
sharply within five weeks, with the water-soluble forms decreasing most
rapidly (B9). Ahlgren and Ahlgren (10) also note this tendency for a
sharp increase, followed by rapid leaching. Scholl, sampling 17 months
after the fire, found no significant differences (82).

Extractable phosphorus was found to increase in the L-F-H
horizons and the surface two centimeters of mineral soil by Smith (89).
The concentrations in these horizons then returned to pre-burn levels
within 15 months. In the deeper mineral horizons, phosphorus levels
were near or below pre-burn levels at the end of the experiment.

Scholl found much higher phosphorus concentrations in the O horizon,

29

though no change could be detected in the mineral horizons (82).
Ahlgren and Ahlgren (10) and Wells et al. (99) report that slight
increases in phosphorus levels in the upper mineral soil are often
reported.

Smith found water-soluble forms of sodium to increase slightly
immediately after burning in the L-F-H horizons, after which the sodium
concentrations in all horizons decreased and remained below pre-burn
levels through the end of the experiment (89). Smith attributed this
to the preferential adsorption of calcium and potassium on exchange
sites formerly occupied by sodium.

Acid-extractable forms of iron and aluminum were found to
increase immediately after burning, while exchangeable hydrogen sharply
decreased. The exchangeable cation concentrations returned to pre-
burn levels, while hydrogen was decreased through the end of the
experiment. In the top two cm of mineral soil, exchangeable cations
sharply increased and then decreased greatly between 13 and 15 months.
Exchangeable hydrogen decreased and then increased during the same
period. The cations slowly leached downward, pushing high exchange-
able hydrogen concentrations just below them. Fifteen months after
burning, both exchangeable cations and exchangeable hydrogen were near
pre-burn levels throughout the mineral soil (89).

Large losses in total nitrogen are often reported due to
burning (10, 44, 99). However, in many cases nitrogen concentrations
of the mineral soil are actually increased (10, 99). This can perhaps
be explained by the rapid mineralization of partially decomposed

organic compounds (10, 99). Nitrogen concentrations may also be

30

increased because of the much more rapid nitrification taking place
due to the increased pH commonly found in a post-fire situation.
Ammonification and denitrification are also often increased by burn-
ing (10). These changes in the nitrogen cycle may last several years,
although this varies (10). Biological nitrogen fixation may also be
responsible for large inputs of this nutrient after fire (10, 99), as

may atmospheric inputs (4).

METHODS AND MATERIALS

Two related survey studies were done. One involved the
examination of 12 areas regenerated by planting. The second concerned
24 locations on which natural regeneration methods were used or had
occurred naturally. These 24 locations included 12 fires (prescribed
and wild) and 12 partial silvicultural harvests (Shelterwood, seed
tree, and strip cuts). These silvicultural situations will be called
plantations, fire areas, and partial cut areas throughout the rest of
this study.

In each study, locations for installation of temporary field
plots were randomly selected from lists of potential study areas.
These lists of potential study areas, to be called locations through-
out the remainder of this study, were obtained by interviewing area
foresters and visiting the sites to determine whether they met the
selection criteria. It was decided that the study would be limited
to the following counties: Crawford, Kalkaska, Montmorency, Ogemaw,

Oscoda, Otsego, and Roscommon.

Plot-Selection Criteria
The plantations were randomly selected to meet the following

criteria:

31

32

1. Greater than 3 and less than 15 years since planting

2. Large enough to contain a 20.12 m by 20.12 m (one chain
by one chain) plot

3. Having a soil type matching the needs of the experimen-
tal design. The experimental design called for three plantations on
coarser-textured soils, three on medium-textured soils, three on finer-
textured soils or soils with fine-textured bands, and three on soils
with relatively shallow water tables. These were restricted to sands
or loamy sands in or near the range of characteristics described as
the Grayling soil series. The locations sampled are listed in Table 2
and plotted on a map in Figure l.

The fire and partial cut locations were selected to meet the
following criteria:

1. Greater than 3 and less than 20 years since last fire or
harvest

2. Large enough to contain a 60.35 m (three chain) by 120.70 m
(six chain) sample grid

3. Having had jack pine as the predominant species in the
last rotation

4. The fires must have had at least a partial overstory of
standing, live mature jack pine at the time of the fire

5. The partial cuts must not have burned in the last rota-
tion

6. Both fires and partial cuts must have had a soil type
matching the needs of the experimental design. The design called for

four coarser-textured soils, four medium-textured soils, and four

33

6 s3

Lake Michigan

Lake Huron

Otsego Co. Montmorency Co.
[

0Q, - / Al pena .

Crawford Co.

\ a'

 

 

 

 

 

 

 

 

 

 

 

‘\ ’ I
Travers City “A Oscoda Co.
0 ..J=="—__—_—
Kalkaska Co. Gray‘ing /
———~_ .a 1!
R C ‘qe
OSCOITITIOH 0. a 0 9111391 CO.
Cadillac f. .

 

 

 

Scale (miles)

LEE 1 l I I I_J
O 10 20 30 4O 50

 

 

I partial cut areas
A burned areas
0 plantations

 

 

 

Figure 1. Locations of plantations, burned areas, and partial
cut areas studied

Table 2. Legal descriptions of plantation, fire, and partial cut

locations sampled in jack pine regeneration study

 

 

 

 

County Township Range Section
Plantations
Crawford 28N 1W 7
Crawford 28N 1W 7
Crawford 25N 1W 14
Crawford 25N 1W l4
Ogemaw 24N 1E 21
Otsego 29N 1W l6
Otsego 29N 1W l6
Roscommon 24N 4W 5
Roscommon 24N 4W 6
Roscommon 24N 2W 17
Roscommon 24N 2W 22
Roscommon 24N 4W 6
Fires
Crawford 27N 2W 32
Crawford 27N 3W 34
Crawford 27N 2W l3
Kalkaska 25N SW 17
Ogemaw 24N 1E 21
Ogemaw 23N 1E 33
Ogemaw 24N 2E 19
Ogemaw 21N 3E B
Ogemaw 24N 2E 29
Otsego 29N 1W 36
Roscommon 24N 2W l7
Roscommon 21N 4W 29
Partial Cuts
Crawford 26N 3W 10
Montmorency 29N 2E 31 B 32
Montmorency 32N 2E 22
Montmorency 29N 1E 5
Oscoda 28N 1E 13
Oscoda 27N IE 5
Oscoda 27N 1E 8 & 9
Oscoda ZBN 1E 32
Oscoda ZBN 1E 27
Otsego 29N 1W l7
Roscommon 21N 3W l3
Roscommon 21N 4W 18

 

35

finer-textured soils or soils with fine-textured bands. This strati-
fication was achieved by using locations with higher than average
gravel and/or coarse sand contents as coarse-textured, those typical
of Grayling medium sand as medium-textured, and those with slight
banding or mottling in the lower profile as finer-textured. In many
cases, the finer-textured soils would have fallen into the Graycalm
soil series (Alfic Udipsamment, frigid, mixed). The locations sampled

are listed in Table 2 and plotted on a map in Figure 1.

Plantation Data Collected

 

After a plantation area was selected and determined to meet
the criteria, a square plot of approximately 0.0405 hectares was
installed, measured, and described. A 20.12 m line between two planted
rows was measured. Lines perpendicular to this line were run from each
end such that they ended between the same two planted rows and were as
close as possible to 20.12 m in length. On this approximately 0.0405
hectare block, henceforth to be called a plot, the area was calculated
and living and dead trees were counted. Other data and samples col-
lected included:

1. understory species and cover percentages, including woody

sprouts
2. soil description
3. sod description
4. soil samples for each horizon to a depth of at least 30 cm
5. numbers and ages of natural jack pine regeneration

36

Fire and Partial Cut Data Collected

After a location was selected and determined to meet the fire
or partial cut criteria, a starting point was randomly located such
that the entire grid of plots would fit into the area. Eight 0.00405
hectare plots were installed; four each spaced 40.24 m apart along
two transects 60.36 at apart. On strip cuts, the transects also had
to cross both a cut strip and a residual strip. At each plot, the
data and samples listed below were collected:

1. overstory species, basal area, and percent cover

2. understory species and percent coverages, including stump
sprouts
stump and snag diameters at ground level or 15 cm
soil description
sod description

soil samples of each horizon to a depth of 30 cm

\lOlU'l-hw

ages and numbers of jack pine regeneration

In some cases, obvious, easily delineated differences in regen-
eration success were observed on a plot. If the “different" area occu-
pied greater than 30% of the plot, the plot was split and treated as
two partial plots. All data were collected for each partial plot
except that two soil pits were not dug and described. Most often,
plots were split where large differences in understory vegetation and
in numbers of seedlings could be seen. The numbers of seedlings on
the partial plots were multipled by an expansion factor to compensate

for the smaller area of these plots.

37

Common Data-Collection Techniques

All readily recognizable understory species were listed. In
some cases closely related species were lumped into one category, as
in the case of the sedges, grasses, lichens, mosses, and blueberries.
Percentages were assigned to those species having greater than 5% area
coverage. These cover percentages were ocular estimates of canopy
cover. As such, few exceeded 70% total.

The percentage of the plot area occupied by stump sprouts was
estimated for each species present. Stump sprouts included all sprouts
from stumps cut or burned in the most recent harvest or fire, including
those harvested in both cuts of a seed tree or shelterwood harvest.

Overstory trees as defined here are those trees, formerly
belonging to the overstory, that survived the most recent cut or burn.
They were identified as to species and their diameters at 1.4 m were
measured to the nearest centimeter. The percentage of the plot
covered by each species was ocularly estimated. An estimate of the
horizontal dimensions of the crown within the plot was divided by the
total plot area to obtain a cover percentage.

Stumps and snags from the most recent fire or harvest were
counted and their diameters measured at 15 cm above the ground. Where
the second shelterwood cut had already taken place, the stumps from
both cuts were counted. The stumps in some of the fires were too
badly burned to allow good estimates of diameter.

A soil pit was dug and described at or near the center of each
plot. For each horizon, depths were measured, textures determined,

and gravel and cobble percentages by volume were estimated. Any change

38

in texture or in gravel or cobble content within the surface 150 cm
of mineral soil was described. Depths to mottles were also recorded.

