3453
..1:1.H...HMM...111.11. 4.. fix. .1:. 1. -
.P‘u‘boo r6111.- c‘hflnn‘vo . c n.» .. .- .. 1 . . . .. . .:.. .1. . - . . _, . ..: . :. 1. . . . .
h .: mono-1M“! can. 1 -w . ... . . . . o 2 . 2.: . .. .. 1... .. h .1. ..:.HnWLHw‘Q‘JIF-: .
0‘ I 9 n I | 1 .‘
. 1

. .
I .: 1N: . : 1
1.: Rump-4.. .. .
Q- Q C I 0v. ‘ O
13.5....“ . .
z.o-v.ttolco.1 I.» 1:.
1.1.1....:...-1..... ha... .L..: 1......
. .1 1 us; ulna:
KY... .1...-..:1.1._..1...1u....:.:m1..1. ... .. .
. w . n. . - . . - .
L. .- ..nfiwmn .....:...:..-1.. he:
1.1“»... . .1 $534.
. .
\ v1 .‘fl. .un¢

. 4 L.)
. . . ..:... :1

..zuw.

.
1
.
. .
1
.1

0

”317521:
3‘1.’ , .
15' l

. .1
. guy-WA“. «mu-“wt.
. . kfimlmbauu-H tannin .
. ”WW1. . ..:...wfiiflmmufismfi
1.! 11.. .1 5.1.11: 91?... .
.. .. clad.“ 1.....a1M1nuf . .1 ..:. .
. 1. :.-1~...u:.1: 2.11.6. Via...» Rummy”...
«wax. . Wt;....-..W»u...1..:1n.
Chou... 11.1.«1wc...:.1......1.a. .. . . . :1
..:“...hrbmfld . . . . . . . . ..1

gr- 1

3-2;}:

a, .1 ,‘. ,
if. ‘9
«#13111:

.

 

‘10:}
1:11

i
a
I.

. 1 ..
. .1..n...~...r.:
. a - QUIVV‘AN:

a...

11"
.
.

f
:
.
‘1.

A . 1
I 0 A v! I O L
a Q ~ . A At A 4 o u d... W \<
.11 - I .. . . : .. . . .- .1 1 t
D n. 1 .. . . .. .
0 0| . 1 . . o v. . . . H Oh'it .
o L V . “‘| i.
a b 1 .- 1 .. q 1.. u . o . 11.6.91
:0 0114011.?!“ . ... . . . . . 1..
15.qu 1... . . : .. . .
h. 1.1.111.“ .I . -. . _ .. . . 1. :. - . . .
. ‘11.“ . . .. .. . .. . . . 1.: . . . - . . . .
”w: ..: : . . , .. . .. .. : .
A u m1 c Q I I C I
h a I . 1 .t
.. 1 . . .

p
. ,

'1'
1.
v
:11
9‘.
I

s
.1
.1-
.21

rfic. .
1~ :
1:;

1

s.

’l
‘3
a

. 1_
.1.
v
v

a

1

i
. . 4.141.. ..
b3 lit. 110'. 4 .
1r1mwfi1m £1.24 ..
. m. .1. ..1.1.-»......-:.¢H.N.fi. . .
:11 5.. .1. ..

 

.
«kt . u 1.
. . .
1! .vhzhtln:
v o ..:o 1(‘1
Wk... 1111..

o u. 1
3333 . 0‘1: . .
'1

. .
.
14.111". . ... . . . .. r .. ”2 . . Q
.17.. : . . . . . . . .1 .
.. .1,.. I. . -.
I ..
I u v
. .

.
1
9:11- 4. V

 

.

111143.
z; .
.1

F...

. Vito.
.1315. .
3;;

“13.1 -
.111: 11 (1111111 «4:51!- ..
1.. '90 1'11: .1! ‘l A lllnffi
0 \ .Vozf Li'vo.lCva
- 1.»...1.\11I11. 11141111113 1
1h-11. :1- ...- . -.
..14..1:1-. 1.11-1.11. 1.111..
1.111 1*- 11.1...~.11-..1.J.1..
I... .111- .11111: 3
11.141. 1112....-. .. 1:.- .
71'“ c4 .11... A! Ll \ “.11....
11.1111 1 111.1 1 1- 11. b1
. 1 11411- 111 .111
9.41:. Sud-.1111 ....111:s.1.11-1r 1.1. ...1. .
.. :uu11-H1u.1.“1: 1. .11-1n- 11111111111121 -

I . “(11411133051011 -

 

 

1 :1

. .1.