A sod description was also done at the soil pit on each sub-
plot. This description involved an estimate of the thickness, and
classification of the relative numbers of roots and density of the sod
mat in each of the surface organic and mineral horizons. Sod density
as used here refers to the rooting intensity and to the cohesiveness
of the sod mat.

All living and dead jack pine trees were counted on the plan-
tations. On the fire and partial cut plots all regeneration was
counted and cut at the soil surface. The age of each tree was deter-
mined in the field if possible. If not, the bottom of the stem was
(brought into the laboratory, stained with dilute methylene blue, if
necessary, and aged using a stereoscope. 0n those plots where seedling
numbers were excessive, a quarter-circle was chosen at random, the
trees in this quarter-circle counted and aged, and the numbers of
trees multiplied by four to represent the entire plot.

Slope, slope position, and aspect were measured for each
plot.

Soil samples were taken from each mineral horizon occurring
in the upper 30 cm of mineral soil. Samples were taken from top to
bottom of each horizon, even if the bottom of the deepest horizon

extended beyond 30 cm from the surface.

39

Laboratory Analyses

Soil samples were air-dried as soon as possible after collec-
tion. Prior to analysis the aggregates were crushed and the samples
passed through a 2 mm sieve. Organic material larger than 2 mm was
disposed of. Gravel contents were determined for all sampled horizons
by weighing the sieved soil samples and the gravel retained in the
sieve.

Soil pH was measured on soil samples from the three upper min-
eral horizons using a glass electrode pH meter in a soil solution with
a soil:water ratio of 1:2.

Cations were extracted from the two uppermost mineral horizons
at each plot using a method similar to that described by Isaac and
Kerber (53), except 200 ml of l N_NH4OAc was used and samples were
shaken for 30 minutes in the extracting solution and did not stand
overnight. Calcium, magnesium, potassium, and sodium concentrations
were then determined using a Spectrametrics SMI III plasma emission
spectrometer.

The method used in the particle-size analysis was similar to
the pipette method described by Day (43). Some modifications were
necessary because of the low silt and clay contents of these soils.
Soil samples of 100 grams were used rather than 10 grams. The soil
samples were dispersed in 20 ml of calgon solution instead of 10 ml.
In the dry sieving process, sieves were shaken for 10 minutes rather
than three. Dispersion was done in a mixer run at low speed for five
minutes in samples which had been allowed to stand for 30 minutes in

calgon solution. Silt plus clay contents were calculated by subtraction

40

of the oven dry weight of the sieved sand from the oven dry weight of
the total sample. Particle size analysis was performed for one plot

at each location in the natural regeneration study, and for all plots
in the plantation study.

Hydrogen peroxide digestion of organic matter was performed
only on the A1&A2 and Al horizons prior to particle size analysis.
Rather than using a filter candle to remove the excess hydrogen peroxide
solution after digestion, the samples were allowed to settle, and the
supernatant was removed when it became clear. Next, a 1:4 mixture of
distilled water and methyl alcohol was added to the sample. Again,
when the supernatant cleared, it was removed by suction. The sample
was then oven-dried.

A porous plate apparatus was used to determine moisture contents
at 0.1 and at 15 bars of pressure for the three surface mineral horizons
on one plot from each location in the natural regeneration study and for
the three surface mineral horizons from each plot in the plantation
study. Samples were allowed to soak overnight and brought to equilib-
rium on a 15 bar plate.

Loss on ignition was determined with a muffle furnace set at
600°C for four hours.

Total nitrogen and total phosphorus contents of the two upper-
most mineral horizons were determined simultaneously on a 0.75 gram
sample. This sample was digested using one Kjeltab (containing
7.5 mg Selenium and 1.5 grams K2S04) and 10 ml concentrated H2504 at
420°C until digestion was complete. Digests were then cooled and

diluted. This digestion procedure is similar to that of Isaac and

41

Johnson (52). The nitrogen and phosphorus contents were then deter-
mined colorimetrically on a Technicon AutoAnalyzer 11 using an
ammonia-salicylate complex for the nitrogen and a phosphomolybdenum

complex for the phosphorus.

Data Analyses

 

The results of this study were analyzed on the CDC 6500 com-
puter at Michigan State University. Regressions, correlations, and
analyses of variance were performed using the SPSS: Statistical Package
for the Social Sciences (70).

Much of the data was transformed prior to analysis to force
data variability into more nearly normal distributions (93). The
square root transformation was used in calculating the square roots
of the numbers of trees. This transformation was also used on many
variables that did not include values of zero. The square root of
the data value plus 0.5 was used as the transformation for all vari-

ables for which zero was a possible value (93).

Description of Study Locations

This study was restricted to the outwash plains of the
Grayling-Roscommon area of northern lower Michigan. A map of the
study area with the distribution of the Grayling-Rubicon-Roselawn soil
association, which is found primarily on outwash plains, is shown in
Figure 2 (100).

The Grayling series is very extensive in northern Wisconsin
and in the upper and northern lower peninsulas of Michigan. It is,

as Eyre (48) says, "typical of the poorer sandy soils." It is most

42

l

Grayling-Rubicon-Roselawn Soil Association

 

 

 

13

a E

 

 

 

 

m
y: \\

 

\\\

 

 

 

 

 

 

Scale (miles)
1 L l I 1

4

Cl 10 20 30 4O 50

 

 

Figure 2. Extent of Grayling-Rubicon-Roselawn soil
association in study area (100)

43

often found on outwash plains and lake plains, popularly known as jack
pine plains or barrens.

This soil principally supports jack pine in the upper penin-
sula and jack pine and scrub oak in the northern lower peninsula of
Michigan. Ground vegetation is dominated by lichens, mosses, winter-
green, sweet-fern, blueberries, bearberry, bracken fern, “dry land
sedge,“ and sand cherry (l, 82). Soil Conservation Service (90) esti-
mates of site index range from 12.5 to 15.2 m (41 to 50 feet) and
average annual growth per acre is approximately 100 to 125 board feet
per acre. Shetron (84), working in the upper peninsula, estimates
site indices for this series to range from 11.9 to 14.9 m (39 to 49
feet), averaging 13.4 m (44 feet). Average volume is estimated to be
118 cubic meters per hectare.

A typical Grayling profile description is as follows (90):

Horizon Depth (cm)_ Horizon Description

A1&A2 O-B Black (N 2/) (Al), and grayish brown (lOYR 5/2)
sand (A2); coated and uncoated sand grains mixed
throughout the horizon, giving a salt and pepper
appearance; moderate organic matter content in
upper part; weak medium granular structure; very
friable; very strongly acid; abrupt smooth
boundary. (5 to 10 cm thick)

 

BZlir 8-23 Dark brown (7.5YR 4/4) sand; weak Coarse granu-
lar structure; very friable: strongly acid;
clear smooth boundary. (10 to 20 cm thick)

822ir 23-38 Strong brown (7.5YR 5/4) sand; single grained;
loose; medium acid; clear irregular boundary.
(10 to 36 cm thick)

B3 38-58 Brown (7.5YR 5/4) sand; single grained: loose;
medium acid; gradual smooth boundary. (8 to 25
cm thick) ,

C 58-152 Light brown (7.5YR 6/4) sand; single grained:

loose: medium acid.

44

The principal competing series are the Rubicon, Croswell, and
the Graycalm. The Rubicon and Croswell can be distinguished from the
Grayling by virtue of their spodic horizons. The Graycalm can be dis-
tinguished by the presence of textural bands in the lower profile. The
Grayling series is classified as a mixed, frigid, Typic Udipsamment (90).

Some of the plots were selected on soils belonging to the
Graycalm series. Very thin bands, usually less than 3 mm in thickness,
were found deep in the profile. Technically, these soils would fall
into the Graycalm series; however, it was believed that these soils
were too little different from the Grayling to be excluded from the
study.

The Graycalm series is found in the central and northern lower
peninsula, the eastern upper peninsula of Michigan, as well as in
north-central Minnesota. It is found primarily in sand on till plains,
moraines, and outwash plains. The vegetation is chiefly oak and white
pine in the southern range, and jack pine and scrub oak in the northern
range, according to the Soil Conservation Service soil series descrip-
tion (91). Site index on this series was found to average 15.8 m
(52 feet), while average volume is 169 cubic meters per hectare (B4).

A typical profile is described below (91):

Horizon Depth (cm), Horizon Description

 

 

Al 0-8 Very dark grayish brown (lOYR 3/2) sand; moder-
ate medium granular structure; very friable;
many fine roots; very strongly acid; clear wavy
boundary. (5 to 13 cm thick)

BZlir 8-16 Dark brown (7.5YR 4/4) sand; weak fine granular
structure; very friable; common fine roots;
strongly acid; clear irregular boundary.

(8 to 30 cm thick)

45

Horizon Depth (cm), Horizon Description (cont'd)

 

822ir 15-33 Strong brown (7.5YR 5/6) sand; weak fine granu-
lar structure; very friable; few fine roots;
medium acid; gradual wavy boundary. (0 to 30 cm
thick)

823ir 33-56 Yellowish brown (lOYR 5/6) sand; single grained;
loose; few fine roots; slightly acid; gradual
wavy boundary. (0 to 64 cm thick)

A2 56-89 Light yellowish brown (lOYR 5/6) sand; single
grained; loose; very few fine roots; slightly
acid; abrupt broken boundary. (0 to 64 cm
thick)

A&B 89-152 Light yellowish brown (lOYR 6/4) sand (A2);
single grained; loose; lamellae and bands of
brown (7.5YR 5/4) and reddish brown (5YR 5/4)
loamy sand (Bt); weak medium subangular blocky
structure; friable; bands are 1/4 to 2 inches
in thickness with a total accumulation of 5

inches; 5 percent by volume of pebbles; slightly
acid.

The principal competing series are the Mosomo, Zimmerman,
Chelsea, Coloma, and Grayling series. The Mosomo and Zimmerman series
contain more fine and very fine sand. Chelsea and Coloma soils are
mesic. The Grayling series lack the Bt lamellae. A soil is described
Graycalm if there is any banding, though the band be as little as
2 mm in thickness. The Graycalm series is a mixed, frigid Alfic
Udipsamment (91).

The stratification used in plot selection in the natural regen-
eration study was intended to assure sampling across the range of tex-
tures found on the Grayling series. However, due to the low variability
in textures that was found, this stratification was ineffective. Dif-

ferences between the strata were so small as to cause differences

46

between the strata not to be statistically significant. Textural dif-
ferences on these sites are not likely to be a factor in regeneration

success. However, there were significant differences among the strata
used in the plantation study (Table 10). This stratification involved

a wider range of soils than that in the natural regeneration study.

RESULTS AND DISCUSSION

This section contains the results of the statistical analyses.
The factors controlling regeneration and the germination patterns on
the burned and partial cut areas are identified and discussed. The
results of the plantation study are also presented and explained.

Site Factors Controlling Regeneration Success
on Partial Cut Areas

The partial cut essentially opens up small areas within a
stand, allowing new vegetation to utilize the site resources formerly
allocated to the harvested trees. Usually, a large proportion of the
seed reaching these sites is dispersed from the residual overstory.
After these sites are reoccupied, the residual overstory is removed
and the rest of the stand is regenerated by seed from the slash and
from the previously established regeneration. This provides for a
slower, less destructive, more aesthetically pleasing regeneration
phase than do most other regeneration techniques.

The regeneration results from the partial cut plots examined
in this study are summarized by location in Table 3. In examining
these results, one finds that the average stocking on these plots is
1346 trees/hectare (545 trees/acre). Given the "clumpy" (31) nature
of this regeneration, it is obvious that, on the average, the partial
cut plots are not well stocked. For the purposes of this study a
stocking level of 2500 trees per hectare (1012 trees per acre) will

47

48

.mpo_n pmwugwq was upon m:_mmwe
on mac mucoENmsnuo cur: .copumoop emu muopn agave zpmumewxogqnm no women memos comumuogo

 

 

 

 

 

 

 

mo.~ N.m oa~.o PN.m N¢.o pm.P o.NMN oemp mmmcm><
coozgmupmgm Np.o m.o mmpo.o eo.m mN.o o.o m.mmN on m mom
voozgmupmnm ep.o ¢.o N¢wo.o mo.m Nm.o m.N o.mcm Now x mom
coozgmupmgm o¢.o p.m mnpo.o mm.¢ ¢_.o ¢.o m.oom mmN_ N who
Neopmgm>o or
.uoozgmNFmsm NN.o N.F mopo.o mN.m mN.o o.~ m.om_ Fem m mmo
xcoumcm>o o:
.mmca comm ep.o 0.0 mNpo.o m~.m mm.o N.N m.mmop emmN v mmo
voozgmp_m;m mm.o m.N capo.o mm.¢ mm.o N.N o.mmm oewN m mmo
«no awgpm oo._ ¢.o_ NFNo.o m—.m ¢_.o m.o m.~mm mmN_ N mmo
noozgmupmcm N¢.o m.op ompo.o o¢.m Np.o m.o o.pmo wap o mmo
uzu argum NN.o o.— pmpo.o mp.m em.o N._ N.mop oem N we:
So 3.5 ummm de mé R86 25 mud ed 93m mam x 82
usu awcum Pm.o o.N voo.o we.m mN.o N.o o.Nmm epm Np mo:
acoumcm>o o:
.uoozgmupmcm mN.o m.o Nmmoo.o em.¢ mm.o m.v m.¢mN NNmp p mmu
Ass Ase A.=\.s..sv
mucmesou vcmwuuwm cam: uguwuuwm cum: ucmwnuwm cam: vgmwummm cum: cowumooA
Ew>ou :mespomzm NNm mo :a Lm>ou zcgmngmmm mcmaszz acwpvmwm

 

cowumqu a; mmpnmwcm> nmaumpmm ecu mgdgEa: mcwpvmom coca uzo .mpugma mo magnesam .m m_amh

49

be considered a regeneration success. Although this stocking level
may seem high relative to those tolerated in plantations, these
naturally regenerated seedlings are not optimally spaced and are
therefore (1) likely to be of poor form and (2) likely not to fully
occupy the site except at somewhat higher stocking levels. Given
this criterion, only two of the locations studied would qualify as
successes, and over half the locations are far below the minimum
acceptable stocking.

As Caveney (31) pointed out, regeneration success is related
to type of competing understory vegetation. He found that percentages
of mil-acre plots stocked with jack pine were well above the study
average if lichen, moss, litter, or bearberry was the dominant under-
story plant on a plot. He found that plots dominated by blueberry
tended to be well below average stocking. Ahlgren (5) reports that
such plants as bracken fern, blue joint grass, sedge, and fireweed
were especially harmful to the growth and survival of reproducing jack
pine seedlings.

Significant positive correlations between numbers of trees and
lichen, sweet-fern, and bearberry coverage were found in this study
(Table 4). Also, multiple regression equations derived from these data
and summarized in Table 5 included bearberry coverage and sweet-fern
coverage as significant independent variables, both with positive coef-
ficients. These relationships can also be seen in the data presented
in Table 3.

Caveney (31) attributed the higher regeneration success on

lichen and moss-dominated plots to the superficial root systems of

50

these species. The jack pine seedling has an excellent chance of
surviving on these plots because its taproot is able to penetrate the

zone in which the lichen or moss is competing for water.

Table 4. Significant correlations between site factors and
regeneration success

 

Correlation Coefficients

 

 

Independent Variable Trees/ha Squ$::e:?ޤ of
Overstory basal area (m2/ha) -0.22* -O.24*
Lichen cover (%) 0.20* 0.21*
Sweet-fern cover (%) 0.27* 0.28*
Bearberry cover (%) 0.28* 0.28*
pH of 822 horizon -O.36* -0.25*

 

*Significant at .05 level.

The reasons for the improvement in regeneration success as
bearberry and sweet-fern cover increase are more speculative. It
may be that these plants, with their woody root systems, do not fully
occupy the surface soil, allowing the young jack pine seedlings to
root alongside them. It may also be that these species have site
requirements similar to those of jack pine. Their presence on a plot
may indicate that unoccupied microsites suitable for jack pine seed-
lings are or recently were available.

Overstory basal area also had a significant effect on regenera-

tion success, appearing in the square root of number of trees regression

51

Table 5. Multiple regression equations developed for partial cut plots

 

Equation 1.

trees/ha = 6800** + 417 ¢xf* + 0.5 - 2514 7x5* + 0.5 + 477 Vx§* + 0.5

- 86(xz)

where x1 percent bearberry cover
pH of 822 horizon

x3 = percent sweet-fern cover

x4 = ppm total phosphorus of 821 horizon
n = 88
R = 0.59**

SEy = 158

 

Equation 2.
Jtrees/ha = 1272** + 74 fo + 0.5 + 114 7x§* + 0.5 - 16.5 7x§* + 0.5

- 481 x3* + 0.5 - 99(xg*) + 114(xg)

 

where x] = percent bearberry cover
x2 = percent sweet-fern cover
x = overstory basal area (mZ/ha)
3 0.1593
x4 = pH of 822
x5 = ppm exchangeable calcium of 821
x6 = A1&A2 thickness (cm)
n = 88
R = 0.62**
SEy = 27.4

 

*Significant at .05 level.

**Significant at .01 level.

52

(Table 5). Overstory basal area increases tend to decrease numbers of
seedlings, as Caveney (31) found. As the residual overstory becomes
more dense, a smaller proportion of the area is unoccupied, and there-
fore able to support regenerating seedlings. These results should be
used carefully, however, because they do not necessarily mean that
heavier first cuts allow greater regeneration success. Rather, they
are probably a result of sampling partial cuts on which the residual
overstory had been removed as well as some on which it had not.

No differences in organic matter content or water-holding
capacity in the upper mineral horizons were found between the fire
areas and the partial cut areas in the natural rengeration study
(Table 6). However, significant differences in profile depth, depth
to mottling, and 822ir gravel contents were found (Table 6). It is
likely that these differences were caused by local differences empha-
sized by heavier sampling for each treatment in the districts in which
one treatment or the other was in favor. This may have caused some
bias in the sampling, and must be considered in the interpretation of
the results.

Differences in soil fertility between burned and partial cut
plots were also found (Table 6). Though the differences in the treat-
ment averages are in some cases not great, it is the great variability
around these means that provided the variability upon which the regres-
sions were based. Nitrogen concentrations in the A1&A2 horizon were
significantly higher on fire areas. This will be discussed in the next
section. Available magnesium and potassium concentrations were sig-

nificantly higher in the 821ir horizon on the partial cut plots.

53

 

 

 

8:0 meozma

 

mzwu

Poo.ov p.4m momp owe mPoN Am=\mmmzov cowowtmcammm
omo.o om.~ cop we.m __P Aggy mgr—Boos o» sagas
moo.o mom.o N.m¢ No.P _.m¢ Aggy goamu m._Loza
_oo.ov Nmeo.o mo.~ ommo.o me.o AEUV goaau Lauu_4
Noo.o emp.o mm._ mmNo.o mum.o ARV Lm>ou saggy;
Nmo.o emp.o «.m. m-.o o.h_ ARV Lm>ou mmumm
_oo.o mmmo.o mmo.o m.~o.o moN.o Am;\~ev mate pammn scopmzm>o
Foo.o NNN.o mm.e mmp.o mN.F ARV Lm>ou scoomtm>o
oPo.o «MP.o mN.m m_mo.o em.~ Aggmwmz Na NV pcmpcou _m>mLm NNm
Noo.o omp.o mm.m mmp.o mo.~ Asaav sapmmaooa mpnmmmcwguxa .Nm
eoo.o _Fm.o om.“ mm¢.o mo.m Asaav E=Wmm=m~e apnemmceguxm PNm
~_o.o m.o~ Fem N.NN ooh Azaav cmaocomc Papas ~<ap<
su.flwwwuoza utnmuuwm cam: uzuwuuwm cow: mpamvta>

 

muopa coccan ucm muopg uao pmmugma cmmzumn mmucmgmwewc accomewcmmm .e «paw»

54

Perhaps these cations increase through a rotation and are subsequently
leached when or if a stand is burned. Partial cuts are not as greatly
disturbed at the end of a rotation and so the cations are not leached.
Though these post-burn cation losses are not reported by Smith (89)

or Scholl (82) for outwash plain soils, it may simply be that Smith

and Scholl both missed them because they did not sample beyond 15 and 17
months after burning, respectively. The burned plots sampled in this
study were sampled between 2 and 19 years, with an average of 7 years
after burning. However, these higher cation concentrations may be
associated with the sampling bias discussed above.