.0 11.1145

II‘CWU'UM

..1I1lA .
.lltv1wfif
I1!

 

. . ...1- .u . . con ..
- '1 . ¢ - . ~n ..: . .1 I. . 1 ATV? -. ..:-1‘
: 10.60. c ..- . . - 1 .. . . - .1 . 13110 1.1.9.1 .I -.
1:11:41 : . . . 1 . . . .. .. . .:..-.-..1:1:-111.1.:.
' 11:14 1 1 .1 . u 1 1 1 .. .11.nv 1 .
(U! c 31.11.. - . .. . u . a 1 1 1
1.1.1.011 . 1.. . . . l 1:.

111-111111111-.1..11. - 1! 1 . .

1..1h1.o.u.al-.:.<1.nf\h.l.11 .

0141111. Un.¢1.,1.1 )0... , . .
I Iv ‘t 1 t , . o

. 1- 1.11.- 11u1vfl11111..‘u.u:.111.1.L.-J.

[’ “|l‘ .o-‘1 ‘1). .0 I.
1111;: .1311 1-3.

11 1-,. .
. .11...

 
 

{.1
'1’ «.1. tt‘U~ - ~. 1 . n 1
(1.1 .2 . 1. 1.1411

1. .hll.F».:1.- .. .. . 1 . : .111»

$1. . 10 0:1 . . _ :1! 31.1. . . . . . z . 1 ..hfs,;\- OTI

s u . - . 1... .- .- $11.81.: 111

It. (1H1 “2.11..“ to. . . . -. . . : . 1 . . . .1113 \ 11130.1 )1

. .-. . . 1 . . . . - 1 1

0 ul .11WV' \ o . u .v 1- :o : . ..1 .~: 0.
. . - u . .

‘9 ..I
.

 

 

:

.
. - . 111'11111‘L-1' ”fluvial . 1 : . .. . J : . . - n . ..c. 1 f . ..u : fidyflfiw

..I- jfl-fi-cilfltl..4 .11."..1). : 11.1. . - . 1 v1.16. - . . ..: . . . . . . .-.. . . . 111-: ..1

-- . 51-11 .1... 1 if“? ..:-1 - . . . .. . .. .

u .- Iv; :‘1 . .1 1

meets , 1‘
l 75"“ #5 fl. tre‘rr
" Liuugwsfid

I

resin: ‘4' irate
L Unwmufy

”TI

This is to certify that the

thesis entitled

FIRST-YEAR RESPONSES OF WILDLIFE AND WILD-
LIFE HABITAT TO SEWAGE SLUDGE APPLICATION
IN A NORTHERN HARDWOODS FOREST

presented by

Anne Husted Thomas

has been accepted towards fulfillment
of the requirements for

 

M.S. degree In Fisheries and

 

Wildlife

 

Date May 6, 1983

 

0-7539 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

MSU RETURNING MATERIALS:
Place in book drop to
LIBRARIES remove this checkout from
__. your record. FINES will

 

 

be charged if book is
returned after the date
stamped below.

 

a.“ 4 1‘ -. 5-: gr 9*”

~_.. ..:/4"? "u it in

 

 

r“.

0 /§/€’) ~ for:

d

FIRST-YEAR RESPONSES OF WILDLIFE
AND WILDLIFE HABITAT TO SEWAGE
SLUDGE APPLICATION IN A NORTHERN

HARDWOODS FOREST

V

/?

by

Anne Husted Thomas

A THESIS

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

MASTER OF SCIENCE

Department of Fisheries and Wildlife

1983

ABSTRACT
FIRST-YEAR RESPONSES 0F WILDLIFE
AND WILDLIFE HABITAT T0 SEWAGE '

SLUDGE APPLICATION IN A NORTHERN
HARDWOODS FOREST

BY
Anne Husted Thomas

A single application of sewage sludge was applied to a
50-year old northern hardwoods forest in Montmorency County,
Michigan, in July 1982. The application technique required
construction of S-m wide trails at 20 m intervals through-
out the forest. Vegetative community composition and
structure were analyzed, small mammal communities monitored
via live-trapping, and red-backed salamanders censused on
9 study plots (3 with trails, 3 with sludge application and
trails, and 3 controls).