These higher cation concentrations are associated with decreased
numbers of seedlings as indicated by the negative correlation between
pH of the 822ir horizon and numbers of trees (Tables 3 and 4). The
negative effect of fertility is also demonstrated by the negative coef-
ficients of the pH of the 822ir horizon and of the total phosphorus
and exchangeable calcium contents of the 821ir in the regression equa-
tions (Table 5). Also, an analysis of variance on plots broken down
by regeneration success found that those plots with greater numbers
of trees had significantly lower pH's in all three surface mineral
horizons.

This negative effect of fertility can be understood if one
considers the ecological niche of jack pine. It is able to survive at
lower nutrient status than any other commonly planted Lake States coni-
fer (102). It is likely that it is most competitive on low-fertility

sites because of the absence of other, more demanding species.

55

Partial Cut Regeneration Patterns

The regeneration patterns for the partial cut areas are plotted
in Figures 3 and 4. Some partial cuts begin to regenerate the year fol-
lowing the first cutting, while on others, there is a one- to two-year
lag. In nearly all cases there are six to seven years of relatively
good regeneration followed by a gradual decrease. Seedlings may con-
tinue to germinate as long as 11 years after cutting. The reasons for
the timing of this process might include differences in yearly seed
crops, weather conditions, competition, and seedbed conditions. It
was attempted to relate partial cut year-to-year regeneration success
to weather information from U.S. Weather Bureau stations in the area,
but no strong relationships were found. The complexity of the response
of seeds and seedlings to weather conditions and the large variation
in precipitation over short distances make this relationship difficult
to define. ‘

Site Factgrs Controlligg_Regeneration
Success on Burns

Although jack pine is naturally adapted to post-fire regenera-
tion, burning does not necessarily guarantee successful regeneration.
First, there must be an adequate seed supply. Second, given an ade-
quate seed supply, site conditions must be such that jack pine is able
to germinate and third, the seedling must be competitive enough to
survive.

Burning greatly alters the condition of the site. Usually,
the overstory is killed. Understory plants may also be killed or

severely damaged, depending on their adaptations to burning and on the

56

mmmgm «no vaugma co mmcppummm mcm>m>czm mo meowuznwgumwu mm<

 

m mo:
N who
m mmo
e mmo
m mmo
N mm:
o mmo
x mo:
N— mo:
N was
p mac
x mom

mcwuuao «new» Loumu memo»

 

.m «camp;

j

cop

.OON

18m

rooc

Wags

.ooo

T8m

 

( Ell/$38.11) Bundmes ;0 an}: o; Butntnans sfiuupaas paaeu iuuafi [luau

57

mmmcm uzo meugma can mmmcm umccan co mmcwpummm acm>w>czm mo mcowpzavcumwu mom mmmcm><

Amsawxv umo>su= Lo 62—» oucvm oewh
m. e. m. N— rp pm a N N o m

p b P h

~Q’
M
N

 

amuse cases;
moose aau :2»an lllu

 

T8N

lace

toow

uooNp

loav—

 

ficomp

(eu/saaai) Butldmes 4o amt: 0: Butataans sfiutlpaas patrolmaafi [[MaN

.e mczmwm

58

severity and type of fire. Lichen coverage, for example, is often
greatly reduced by burning, while sedge coverage is usually increased
(Table 6).

This rapid mortality leaves the site largely unoccupied.
However, burning often kills only the aboveground portions of a plant.
As Darlington (42) pointed out, the great majority of the plants on
the outwash plains have some type of deep roots or rootstocks able to
survive a fire and reproduce vegetatively afterward. Ahlgren (10)
discussed the abilities of blueberry and bracken fern to do this.

Besides killing or damaging the vegetation, the fire also con-
sumes a large proportion of the forest floor. This study found the
depth of the forest floor to be significantly reduced on the burned
areas relative to the partial cut areas (Table 6). This in itself is
an important effect, as a large body of Canadian literature has shown
that the reduction of the forest floor improves germination and sur-
vival. According to Chrosciewicz's classification (39), the average
burned plot in this study was a high-quality seedbed, while the average
partial cut plot was of only moderate quality. The differences in
regeneration success between the two treatments are probably partially
due to differences in litter depth, although it was not asignificant
variable in any of the regressions or correlations with numbers of
trees. This, however, is probably due to the analysis of the two
data sets separately.

A secondary effect of the reduction in litter depth is the
subsequent release of large amounts of nitrogen into the mineral soil.

Partial combustion changes many of the organic nitrogen-containing

59

compounds in the remaining litter into easily mineralized forms (10,
99). This process manifested itself in this study as a significantly
higher total nitrogen content in the A1&A2 horizon on the burned plots
(Table 6).

Significantly more seedlings were found on the burned plots
than on the partial cut plots. This can be seen in Table 7, where
7 of the 12 locations were "successes," as measured by the criterion
of 2500 trees/ha established earlier. Of those locations not meeting
the criterion, only two are far below it. The average stocking of 7123
seedlings/ha on the burned plots is well above the minimum considered a
success. These results are in sharp contrast to the partial cut stock-
ing levels.

As was mentioned earlier, sedge cover was found to be signifi-
cantly higher on the burned than on the partial cut areas (Table 6).
Sedge cover is also more strongly correlated with numbers of trees
than any other variable and it has a negative correlation (Table 8).
It is also a significant variable in the multiple regression equations,
with negative coefficients (Table 9). In short, more sedge means less
trees, and this factor may have greater control over tree numbers than
any other factor measured.

The negative effect of the sedge cover can be seen in Table 7.
The location with the highest average stocking is the lowest in sedge
cover. In contrast, the location with the lowest stocking has the
greatest sedge cover. This relationship can generally be seen at the
other locations also, although averaging all of the plots at a location

and compensation by other factors in some cases weakens it.

60

.mpopa pmmusma use some mcwmmws
op man mucwsumsnum saw: .copamoop Ema moo—a ugmwm apmpmewxogaam co comma memos cowpmqum

 

 

 

Fn.o e.p ce.~ mm.m Fm.m m.NN pmoN. mN_N mmmcm><
mgwmupmz mo.o P.o F¢F.o NN.N mm.o m.om mmN ommp a mom
scan cmnwgummgm ¢~.o m.o mNP.o mm.m mm.o m.¢¢ m.mm mom m mom
«Eweupwz ep.o m.o mop.o mm.m mN.o o.m mmN ompN F mho
xcoumgm>o m>w4
mngnpwz agar; mm.o m.P cop.o oN.N mn.o m.mN mop Fem a duo
msweupws ¢F.o v.0 Pup.o um.m Nm.o o.oN oNpN oNom m duo
wow...» 83 .525
scan umawcummga eN.o o.o mop.o oN.m ew.o N.Ne mmN oppN N mwo
wgwmupvz mo.o o.m om~.o Fe.w mm.o N.m coon ooNeN m use
mgpeupw: o.o o.o NmN.o No.N me.o m.cm omop oNomp N mac
mgwwuppz o.o o.o mop.o m¢.m mu.o m.mp mNm omme N m<¥
wg_mnpw3 ¢¢.o N.p mmm.o mm.w Nm.o w.mm mow ommN m dmu
mgwmqu: mm.o m.m mm_.o em.—_ eo.p m.om ONmF ONNFP m duo
mcmvawz Np.o m.o Nop.o mm.pp F¢.o o.wN cum oommp x emu
Asuv Am;\mmmgpv
mangEou ecumuuwm cam: ucwwnuwm new: uguwumwm cam: uguwumwm cam: cowumuog

 

gm>ou accmagmmm

 

mmmcquzh me

 

Lm>ou mmvmm

 

mg¢g§=z mcwpummm

 

coppmoop An mmpnmwgm> cmuumpmm ucm mamasac mc_Fummm mega more we oxgmssam .N open»

61

Table 8. Significant correlations between site factors and
regeneration success on burned plots

 

Correlation Coefficients
Square Root of

 

Independent Variable
Trees/ha

 

Trees/ha
Sedge cover (%) -0.37* -0.37*
821 thickness (cm) 0.21* 0.22*
Sweet-fern cover (%) 0.18 0.24*
Bearberry cover (%) 0.18 0.20*
Depth to mottling (cm) -0.16 -0.26*

 

*Significant at .05 level.

It is easy to see why sedge is so important. Like jack pine,
it is well-suited to regenerating under post-fire conditions. Although
sedge is often found in older stands, it is less common as stands grow
denser, as indicated by its significant negative correlations with
overstory cover and overstory basal area on the partial cuts. Ahlgren
(9) discusses its low frequency in mature red and white pine stands.
Nonetheless, Ahlgren found live seed in the forest floors of these
stands. He also found high sedge frequencies and seed numbers in
recently prescribed-burned jack pine stands in northeastern Minnesota.
Personal observation suggests that this species may also reproduce
vegetatively after light to moderately severe fires.

Sedge is an extremely competitive species under post-fire con-
ditions. It forms a dense root mat in the Al&A2 and upper 821ir hori-

zons. Field sod descriptions indicated that this mat was usually much

62

Table 9. Multiple regression equations developed for burned plots

 

Equation 3.
trees/ha = -2518 fo‘ + 0.5 + 7824 7x3; + 0.5 + 2077 7x3 + 0.5

 

where x1 = sedge cover percentage
x = thickness of 821 horizon (cm)
2 2.54
x3 = bearberry cover percentage
n = 92
R = O.67**
SEy = 1017

 

Equation 4.
Jtrees/ha = 1437** - 242 fo* + 0.5 + 125(x§*) + 296 7x§* + 0.5

- 51.3(xz*) + 1.85 7kg + 0.5

where x1 - sedge cover percentage

 

x2 = profile depth (cm)
x3 = bearberry cover percentage
x4 = depth to mottling (cm)
= total basal area (m2/ha)

x5 0.1593

n = 88

R = 0.66**
SEy = 72.0

 

*Significant at .05 level.

**Significant at .01 level.

63

thicker on recent burns than on partial cut areas. The importance of
this sod is magnified by the low water-holding capacity of these soils.
Soil moisture depletion is never more than a few days away, and the

jack pine seedling must get a taproot beneath this sod mat and into

a more stable, less competed-for water source. The deeper this sedge
root mat is, the lower is the probability that the seedling will survive
long enough to sink a taproot through it. It may be that the jack

pine seedling's chances of survival are further reduced by an allelo-
pathic effect of the sedge.

In many cases during the field data collection, plots were
split where one section of the plot was dominated by sedge and another
by other understory species, with in almost all cases far fewer sur-
viving jack pine in the portion of the plot dominated by the sedge.

The sedge is excluded from certain areas in stands in which it is
otherwise the dominant understory plant. The reason for the exclu-
sion in these areas may be that a denser overstory decreased the amount
of sedge in the understory prior to the burn, thereby decreasing sedge
seed supplies and increasing the competitiveness of other species.

It may also be due to more intense burning around the deeper litter
layers and heavier slash accumulations occurring where the stands are
denser. There, hotter fires might kill more of both the seed in the
forest floor and the sprouting underground portions of the plant.

Sedge cover was found to have had a correlation of -O.55 with
years since burning. This may be due to the fact that on some of the
older burned plots the regeneration had closed crowns, thereby begin-

ning to exclude sedge from the understory.

64

Bearberrycover was found to be significantly and positively
correlated to regeneration success (Table 8). It was also a signifi-
cant variable in the regression equations (Table 9). Again, the posi-
tive effect of the presence of this species may be due to its less
competitive relationship with jack pine seedlings, or may be due to
its tendency to grow under conditions similar to those favoring jack
pine.

Two interesting and apparently very important variables in the
regression equations are profile depth and thickness of the 821ir
horizon (Table 9). It is relatively easy to understand the importance
of the thickness of the 821ir, since this horizon is probably the site
of a large portion of the jack pine root systems. The thicker this
horizon is, the more site resources are available to the jack pine
seedling. The resources in this deeper horizon are by and large not
available to the shallower-rooted herbaceous competition. The most
important of the resources being discussed here is probably water,
since higher fertility would favor other woody species and would put
the jack pine at a competitive disadvantage. The total profile depth
is probably important for similar reasons. Profile depth is signifi-
cantly related to depth of the 821ir and is probably roughly equivalent
to rooting depth. Again, deeper profiles provide more site resources
to the trees, but are of less value to the competing herbs. It is
also possible that the deeper profiles are able to support a denser
overstory, shading out the sedge and yielding more slash and litter.
Thus, the effects of the deeper profiles would favor the seedling

indirectly.

65

Depth to mottling is significant in the regression and simple
correlation of square root of tree numbers (Tables 8 and 9). The
coefficients are negative, meaning that deeper mottles tend to result
in fewer trees. Where mottles are quite shallow, they indicate the
presence<rfstanding water during some part of the year within reach
of the seedling root systems. Where mottles are deeper, they are not
likely to affect the seedlings themselves; however, their presence
may have an indirect effect by supporting a denser parent stand.

Total basal area of the pre-fire stand is also a significant
variable in the square root regressions (Table 9). Its effect is
positive. Total basal area is a variable combining the basal area of
the live overstory with stump cross-sectional areas and the cross-
sectional areas of snags at stump level (15 cm). This variable is
thus a rough measure of the density of the pre-fire stand. Its
importance, discussed above, is that heavier pre-fire stocking results
in a less vigorous understory and in some cases in more intense fires
due to heavier slash loading and deeper litter layers. Also, more seed
may be available.

The correlation coefficients obtained for the multiple regres-
sion equations developed in this natural regeneration study are rather
low, although not unusually so for this type of study. This means that
a fairly large proportion of the variability in stocking cannot be
explained by the factors measured. Large differences in seed availa-
bility were no doubt partially responsible for this unexplained varia-
tion. Also, the effects of weather were not accounted for by variables

in the equation. Mortality caused by insects, rodents, and other

66

herbivores was also likely to have been a contributing factor.
Perhaps the most important reason for this unexplained variability,
however, is the fact that the site factors were often measured long
after much of the regeneration became established, during which time

site conditions are likely to have changed.

Burn Regeneration Patterns

One of the most interesting results of this study is the post-
fire germination pattern of surviving jack pine seedlings. The litera-
ture on germination patterns is dominated by reports of heavy germina-
tion the year following a fire, with small numbers of germinants in
the years immediately following (5, 32, 35). Beaufait (26), however,
noticed a lag in germination in northern Michigan. This lag was
obvious in this study also, as shown in Figures 4 and 5. In most
cases, either few seedlings germinate or few survive in the first year
following the fire. Maximum germination of surviving seedlings then
occurs in the second, third, fourth, or fifth year after burning.
Figure 4 shows that the fourth year has the highest germination, on
the average. The peak at two years is entirely due to area OGF 8.

Much of the seed germinating beyond the fifth year is probably from
cones on the regenerating seedlings.

Beaufait (26) attributed this late germination to "apparent
dormancy." Yet, high germination is often reported in other studies,
often within days of dispersal. It is possible that the early-
germinating seeds do not survive on these sites, due to the low osmotic

potentials caused by the high cation concentrations in the post-burn

mmmcm umccan co mmcwpummm mcw>m>czm do mcowpznwcpmwu mm< .m meamwm

mcwcgzn seven memo»

 

67

 

m— Np pp op m w n m m e m N p o
A; \LAH . : r111. . - - .-/ .1 -. -ia . o
:1... Y7 7114! - /://:l - 1-..... :llsJUANMrJ/li. LU- ./.\.../ \/ \ MAN.“ W
474/17 x. //.4..- u I.”- u W314”. \ but
«I I VAT N ..\ . .- .D
77 /. a \ r dilx 44‘s \ a
z; , /. N , .\ .W . - 82 W
./ cl / \ . \ ~ . u
)2 1L \N _ N m
11414177121111) .’ as“ N \ U.
» y N ‘ I . ocoN “u
x . , . x m
o . u
x . , W \ .w
m “.01 41111 )2 K4 ’ _ \ j m M
m “.8 .1.1 J, 5...... .. N \ w
.. , u
Pu—PO ulllll <\ w .. N .. 6
as .. . L. . L .8... m.
:8 ...i ... ,_ .. I m
m 28 ll..- .. . x. f 88 ......
N .8 - ...... ,_ :__ m
NL§ III- , .\\ m
x as 3::- L .\ 88 H
mama .1-.. ,\ m
m egg .1..... ( Mm
w.
w coca

 

68

ash (5), to the high temperatures on the blackened soil surface, or
due to other causes. However, these conditions are not likely to per-
sist for more than one or two growing seasons and thus do not explain
the longer lags. Also, if germination followed by high mortality were
occurring, there would be little seed left behind to germinate in sub-
sequent years.

A better explanation might be that the conditions necessary
for germination do not commonly occur until several growing seasons
after burning. The characteristics of the seedbed likely change a
great deal in the time immediately after the fire. The partially
consumed litter is rapidly decomposed, charcoal is washed out and
blown away, the released nutrients are fixed or leached downward, and
a dense understory often develops. The moisture supply available to
the seed might be increased enough to allow germination by any of
these.

Jack pine seems to be one of the few tree species in which the
seed is able to survive in the soil until conditions suitable for ger-
mination develop. Perhaps this ability to survive for long periods
without germinating is related to its ability to survive for several
years sealed in the cone. It may also be the result of the slower ger-
mination sometimes reported for older seed (2, 25, 35).

Germination requires some minimum number of days above a cer-
tain temperature in which plentiful water is available to the seed
(19, 48). However, litter seedbeds often have poor water relations,
as discussed in the Literature Review. The sandy soils beneath the

forest floor on these sites are likely to worsen the problem because

69

they are not able to replenish moisture levels as the forest floor
dries. The seed might thus be stored, with its moisture absorption
requirement only partially met, until the next spring, and perhaps the
next, and so on. The same dry conditions delaying germination would
tend to reduce damage to the seed by fungi and other seed-killing

organisms.

Plantation Regeneration Success

 

The differences between plantation silviculture and natural
regeneration are great. Yet, by knowing how jack pine responds to site
factors under natural conditions, one can better understand its response
to the various aspects of plantation silviculture.

The need for site preparation prior to burning on sod-dominated
sites has long been known. This has most often been done using a
V-blade or a disk. This treatment peels away the sod and the sod-
supporting A1&A2 horizon. It ensures that the roots of the seedling
are placed directly into that part of the soil in which they are most
competitive, and that they are placed there essentially free of compe-
tition.

There seems to have been a rather common practice in the past
of furrowing, allowing several years to pass, and then planting the
furrows. Although a jack pine seedling may still survive under these
conditions on dry sites because of its competitive advantage, it will
grow more slowly because competition has already begun to reinvade
the furrows. On somewhat better sites, these furrows revegetate more

rapidly and more heavily and are likely to cause significant mortality

70

of planted seedlings. Therefore, planting should follow furrowing
within a reasonably short time.

This study examined plantations on a range of textures and on
soils influenced by a water table. There were significant differences
in vegetation and fertility among the different textures and drain-
ages (Table 10); however, site preparation prior to planting apparently
made these differences in site factors unimportant to the survival of

planted seedlings (Table 11).

Table 10. Variables significantly different among texture and
drainage strata in plantation study

 

Litter depth (cm)**
A1&A2 exchangeable magnesium (ppm)**
A1&A2 exchangeable calcium (ppm)***

Sedge cover (%)*

 

*Significant at the .10 level.
**Significant at the .05 level.

***Significant at the .01 level.

Table 11 shows the similarity in survival rates among the
strata. The reasonably high survival percentages (averaging 75.4%)
indicate that planting is a viable regeneration technique on all these
types of sites. The table also shows the high variability in tree
numbers among locations.