First year results indicated that tree seedling densities
were lowered in trails by trail construction and use, and
in forest interiors by sludge damage. Vertical vegetative
cover was reduced in tall strata by tree removal for trail
construction, and in low strata by sludge damage. Trails
appeared to provide suboptimal salamander above-ground
foraging habitat; but did not affect small mammal habitat
use. Peromyscus mice increased in number on sludge—treated
plots, compared to trail-only and control plots, presumably
due either to differential survival or to a behavioral re-

sponse to some change in their environment.

ACKNOWLEDGEMENTS

There are two people without whom this thesis would
not have been written. I am deeply indebted to my advisor,
Dr. Jonathan Haufler, for his expert guidance and for his
faith, support, and friendship. I would also like to
thank my father, Dr. William Thomas, whose fine example
as a naturalist, combined with loving encouragement were
invaluable in inspiring my work.

A field study could rarely take place without the
help of many people, but in this case a truly herculean
task was accomplished only with the concerted effort of
a uniquely hardeworking, cheerful field crew. I extend my
deepest appreciation to Jack Dingledine, Linda Doolittle,
Ginny Lilly, Elena Seon and Dave Woodyard for the many
days and nights they gave so willingly to my project, and
for the friendship which made the work more enjoyable.

I thank the members of my committee, Dr. Ben Peyton
and Dr. Donald Straney, for sharing their enthusiasm,
expertise, time and support. They were everthing one
hopes for in a committee, and more. I am also grateful to
Drs. Thomas Burton and Phu Nguyen, who contributed valuable

advice and enthusiasm. Dr. Daniel Talhelm has devoted

much energy and a listening ear.

ii

Special thanks are due Rique Campa, Stan.Koster,
Dave Gordon and Elena Sean, and the many others whose
affection and friendly criticism.and support have en-
riched my years at MSU.

My friend Dave Luukkonen has contributed immeasurably
in time, energy, critical advice and kind understanding,
and has had a major influence on the completion of this
thesis. .

Thanks, finally, to Ms. Susan Hazard for her expert
typing.

This project was funded by the U.S. Environmental

Protection Agency.

iii

TABLE OF CONTENTS

Page
LIST OF TABLES ...................................... .v
LIST OF FIGURES ..................................... viii
INTRODUCTION ........................................ 1
OBJECTIVES .......................................... 7
SITE DESCRIPTION .................................... 8
METHODS AND MATERIALS ............................... 12
Experimental Design ............................ 12
Trail and Sludge Treatment ..................... 12
Vegetative Community Composition ............... 16
Vertical Vegetative Cover ...................... 17
Annual Plant Production ........................ 18
Small Mammal Populations ....................... 19
Data Analysis .................................. 21
RESULTS ............................................. 24
Vegetative Community Composition ............... 24
Foliage Height Diversity ....................... 44
Plant Annual Production ........................ 44
Small Mammal Community ......................... 48
Salamander Populations ......................... 54
DISCUSSION .......................................... 56
Vegetative Community Response .................. 56
Small Mammal Response .......................... 63
Salamander Response ............................ 69
SUMMARY AND RECOMMENDATIONS ......................... 71
LITERATURE CITED .................................... 75
APPENDIX ............................................ 80

iv

LIST OF TABLES

Number Page

1 Mean element concentrations in wet sludge from
Rogers City, MI, and mean loading levels
applied to the soil on the northern hardwood
study site, Montmorency County, MI in July 15
1982 ............................................

2 Density (stems/ha) of all woody plants in entire
plots (interiors and trails) per size class:
for controls, plats with trails, and plots
with trails and sludge application in
Montmorency County, MI, July, 1982 .............. 25

3 Density (stems/ha) of each species in the O-BOcnn
size class in interiors: for controls, plots
with trails, and plots with trails and sludge
iggéication, in Montmorency County, MI, July 27

4 Density (stems/ha) of each species in the 0-
30 cm size class in trails: for controls,
plots with trails, and plots with trails and
sludge application, in Montmorency County, MI,
July 1982 ....................................... 28

5 Density (stems/ha) of each species in the 30¢nn-
l m size class in interiors: for controls,
plots with trails, and plots with trails and
sludge application, in Montmorency County, MI,
July 1982 ....................................... 29

6 Density (stems/ha) of each species in the 30¢nn-
l m size class in trails: for controls, plots
with trails, and plots with trails and sludge
aggéication, in Montmorency County, MI, July 31

7 Density (stems/ha) of each species in the la:-
2 m size class in interiors: for controls,
plots with trails, and plots with trails and
sludge application, in Montmorency County,
MI, July 1982 ................................... 32