Planting therefore offers a simple means of reestablishing

jack pine. Little attention need be given to the site factors,

71

Table 11. Plantation data summarized by location and by stratum

 

 

 

 

 

 

 

 

 

 

 

 

Number of Trees Survival
. Standard Standard
Stratum Location Mean Error Mean Error
(trees/ha) (%)
Coarse-textured CRP l 3733 -- 57.6 --
CRP 2 3246 -- 82.1 --
OTP 2 1388 -- 72.2 --
Total 2789 289.3 70.6 7.10
Medium-textured CRP FH 2437 -- 88.8 --
CRP 5 1519 —- 68.1 ~—
ROP 3 978 -- 79.6 --
Total 1645 172.5 78.8 5.97
Fine-textured CRP 2H 1439 -- 65.6 --
ROP l 747 -— 85.7 --
ROP 4 1776 -- 86.1 --
Total 1321 122.8 79.1 6.78
Water-table— CRP FL 2014 -- 82.6 --
‘nf‘uenced CRP 2L 1010 -- 56.3 --
OTP l 771 -- 80.0 -—
Total 1265 154.1 73.0 8.36
Grand total 1755 112.03 75.4 3.228

 

72

relative to that needed in the use of the natural regeneration methods.
It is more applicable to large areas for this same reason. However,
planting introduces some of its own problems. Poor planting quality
was found by Rudolf (77) to be the major cause of plantation failure-
This is a problem because tree planters are often poorly paid and not
motivated to insure the success of the operation. Poor seedling
quality may be a problem also. This can usually be traced directly
back to the nursery, although shipping and storage of stock may also
be factors. Last, heat and drought will kill large numbers of seed-
lings in some plantations. This factor cannot be controlled, but its
effects can be minimized by creating otherwise ideal conditions for

the survival of the seedling.

SUMMARY AND RECOMMENDATIONS

The primary objective of this study was to identify the site
factors exercising greatest control over jack pine regeneration success.
A second objective was to develop criteria for selection of regenera-
tion method by site types.

Temporary field plots were installed at 36 locations on sand
plain sites throughout the Grayling—Roscommon area. Regeneration
success was measured by counting all living and dead seedlings on the
plots. The seedlings were cut and aged, except on the plantations.
Vegetation and soils were described and soil samples taken. Physical
and chemical properties of the soils were determined in the laboratory.
'Multiple regression equations were developed, analyses of variance
performed, and correlations computed to determine what site factors
were related to regeneration success.

Jack pine regeneration can be obtained only by meeting the
requirements of a series of three steps in the process of seedling
establishment. These steps are:

1. Seed dispersal--seed availability often limits regeneration
success because of seed locked up in serotinous cones, because of seed
mortality due to slash burning, or because of a lack of seed in the
parent stand.

2. Germination--germination depends upon water availability

during periods of warm temperatures. The water requirement for

73

74

germination may not be met because of a lack of water due to inadequate
precipitation or because of seedbed conditions decreasing the availa-
bility of water to the seed.

3. Establishment--after germination, the seedling's taproot
must quickly penetrate the litter to reach a more reliable source of
moisture. However, it often encounters intense competition with under-
story species for the moisture in the surface soil, so the seedling
must root even deeper, below the reach of the roots of the understory
plants. If water becomes limiting at any time before the seedling
penetrates the litter and the zone of intense competition, the seedling
is not likely to survive.

Planting of seedlings is often more reliable than the natural
regeneration techniques because the first two steps described above
are not left to chance. This study found planting to be a viable
regeneration technique on all of the site types likely to be encountered
on the outwash sands of the Grayling-Roscommon area. A stratification
by texture and drainage failed to find differences in stocking or in
survival, although significant fertility and vegetation differences
were found among the strata. Survival averaged 75.4% on the planta-
tions sampled and average stocking was 1755 trees/ha (711 trees/acre).
Stocking was highly variable, ranging from 747 trees/ha (303 trees/acre)
to 3733 trees/ha (1512 trees/acre).

Stocking on the partial cut areas averaged 1346 trees/ha
(545 trees/acre). A minimum acceptable stocking level of 2500 trees/ha
(1012 trees/acre) was suggested. Two of the partial cut locations

were able to meet this stocking requirement (Table 3). Seedling numbers

75

were found to be positively related to percent bearberry, lichen, and
sweet-fern cover, probably because the seedling is better able to com-
pete with these species. Seedling numbers were inversely related to
overstory basal area. Fertility, as measured by pH of the 822 and by
total phosphorus and exchangeable calcium of the 821 horizons, was
found to be inversely related to regeneration success. This negative
effect may be a result of the increased ability of associated species
to compete as soil fertility increases.

The partial cut areas usually begin to regenerate within one
or two years after the first cutting. Seedling establishment is then
relatively good for six to seven years, after which it gradually
decreases. No strong relationships between weather conditions and
year-to-year seedling establishment were found.

The use of partial cuts is likely to often result in failure.
The reasons for the successes of two partial cut locations in this
study are not clear. They are not exceptionally high or low in any
of the site factors found to have significant effects on regeneration.
Because successes are few and cannot be explained when they occur, the
use of these regeneration techniques is not recommended. If, however,
the decision is made to regenerate using partial cut methods, the
highest probability of success exists on those sites with greatest bear-
berry, sweet-fern, and/or lichen cover. If nonserotinous cones are
present, trees bearing such cones should deliberately be left in the
residual overstory. If nonserotinous cones are few or not present,

if sedge cover is heavy, or if a large oak component is present in the

76

stand, partial cut techniques are likely to fail to regenerate jack
pine at an acceptable level.

This study found an average of 7123 trees/ha (2885 trees/acre)
on burned sites with adeqUate seed trees. Seven of the 12 locations
examined had greater than 2500 trees/ha and may, therefore, be con-
sidered successes. Only two of the locations were far below accept-
able stocking levels. This treatment is therefore more likely to be
successful than the partial cut techniques.

Regeneration success on burns is inversely related to sedge
cover. The sod formed by sedge is apparently very competitive with
jack pine seedlings in the surface horizons, killing the seedlings
before they are able to penetrate the sod. Sedge cover seems to be
decreased where fires burn hotter.

Seedling numbers on burns are positively related to bearberry
and sweet-fern cover, as they were on the partial cut plots. They
are also positively related to basal area of the parent stand. This
is attributed to hotter fires, less sedge, and possibly greater seed
availability where stands are denser. A positive relationship between
seedling numbers and 821 horizon thickness and/or profile depth was
also found. Shallower mottles tend to increase regeneration success
as well.

Seedling establishment on these burned sites was found to be at
a maximum two to five years after burning. This regeneration pattern
contradicts much of that found in the literature. It is likely to be

caused by some change in seedbed conditions or by some physiological

77

requirement within the seed that must be met before germination will
occur.

Beaufait (26) suggests that the use of slash burning under
seed trees be avoided in stands with less than 22.9 square meters of
basal area per hectare or yielding less than 48 cm of evenly distributed
slash. He suggests leaving 30 to 123 seed trees per hectare. However,
30 trees per hectare are not considered sufficient on these sites.

The trees left in the overstory for this treatment should be the best
in the stand and able to survive the shock of release, to resist wind-
throw, and to improve the genetic quality of the next rotation. They
should also be selected based on abundance of serotinous cones. If
the stand has a large proportion of nonserotinous cones, other regen-
eration techniques should be considered.

This treatment may be restricted by the danger of potential
wildfires and by the infeasibility of burning large areas. With the
lack of consolidated ownerships on much of the state land in the
northern lower peninsula, escaped fires could be an especially serious
problem. Also, this method is strictly limited by weather conditions
since only a few days of each year are appropriate for controlled
burning.

Potential exists for'utilizing this method for Kirtland's
Warbler management. The warbler nests in clumps of dense regeneration
in areas with a fire history. Rather than managing large areas of
public lands in the range of the warbler for both timber and warbler
habitat, it might be more effective to manage individual areas for

one use or the other. Dense regeneration could be fairly consistently

78

obtained if small, dense, unharvested isolated stands were burned under
dry conditions. The burn would have to kill the overstory, kill as

much of the understory as possible, and release the seed from serotinous
cones in the crowns. Although this method would cause the loss of some
merchantable jack pine timber, it might provide for the needs of the
warbler on smaller areas of land, allowing a timber orientation in the
management of the remaining area, and overall productivities might
increase.

In summary, planting can be expected to succeed on any site
found on the outwash plains of this area, given adequate site prepara-
tion, good planting stock, and a careful planting job. The partial
cut methods are very likely to fail on these sites. Burning without
a harvest or with a partial harvest is more likely to succeed, with the
probability of success increasing with greater stocking in the residual
stand. This method should be restricted to well-stocked stands domi-
nated by trees with serotinous cones.

This study has identified needs for further research in sev-
eral areas. Given the strong effect of competition on survival, the
use of herbicides should be considered. If the competing vegetation
in the A1&A2 horizon is killed, jack pine might be able to root in
this moister, more fertile horizon, and thereby survive and grow
faster. If a sufficient increase in site index and stocking is
obtained, it might make jack pine management a more economical opera-
tion.

The minimum number of trees to be left in stands being

burned as suggested by Beaufait seems to be too low for these sites.

79

Research is necessary to determine the minimum number of trees to be
left.

The use of the scalper-seeder for spot seeding deserves more
study. This method should work with good-quality seed and appropriate
seed treatment, which may be the factors limiting this technique.
Again, the use of herbicides, allowing placement of the seed into the
Al&A2 horizon, would improve the moisture conditions in the soil sur-
rounding the seed, perhaps providing for more reliable germination.

There is a great need to determine the cause of the late ger-
mination observed on the burned sites in this study. If a change in
seedbed conditions is occurring, knowing what that change is might
make it possible to set seedbed requirements for germination. If
there is some physiological reason for the delayed germination of the
seed that can be overcome, it might explain the long history of seed—
ing failures in this area and might bring cheaper methods of regenera-

tion involving seeding back into consideration.

LITERATURE CITED

80

10.

11.

LITERATURE CITED

Abrams, M. D. and D. I. Dickmann. 1980. The effect of prescribed
burning on plant succession on jack pine sites in northern
lower Michigan. Unpub. Progress Report.

Ackerman, R. F. and J. L. Farrar. 1965. The effect of light and
temperature on the germination of jack pine and lodgepole pine
seeds. Tech. Rpt. No. 5, Faculty of Forestry, Univ. of Toronto.