3 Density (stems/ha) of each species in the >2 m
tall, <10 cm dbh size class in interiors: for
controls, plots with trails and plots with
trails and sludge application, in Montmorency
County, MI. July 1982 .......................... 33

Number

10

ll

12

13

14

15

16

17

Density (stems/ha) of each species in the

>2 m tall, >10 cm dbh size class in in-

teriors: for conjrols, plots with trails,

and plots with trails and sludge application,

in Montmorency County, MI, July 1982 ........... 34

Density (stems/ha) of each species in the

0-30 cm and 30 cm - 1‘m size classes, in

interiors and trails, for plots with trails-

only, Montmorency County, MI, July 1982 ....,... 36

Density (stems/ha) of each species in the 0-30

cm and 30 cm - l'm size classes, in interiors

and trails, for plots with trails and sludge
application, in Montmorency County, MI,

July 1982 ...................................... 37

Foliage height diversity (Shannon Wiener Index
values) for control plots, plots with.trails,

and plots with trails and sludge application,

in Montmorency County, MI, July 1982 ........... 45

Above ground net annual production (kg/ha) <2un

for controls, plots with trails, and plotS‘

with trails and sludge application in

Montmorency County, MI, August 1982 ............ 46

Above ground net annual production (kg/ha)

<2m, from interiors: for controls, plots

with trails and sludge application in

Montmorency County, MI, August 1982 ............ 47

Above ground net annual production (kg/ha)

<2 m, from trails: for controls, plots with

trails and sludge application in Montmorency
County, MI, August 1982 ........................ 49

Total small mammals known to be alive in August
1981, in controls, plots designated to have

trails, and plots designated for trails and

sludge application in Montmorency County, MI ... 50

Average number of small mammals known to be

alive in August, 1981, major species only, in
controls, plots designated to have trails,

and plots designated for trails and sludge
application in Montmorency, County, MI ......... 51

vi

Number

18

19

20

Total Small mammals known to be alive in
summer, 1982, by species, in controls,

plots with trails, and plots with trails and
sludge application in Montmorency County, MI

Average number of small mammals known to be
alive in summer, 1982, all species combined,
in controls, plots with trails, and plots
with trails and sludge application in

Montmorency County, MI .......................

Average number of small mammals known to be
alive in summer, 1982, major species only,
in controls, plots with trails, and plots
'with trails and sludge application in

Montmorency County, MI .......................

vii

.52

...53

...55

Number

LIST OF FIGURES

Mmp showing location of study site in
Mbntmorency County, MI ...................... 9

Average precipitation and temperature, by
'month for normal period and study period,
1981-1982, Atlanta, MI

Study plot arrangement, northern hardwoods
study site, Montmorency County, MI .......... 13

Frequency of commonly occurring herbs and

shrubs (Z of quatrats in which the plant
occurred) in control plots, plots with

trails, and plots with trails and sludge
application, Montmorency County, MI,

July 1982 (X with SE) ....................... 38

Percent vertical vegetative cover in 4

height strata in entire plots: for control
plots, plots with trails, and plots with

trails and sludge application, _

Monomorency County, MI, July 1982 (X

with SE) .................................... 40

Percent vertical vegetative cover in 4

height strata in forest interiors: for

control plots, plots with trails, and plots

with trails and sludge application,

Montmorency County, MI, July 1982 (X with 42
SE .........................................

Percent vertical vegetative cover in 4

height strata in application trails: for

control plots, plots with trails, and

plots with trails and sludge application,
ggntmorency County, MI, July 1982 (X with 43

viii

INTRODUCTION

Sewage waste disposal poses a growing problem for many
municipalities. Technological advances can only partially
solve these problems by modifying processes to reduce the
amount of waste generated and to make it less harmful to
humans and to the environment. Advanced secondary sewage
treatment uses filtration and aerobic or anaerobic bacterial
digestion to break down organic material to its inorganic
constituents. The process destroys or immobilizes most
toxins, pathogens and parasites (Sopper and Kardos 1979).
Much water can be reclaimed from sewage, but the remaining
solids must be dealt with. Final sewage solids, called
sludge, can be disposed of by: (1) discharge into rivers,
lakes, or oceans, (2) incineration, (3) landfill, or
(4) land application. Federal clean water criteria dictate
against sludge disposal in any manner that might affect
the inland or coastal navigable waters of the United States
(Schmid et a1. 1975). Offshore ocean dumping is un-
desirable, since sludge constituents reduce seabed oxygen
and in other ways affect benthic organisms (National
Academy of Sciences 1978). Incineration leaves noncombus-

tible material which must be disposed of, wastes raw

materials, and creates air pollution (Turk et al. 1972,
National Academy of Sciences 1978). Landfill, if properly
done, can be a safe sludge disposal method (National Aca-
demy of Sciences 1978). As with incineration, however,
landfill wastes raw materials. In addition, safe land-
fill sites, with no threat of groundwater contamination,
are scarce. Burial necessitates removing existing vege-
tation and stockpiling an amount of excavated material
equal to the volume of sludge to be buried. Stockpile
sites also will be temporarily devoid of vegetation
(Schmid: et a1. 1975). Thus, available sites meeting all
environmental and economic criteria for landfill are
quite limited.'

Sludge disposal by land application is the remaining
alternative. Terrestrial ecosystems can act as living
filters for the compounds and elements in sludge (Sopper
1975), while sludge-borne nutrients can promote plant
productivity. Sludge or sewage effluent has been applied
to agricultural lands, often resulting in increased yield
and higher nutrient content of crop plants (Sopper and
Kardos 1979). However, for human health and aesthetic
reasons, and because they are less effective in retaining
the mineral elements applied (WOodwell 1977), agricultural

systems are not always the ideal candidates for sewage

sludge disposal.

Forests, in contrast, are generally considered stable,
nutrient-conservative ecosystems with greater potential
to act as living filters of mineral enriched discharges
(Sopper 1975, WOodwell 1977). Forests are attractive as
potential sewage disposal sites because of their eco-
system.longevity, stability, and their tendency to be
nitrogen-limited (Wollum.and Davey 1975, Bormann and
Likens 1979). Sewage effluent or sludge applied to aggra-
ding northern hardwood forests has been shown to achieve
significant nutrient removal, primarily by the ecosystem's
plant communities (Woodwell 1977, Sopper and Kardos 1979).
Sludge fertilization can result in increased forest pro-
ductivity, particularly by trees (Saffort 1973, Weetman
and Hill 1973, Woodwell 1977, Sopper and Kardos 1979).
However, fertilizer effects on forest plant and animal
community composition in terms of numbers of individuals,
dominance, and structure can alter the integrity and
stability of the ecosystem.(Weetman and Hill 1973), thus
pointing to the need for ecosystem study and monitoring.

Past studies of forest fertilization effects have
concentrated on tree production, and usually have dealt
with tree plantations and coniferous forests. In the de-
ciduous forests studied, responses of tree growth have
ranged from. negligible growth increases in 50-70 year
old stands (Koterba et a1. 1979, Sopper and Kardos 1979,
Stone et a1. 1982) to threefold growth increases compared

to controls in a younger stand (Sopper and Kardos 1979).

There is some evidence that sludge or effluent fertili-
zation initiates plant community composition and structure
changes, with a general acceleration in canopy closure,

an increase in leaf area index and in herbaceous cover,

a decrease in midstory cover, increased dominance by

shade tolerant understory species, and a decline in

plant species diversity (Anthony and weed 1979, Sopper

and Kardos 1979).

The sewage sludge application process for a mature
forest may involve clearing narrow (5m.wide) application
trails at intervals throughout the forest in order to
allow access by a spray vehicle.‘ This procedure would
produce a unique forest thinning effect. The influence
of this particular thinning pattern on forest communities
has not been investigated, either alone or in conjunction
with sludge fertilization. One would expect, however,
the invasion of post-disturbance plant species,(e.g.
brambles, cherry) a flourish of germination and growth by
species intermediate in shade tolerance, and stump sprouting
by species that characteristically use that form of re-
generation.

Very little work has been done on animal community
responses to forest sludge fertilization. Bierei et a1.

(1975) found that Peromyscus leucgpus population densities

 

on effluent and sludge-injected effluent treated areas

were significantly greater than control area densities

in the fall, but not in spring. The authors attributed this
to increased herbaceous growth following fertilization,
which improved summer food and cover resources that were
unavailable in winter. WOod and Simpson (1973) conducted
pilot studies on sewage effluent forest irrigation effects
on wildlife and wildlife habitat, and their data suggested
no difference in P. leucopus populations between treatment
and control sites. They noted greatly enhanced herbaceous
vegetation growth due to sewage treatment, theoretically
improving habitat for herbivores, which they did not
mmnitor. Sludge treatment in a 40-year-old Douglas fir
forest, according to early reports, resulted in lower
herbivore numbers on treated plots, apparently due to
reduction of the animals' required food and cover plant
species by sludge treatment (West 1981). Woodyard (1982)
monitored small mammal populations for 2 years following
sludge fertilization of a 4-year-old jack pine clearcut.
He found increases in small mammal species diversity and
foliage height diversity on treated plots, with 3 small
mammal species showing increased coloniZation on

treatment plots.