Adams, J. L. 1966. Prescribed burning techniques for site prepa-
ration in cut-over jack pine in southeastern Manitoba. Pul

g9: gaper Magazine of Canada, Woodlands Review, December 1966):

- 84.

Adams, P. W. and J. R. Boyle. 1980. Effects of fire on soil
nutrients in clearcut and whole-tree harvest sites in central
Michigan. Soil Sci. Soc. Am. J. 44:847-850.

 

Ahlgren, C. E. 1959. Some effects of fire on forest reproduction
in northeastern Minnesota. J. For. 57:194-200.

1960. Some effects of fire on reproduction and

 

growth of vegetation in northeastern Minnesota. Ecology 41:
431-445.

1976. Regeneration of red and white pine follow-

 

ing wildfire and logging in northeastern Minnesota. J. For.
74:135-140.

1979a. Buried seed in the forest floor of the

 

Boundary Waters Canoe Area. Minn. For. Res. Notes No. 271.

1979b. Buried seed in prescribe-burned jack pine

 

fbrest soils, northeastern Minnesota. Minn. For. Res. Notes
No. 272.

Ahlgren, I. F. and C. E. Ahlgren. 1960. Ecological effects of
fires. Bot. Rev. 26:483-533.

1965. Effects of prescribed

 

burning on soil microorganisms in a Minnesota jack pine forest.
Ecology 46:304-310.

81

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

82

Alban, D. H. 1977. Influence on soil properties of prescribed

burning under mature red pine.

USDA For. Serv., North Cent.

For. Expt. Sta., Res. Pap. NC-l39.

Amundson, D. C. and H. E. Wright, Jr. 1979. Forest changes in
Minnesota at the end of the Pleistocene. Ecol. Monqgrephs

49:1-16.

 

Anderson, R. C., L. L. Loucks, and A. M. Swain. 1969. Herbaceous
response to canopy cover, light intensity, and throughfall
precipitation in coniferous forests. Ecology 50:255—263.

Anon. 1931. Spring fires on the jack pine plains. USDA For.

Serv., Tech. Notes No. 44.

. 1937. Heat more injurious than lack of moisture during
drought. USDA For. Serv. Tech. Notes No. 125.

1938. Deep rooting and plantation survival. USDA For.

Serv. Tech. Notes No. 140.

1939. Beneficial effect of partial shade on field survival.

USDA For. Serv., Lk. Sts. For. Expt. Sta., Tech. Notes No. 144.

1939. When do jack pine seeds germinate? USDA For. Serv.

Tech. Notes No. 149.

Arend, J. L., N. F. Smith, S. H. Spurr, and J. W. Wright. 1961.
Jack pine geographic variation--five-year results from lower
Michigan tests. Pap. Mich. Acad. Sci,, Arts, and Letters

46:219-238.

Bensend, D. W. 1943. Effect of nitrogen on growth and drouth
resistance of jack pine seedlings. Minn. Agric. Expt. Sta.

Tech. Bu11. 163.

Benzie, J. W. 1968. Regeneration of cutover jack pine stands.
USDA For. Serv. Res. Note NC-49.

 

1977. Manager's handbook for jack pine in the

north central states. USDA For. Serv. Gen. Tech. Rpt. NC-32.

Beaufait, W. R. 1960a. Crown temperatures during prescribed
burns in jack pine. Proc. Mich. Acad. Sci.) Arts and Letters.

46:251-258.

1962a. Some effects of high temperatures on

 

cones and seeds of jack pine.

For. Sci. 6:194-199.

L,“

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

83

1962b. Procedures in prescribed burning for

 

jack pine regeneration. Mich. Coll. of Mining and Technology
Tech. Bull. No. 9.

Beaufait, W. R. and R. T. Brown. 1960b. Phyteogeography of a
representative jack pine site. Mich. Acad. Sci., Arts, and
Letters 47:201-209.

 

Burns, R. M. and E. A. Hebb. 1972. Site preparation and refores-
tation of droughty, acid sands. USDA Agr. Handbk. 426.

Cameron, H. 1953. Melting point of the bonding material in
lodgepole pine and jack pine cones. Can. Dep. Res. Dev.,
For. Br., For. Res. Div., Silv. Leafl. 86.

Candy, R. H. 1951. Reproduction on cut-over and burned-over
land in Canada. Can. Dep. Res. Dev., For. Br., For. Res. Div.,
Silv. Resource Note No. 92.

Caveney, E. W., Jr. 1969. Reproducing jack pine by the shelter-
wood method in northern lower Michigan. Unpub. M. S. thesis,
Mich. State Univ.

Cayford, J. H. 1963. Some factors influencing jack pine regen-
eration after fire in southeastern Manitoba. Can. Dep. For.
Publ. 1016.

1966. Operational trials of regeneration methods

 

for jack pine in southeastern Manitoba. Can. Dep. For. Pub.
1165.

1970. The role of fire in the ecology and silvi-

 

culture of jack pine. Proc. Tall Timbers Fire Ecol. Conf.
10:221-244.

Cayford, J. H. and J. M. Jarvis. 1963. Furrowing and sheltering
to improve early survival of planted red pine on dry sites,
southeastern Manitoba. Tree Planters' Notes 59:21-24.

Cayford, J. H., 2. Chrosciewicz, and H. P. Sims. 1967. A review
of silvicultural research in jack pine. Can. Dep. For. and
Rural Dev., For. Br., Dep. Pub. No. 1173.

Chrosciewicz, Z. 1963. The effects of site on jack pine growth
in northern Ontario. Can. Dep. For., Publ. 1015.

1967. Experimental burning for humus disposal

 

on clear-cut jack pine sites in central Ontario. Can. Dep.
For. Rural Dev., For. Br., Publ. No. 1181.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

84

1974. Evaluation of fire-produced seedbeds

 

for jack pine regeneration in central Ontario. Can. J. For.

Res. 4:455-457.

Cooley, J. H.
survival of
No. NC-159.

 

sands. USDA For. Serv. Res. Note No. NC-l38.

Darlington, H.

 

1971. Planting technique and care of stock affect
planted red pine. USDA For. Serv. Res. Note

1972. Site preparation for jack pine on Grayling

T. 1945. Taxonomic and ecological work on the

higher plants of Michigan. Mich. State College, Agr. Expt.

Sta., Tech.

Bull. 201.

Day, P. R. 1965. Particle fractionation and particle-size
analysis. Ig_C. A. Black (ed.) Methods of soil analysis.
Part 1. Agronomy 9:545-566. Am. Soc. of Agron., Madison,

Wis.

De Bell, 0. S.

and C. W. Ralston. 1970. Release of nitrogen by

burning light forest fuels. Soil. Sci. Soc. Am. Proc. 34:

936-938.
Ellis, L. M. 1

 

911. Some notes on jack pine (Pinus divaricata)

 

in western Ontario. Forestgy Quart. 9:1-14.

 

Eyre, F. H. 1936. Dispersal of jack pine seed from seed trees

and slash.

 

J: For. 36:1

Mich. Acad. Sci.,eArts, and Letters Pap. 21:279-384.

1938. Can jack pine be regenerated without fire?

067-1072.

Eyre, F. H. and R. K. LeBarron. 1944. Management of jack pine

in the Lake

Harper, R. M.

States. USDA Tech. Bull. 863.
1918. The plant population of northern lower

Michigan and its environment. Torrey Bot. Club, Bull. 45:23-42.

Hodges, J. P. and R. L. Scheer. 1962. Soil cover aids germination
of pine seeds on sandy sites. USDA For. Serv., Tree Planters'
Notes 54:1-3.

 

Hyun, J. O. 1977. Geographic variation of jack pine (Pinus
banksiana Lamb.). Proc. Lake States Tree Improvement Conf.

13:|07-|16.

Isaac, R. A. and Johnson. 1976. Journal of the AOAC, Vol. 59.

from Anon.

1978. Digestion of plant tissue for determination

of total nitrogen and/or phosphorus with Tecator block digestor

model DS-40.

Tecatopic 4/Rev l, Tecator, Inc.

-//”

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

85

Isaac, R. A. and J. D. Kerber. 1971. Atomic absorption and flame
photometry techniques and uses in soil, plant, and water analy-
sis. In L. M. Walsh (ed.) Instrumental methods for the analysis
of soil? and plant tissue. Soil Sci. Soc. Am.

Jameson, J. S. 1961. Observations on factors influencing jack
pine reproduction in Saskatchewan. Can. Dep. For., For. Res.
Div., Tech. Note No. 97.

1963. Relation of jack pine height-growth to

 

site in the mixedwood forest section of Saskatchewan. Can.
Dep. For., For. Res. Br., Contribution No. 564.

Lambert, J. L., J. R. Boyle, and W. R. Gardner. 1972. The
growth response of a young pine plantation to weed control.
Can. J. For. Res. 2:152-159.

 

LeBarron, R. K. 1944. Influence of controllable environmental
conditions on regeneration of jack pine and black spruce.
J. Agr. Res. 68:97-119.

LcBarron, R. K. and F. H. Eyre. 1938. The influence of soil
treatment on jack pine reproduction. Pap. Mich. Acad. Sci.
23:307-310.

 

Logan, K. T. 1966. Growth of tree seedlings as affected by light
intensity. II. Red pine, white pine, jack pine, and eastern
larch. Publ. Dep. For. Can. No. 1160.

Lopushinsky, W. and G. 0. Klock. 1974. Transpiration of conifer
seedlings in relation to soil water potential. For. Sci.
20:181-186.

Lotan, J. E. 1974. Cone serotiny--fire relationships in lodgepole
pine. Proc. Tall Timbers Fire Ecol. Conf. 14:267-278.

MacArthur, J. D. 1965. Planting methods to overcome strong com-
petition form dense, herbaceous vegetation. Can Dep. For.,
For. Res. Br., Contribution No. 638.

MacHattie, L. B. and K. W. Horton. 1963. Influences of micro-
climates on mortality and growth of planted white spruce, jack
pine and white pine. Can. Dep. For., For. Res. Br., Contribu-
tion No. 511.

Maissurow, D. K. 1941. The role fire in the perpetuation of
virgin forests of northern Wisconsin. J. For. 39:201-207.