Although research has not examined mammal responses
to the forest thinning pattern created by cutting appli-
cation trails, more extensive thinning or clearcutting
in eastern deciduous forests has prompted herbivore in!

crease and immigration and changes in numbers within.

granivore-omnivore species (Krull 1970, Kirkland 1977).

A study of mammal use of 15-mrwide undisturbed winter roads
and surrounding forest in Canada revealed increased
Microtus captures in the predominantly grass and sedge

dominated roads, and fewer Clethrionomygggapperi captures

 

on roads than in forest (Douglass 1977).

The effects of forest fertilization and/or thinning
on reptile and amphibian population size and distribution
have not been documented.

Thus, the research to date leaves many questionstxzbe
answered concerning sludge fertilization effects on forest
ecosystems. Evidence does suggest that such a manipu—
lation can potentially alter forest plant and animal com-
munity composition, structure, and dominance and hence
the ecosystem's integrity and stability. Depending on
the goals of forest and wildlife managers, the effects may
be positive or negative, and in either case society would
benefit greatly by increasing its understanding of and
ability to predict the short and long range environmental

soundness of this form of sludge disposal.

OBJECTIVES

This study was undertaken with the primary objective
of determining the first year impact of sewage sludge
fertilization of a northern hardwoods ecosystem, on
various plant, small mammal, and terrestrial salamander
communities. Specifically, the responses of plant community
structure and composition, plant current annual pro-
duction, small mammal community size and composition, and
terrestrial salamander populations to sludge application

were investigated.

SITE DESCRIPTION

The area studied was part of the Thunder Bay River
State Forest in Montmorency County, northern lower
Michigan (Fig. 1). The site is an approximately 25 ha,
50-year old northern hardwoods forest located 16 km north-
east of Atlanta — roughly halfway between Gaylord and
Alpena. It occupies the SW % of the SW % of section 19,
and the NWkof the NW'k of section 30, T. 32 N.., R. 2 E.,
Montmorency Comty. The land is flat to gently sloping.
Soils were of the Mancelona, Melita and Menominee series,
which are deep, well-drained soils in sandy material (MSU
Forestry Dept. , unpubl.-~data. Elevation is approximately
300m above sea level. 7

Sugar maple (,Acer saccharum) and red maple (Acer

 

rubrum) were the dominant overstory tree species, with

beech (Fagus grandifolia), basswood (Tilia americana),

 

birch (Betula lutea and Betula papyrifera) , hemlock (Tsuga

 

canadensis) , red oak (Quercus rubra) and white ash (Fraxinus

 

 

 

americana) as subdominants.(MSU Forestry Dept. unpubl. data).

 

The study area was located just north of the Polar-
Equator Trail, which marks the 45th parallel. Atlanta,

Michigan (44° 59' N, 84° 10' W) and has a climate typical

Montmorency
County

 

Figure 1. Map showing location of study site in Montmorency
County, MI.

10

of northern lower Michigan, with long, severe winters,
short, cool summers and an abbreviated growing season.

The average annual precipitation (including melted snow)
is 76.66 cm. The mean annual temperature is 5.830 C.
Average temperature extremes range from -7.4°C in January
to l9.6°C in July (NOAA 1981) . Average precipitation
gradually declines from a monthly high of 8 cm in June to
a February low of 3.4 cm. During the study period (August,
1981 through August, 1982), temperatures closely followed
the average except during the winter, which was unusually
cold. Precipitation from fall to January tended to equal
or exceed the normal, but winter and spring levels were

abnormally low (Fig. 2) (NOAA 1982).

1
J-

inJ

 

 

 

 

53 Si"
3‘- “a
.3, ET
5 '3
5 2
<
a
ICIOI a
I
<
3
II.
N
"0qu . 'J a
II. a
2
IIIIIIIIIIIII 0
Cl)
8
< 2
a o o h a r v 'n N .-
8::=“-v«r?????
(we) uonoudiaud (0.) 01")!"de

Figure 2. Average precipitation and temperature, by month
for normal period and study period, 1981-1982,
Atlanta, MI.

METHODS AND MATERIALS

Experimental Design

The study was organized as a completely randomized de-
sign. The study area was divided into 9 study plots, each
1.5 ha in size and separated from each other by at least
20u1 buffer zones. In order for the sludge application
vehicle to gain access to the interiors of the study plots,
S-m wide application.trails were needed, running the
length of each plot and situated at 20o: intervals. Be-
cause these trails were expected to have a treatment effect
independent of the sludge effect, the study was designed
with 2 treatments and a control in orderto be able to
separate trail effects from response to sludge application.
The 9 study plots were randomly divided into 3 plots
with trails-only treatment, 3 plots with trails and
sludge, and 3 control plots which received no manipulation

at all (Fig. 3).

Trail and Sludge Treatment
In September 1981, application trails were created
on the 6 treatment plots, along with an east-west access

trail. All trees and shrubs greater than Zrn in height

12

l3

 

 

 

 

 

 

 

 

_ ...-IIIIIIIIQIQIIOIIIII

 

S

. T
..-..-...------------------------I

    
 

     

 

 

 

 

 

nfl

oflr

mmum

o- nu

3

M03

w..—
c

a...

.

I

'

'

'-

u

n»

 

 

J

rn hardwoods studv

h
N e
.... at
a r
0M“
m. nv“
.m .t
s t
. nm
s as
T. e
coy
mm
vie
am
.8
m mm
H. 1%
w BM”
n. v.,
Ade
ut
r71
cos
M .
fl “3
n
o e
.. u
n» .5
.1
F

l4

removed from.the trails by felling with chain saws and
skidding cut material to a site outside the study area.
Sludge application, scheduled for fall 1981, was post-
poned until summer 1982. In late June and early July
1982, each sludge treatment plot received approximately
14 metric tons (224,601 liters) of anaerobically digested,
municipal sewage sludge from.Rogers City, Michigan. Three
tanker trucks, making repeated trips, transported the
sludge to a field near the study area. Sludge was trans-
ferred from.these trucks to a smaller tank pulled by a
tractor originally designed for logging operations. The
tank sprayed sludge onto the adjacent forest strips,

designated as "interiors,' as it moved slowly along the
application trails. In order to achieve the desired
nitrogen loading level on the forest floor, the application
vehicle had to make several passes around each interior.
Application trails themselves did not receive sludge.
Application was accomplished in 2% weeks.

The U.S. Forest Service - MSU Cooperative Analytical
Laboratory analyzed sludge samples to determine element
content and loading levels, which are listed in Table 1.
Concentrations reflect the fact that this was "clean"
sludge, from a nonindustrial source, with very low levels
of trace elements - well below the maximum metal limits

allowable for sludges used on food crops (Chaney 1973).

,Heavy and trace metal loading levels were low, while

15

Table 1. Mean element concentrations in wet sludge from
Rogers City, MI, and mean loading levels applied
to the soil on the northern hardwood study site,
Montmorency County, MI in July 1982.

 

 

Chemical Loading levels

goncentration (X kg/ha for
Element (X for 3 plots) 3 plots)
Solids (%) 5.04 9210
Nitrogen (%) 0.427 783.1
Phosphorus (%) 0.21 383.7
Zn (ppm) 47.63 8.60
ca (ppm) ‘ ~ 0.42 0.08
Mn (ppm) 9.18 1.66
B (ppm) 1.50 0.27
Fe (ppm) 2568 465.9
A1 (ppm) 440.7 79.80
Mg (ppm) 275 49.84
Cu (ppm) 59.7 10.82
K (Ppm) 65.4 ' 11.89
Ca (ppm) 2781 503.0
Ni (ppm) 1.17 0.21
Cr (ppm) 3.22 0.58

Na (ppm) 102. 4 18.57

 

16

the application rates of nitrogen, phosphorus, calcium
and potassium.were high and are comparable to high conven-
tional fertilization treatments (MSU Forestry Department

unpubl. data).

Vegetative Community Composition

Vegetation sampling used a stratified random sampling
design since plant populations were expected to vary
in a predictable pattern among the application trails
and the interior forest strips between trails. Trails
were sampled as l stratum and interiors as another, in
order to keep variation within strata small and avoid in-
flating the sampling error of the estimated population
mean (Steel and Torrie 1960). In addition, separate
stratum means could be estimated and compared to each
other (e.g. sludged interiors vs. unsludged interiors vs.
controls).