Martin, R. E. and L. S. Davis. 1960. Temperatures near the
ground during prescribed burning. Proc. Mich. Acad. Sciag
Arts and Letters 46:239-250.

 

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

86

McCumber, J. E. 1979. Timber management guide for the Huron-
Manistee jack pine type. Unpub. Rpt., Huron-Manistee National
Forest, USDA For. Serv.

Miller, E. L. and G. Schneider. 1971. First year growth response

of direct seeded jack pine. Mich. St. Univ., Ag. Expt. Sta.
Res. Rpt. 130.

. 1971. Container-grown jack pine
establishment on modified sites in lower Michigan. Mich. St.
Univ. Ag. Expt. Sta., Res. Rpt. 137.

 

Mullin, R. E. 1979. Plantation performance averages for jack
pine. Ontario Ministry of Nat. Res., For. Res. Note No. 20.

Nie, J. 0., C. H. Hull, J. G. Jenkins, K. Steinbrenner, and
D. H. Bent. 1975. SPSS:Statistical package for the social
sciences. (2nd ed.) McGraw—Hill Book Co., New York.

Noakes, J. W. 1946. Effect of different methods of slash
disposal on jack pine reproduction. Can. Dep. Mines and Res.,
Lands, Parks, and Forests Br., Silv. Res. Note No. 78.

Nordin, J. 0. and D. F. Grigal. 1976. Vegetation, site and fire
relationships within the area of the Little Sioux fire, north-
eastern Minnesota. Can. J. For. Res. 6:78-85.

 

Pawluk, S. and H. F. Arneman. 1961. Some forest soil character-
istics and their relationship to jack pine growth. For. Sci.
7:160-173.

Pereira, J. S. and T. T. Kozlowski. 1977. Water relations and
drought resistance of young Pinus banksiana and Pinus resinosa
plantation trees. Can. J. For. Res. 7:132-137.

 

Ralston. R. A. 1951. Successful natural jack pine regeneration.
USDA For. Serv., Lk. Sts. Expt. Sta., Tech. Notes No. 363.

Rudolf, P. 0. 1937. Lessons from past forest planting in the
Lake States. J. For. 35:72-76.

. 1939. Why forest plantations fail. J. For.
:3 - 83.

. 1950. Forest plantations in the Lake States.
USDA Tech. Bull. No. 1010.

. 1958. Silvical characteristics of jack pine.
USDA For. Serv., Lk. Sts. For. Expt. Sta., Sta. Pap. No. 61.

 

)

80.

81.

82.

83.

84.

85.

86.

87.

89.

90.
91.
92.

93.

94.

87

Schantz-Hansen, T. and R. A. Jensen. 1952. The effect of source
of seed on growth of jack pine. J. For. 50:539-544.

Schoenike, R. E., T. D. Rudolph, and T. Schantz-Hansen. 1959.
Cone characteristics in a jack pine seed source plantation.
School of For., Univ. of Minn., Minn. For. Note No. 76.

Scholl, D. G. 1966. The effect of prescribed burning on the
properties of the Grayling soil series in upper Michigan.
Unpub. M. S. thesis, Mich. St. Univ.

Severson, R. C., D. F. Grigal, and H. F. Arneman. 1975. Perco-
lation losses of phosphorus, calcium, and potassium from some
Minnesota soils. Soil Sci. Soc. Am. Proc. 39:540-543.

 

Shetron, S. G. 1972. Forest site productivity among soil
taxonomic units in northern lower Michigan. Soil Sci. Soc.
Am. Proc. 36:358-363.

 

1974. Distribution of free iron and organic

 

carbon as related to water in some forested sandy soils.
Soil Sci. Soc. Am. Proc. 38:359-362.

Sims, H. P. 1964. Root development of jack pine seedlings on
burned-over dry sites in southeastern Manitoba. Dept. For.,
Canada, Publ. 1061.

1975. Temperature and moisture conditions on a

 

plowed jack pine strip cut in southeastern Manitoba. Can. J.
For. Res. 5:541-545.

Sims, H. P. and D. Mueller-Dombois. 1968. Effect of grass com-
petition and depth to water table on height growth of coniferous
tree seedlings. Ecology 49:599-603.

Smith, D. W. 1970. Concentrations of soil nutrients before and
after fire. Can. J. Soil Sci. 50:17-29.

 

Soil Survey Staff. 1976. Grayling series.

1977. Graycalm series.

 

Soper, E. K. 1919. The peat deposits of Minnesota. Minn. Geol-
Surv. Bull. 16.

Steel, R. G. D. and J. H. Torrie. 1960. Principles and proce-
dures of statistics. Hill Book Co., New York.

Sterret, W. D. 1920. Jack pine. USDA Bull. 820.

95.

96.

97.

98.

99.

100.

101.

102.

103.

88

Tryon, E. H. 1948. Effect of charcoal on certain physical,
chemical, and biological properties of forest soils. Ecol.
Monog. 18:81-115.

Urie, D. H. 1959. Pattern of soil moisture depletion varies
between red pine and oak stands in Michigan. USDA For. Serv.,
Lk. Sts. For. Expt. Sta., Tech. Notes No. 564.

Van Wagner, C. E. 1966. Three experimental fires in jack pine
slash. Can. Dep. For., Dep. For. Pub. No. 1146.

Voss, E. G. and G. E. Crow. 1976. Across Michigan by covered
wagon: A botanical expedition in 1888. The Michigan Botanist
15:3-70.

Wells, C. G., R. E. Campbell, L. F. DeBano, C. E. Lewis,
R. L. Fredericksen, E. C. Franklin, R. C. Froelich, and
P. H. Dunn. 1979. Effects of fire on soil. USDA For. Serv.
Gen. Tech. Rpt. WO-7.

Whiteside, E. P., I. F. Schneider, and R. L. Cook. 1956. Soils
of Michigan. Mich. St. Univ. Ag. Expt. Sta. Special Bull. 402.

Wilde, S. A. 1964. Changes in soil productivity induced by pine
plantations. Soil Sci. 97:276-278.

1966. Soil standards for planting Wisconsin coni-

 

fers. a: For. 64:389-391.

Wilde, S. A., J. G. Iyer, C. Tanzer, W. L. Trautmann, and
K. S. Watterston. 1964. Growth of jack pine, Pinus banksiana
Lamb., plantations in relation to fertility of non-phreatic
sandy soils. Soil Sci. 98:162-169.

APPENDIX A

SCIENTIFIC NAMES 0F SPECIES ENCOUNTERED
IN THIS STUDY

89

Tree and shrub species1

APPENDIX A

SCIENTIFIC NAMES 0F SPECIES ENCOUNTERED

 

 

1. Alder

2. Aspen

3. Balsam fir

4. Basswood

5. Beaked hazel

6. Black spruce

7. Bur oak

8. Eastern white pine
9. Hazel
10. Jack pine
11 Lodgepine pine
12. Northern pin oak
13. Northern red oak
14. Paper birch
15. Red pine
16. Sugar maple
l7. Trembling aspen
18. White spruce

1
(5th ed.).

IN THIS STUDY

Alnus rugosa (L.) Gaertn.

Populus spp.
Abies balsamea (L.) Mill.

 

Tilia americana L.

Corylus cornuta Marsh.

 

Picea mariana (Mill.) B.S.P.

 

Quercus macrocarpa Michx.

 

Pinus strobus L.

 

Corylus americana Walt.

 

Pinus banksiana Lamb.

 

Pinus contorta Dougl.

 

Quercus ellipsoidalis E.J. Hill

Quercus rubra L.

 

Betula papyrifera Marsh.

Pinus resinosa Ait.

 

Acer saccharum Marsh.

 

Egpulus tremuloides Michx.

Picea glauca (Moench) Voss

Harlow, W. M. and E. S. Harrar. 1969. Textbook of Dendrology
McGraw-Hill Book Company, New York.

90

8. Ground Cover Species

2

 

l.

._J ._I _I .—a ._a ....1 ....1
O5 0" b 00 N -‘
. C C O . C

17.

O NO CD \4 01 01 h to N
o e e o o o o e o

Bearberry

Blue joint grass
Blueberry
Bracken fern
Bunchberry

Dry land sedge
False lily-of-the-valley
Fireweed
Labrador-tea
Leatherleaf
Raspberry
Reindeer lichen
Sand cherry
Sedges
Sweet-fern
Twinflower

Wintergreen

C. Insect Species3

 

1. Jack pine budworm
2. Sawfly
2Fernald, M. L. 1950.
Company, New York.
3viison, L. F. 1977.

the Lake States.

91

Gray's Manual of Botany.

Arctostaphylos uva-ursi (L.) Spreng-

 

Calamagrostis canadensis (Michx.)

 

Vaccinium spp.

Pteridium eguilinum (L.) Kuhn

 

Cornus canadensis L.

 

Carex peneylvanica Var. distans Peck

 

Maianthemum canadense Desf.

 

Epilobium angustifolium L.

Ledum ggoenlandicum Oeder

 

Chamaedaphne calyculata (L.) Moench

Rubus idaeus L.

 

Cladonia spp.

Prunus pymila L.

 

Carex spp.

Comptonia peregrina (L.) Coult.

 

Linnaea borealis L.

 

Gaultheria procumbens L.

Choristoneura pinus Freeman

Neodiprion lecontei (Fitch)

American Book

A guide to insect injury of conifers in
USDA For. Serv., Ag. Handbk. No. 501.

92

 

 

D. Pathogen species4
1. Pine-oak rust - Cronartium quercuum (Berk.)
4

Ibid.

APPENDIX 8

UNIT CONVERSIONS

93

APPENDIX 8

UNIT CONVERSIONS

English to Metric

 

1 inch

1 foot

1 chain

1 acre

1 pound

l pound/acre
1 foot 2lacre

2.540 centimeters
0.3048 meters

20.12 meters

0.4047 hectares

0.4536 kilograms

1.121 kilograms/hectare
0.2294 meters 2/hectare

Metric to English

1 centimeter

1 meter

1 meter

1 hectare
lkilogram

1 kilograglhectare
1 meter /hectare

0.3937 inches

3.281 feet

0.04971 chains

2.471 acres

2. 205 pounds

0.8921 pougds/acre
4359 feet /acre

94

   

"llllllllllliii“