Plant community composition was characterized using
nested quadrats, placed according to the above-mentioned
stratified random sampling design. Woody plants were
counted and recorded, by species.ixi each of 5 size
classes: 0-30 cm tall, 30 cm - 1m tall, 1m- 2m tall,
>2m tall'but <10 cm dbh, and >2m tall and >10 cm dbh.

The height classes chosen conformed to the natural
life forms which occurred on the site, and to correspond

with the vertical cover heights also measured - heights

17

considered to be of particular importance to the small
mammals océupying the habitat (M'Closkey and Lajoie 1975).
Quadrats (plots) used to sample the smallest size class
(0-30 cm) were In X 10111, class II (30 cm - 1m) were

2m X 20111 in interiors and controls and 1m X 30111 in

the trails, classes III (1m - 2m) and IV (> 2m high, <10 cm
dbh) were also 2 m X 20 m, and class V (>2 m high, < 10 cm
dbh) were4 m X 20 111. Long, narrow rectangular plots were
found to be most effective, since much of the vegetation
occurred in a clumped distribution. Quadrats were randomly
located in trails, interiors, and controls. Trail quadrats
were placed lengthwise along the trails, while interior
quadrats angled across the interiors in order to fit

2001 plots into a 15 mrwide interior. Frequency of
herbaceous vegetation was recorded, by species, using

2m X 5m quadrats.‘
Vertical Vegetative Cover

The line intercept method (Gysel and Lyon 1980) was
used to estimate vertical cover and foliage height diversity.
All vegetative cover 1 cm or greater intercepting an
imaginary vertical plane rising from 1 edge of a 20 m-long
tape was recorded. Gaps of less than 10 cm were ignored.
Cover was measured for each of 4 height strata known to be
important to small mammals: 0-10 cm, 10-30 cm, 30 cm - 2pm

and >21n (M'Closkey and Lajoie 1975). Cover lines were

18

randomly located in control and treatment plots, and in
treatment plots they were placed so that they ran across
a trail and an adjacent interior. Cover across the

15u1 wide interiors was recorded in Sui segments, so

that an edge profile could be constructed. Segments ad-
jacent to a trail were expected to show an edge effect.
Vegetation measures were taken during the last week in
July, 1983, after full leaf-out but before senescence had

led to significant leaf loss.

Annual Plant Production

To estimate primary production below 2 m in height.
samples of above-ground vegetative growth to up to 2 m
were collected. Quadrats 35m x 15m were randomly located
in control plots. In treatment plots, 55m x 20m plots
were randomly located across plots so that they spanned an
entire interior (15 m) and an adjacent trail (5 m). Within
a quadrat all living herbaceous vegetation was clipped, as
was all current annual growth from living woody plants,
from.ground level to a height of 2 mm Tissue taken from
interior strips was kept separate from.that collected in
trails. Seven categories of vegetation were chosen, based
on consistent abundance throughout all study plots: Hop-
hornbeam (Ostrya virginiana), beech, white ash, sugar
maple, and bracken fern were segregated into individual

groups. All other woody species were combined (e.g. red

19

oak, red maple,.basswood) into a single group, and all
herbaceous Species were combined together. Plant production
above 2 m.in height was not collected, since it was assumed
that such material would be beyond the reach of foraging
herbivorous mammals expected to inhabit the area. Vege-
tation was collected during the last week in August, 1982.
Samples were placed in paper bags, oven-dried at 60°C until
they no longer lost weight (@ 24 hr), at which time dry '

'weights were recorded.

Small Mammal Populations

Small mammal populations were monitored by conducting
several periods of live-trapping, each lasting for 5 con-
secutive days and occurring once per month.

In mid-August 1981, a single 5-day trapping session
was conducted in an attempt to gather pre-treatment, base-
line data on the study area's small mammal community. A
5 x 5 grid of trap stations was centered within each study
plot. Stations were 15 m apart and had a single Sherman
live—trap (H. B. Sherman Co., Tallahassee, FL) (13 x 13 x
38 can. Bait consisted of rolled oats, raisins, and anise
extract. Traps were left open throughout the 5-day period,
and were checked and re-set each morning. All newly cap-
tured animals were marked with ear tags or toe clipping;
and for each capturedindividual ID, species, sex, relative

age and trap station were recorded.

20

Trapping resumed in May, 1982, after trail cutting
but before sludge application. In order to accommodate
the high mammal populations indicated by the previous
August's results, and also to standardize methods with
concurrent trapping on 3 other sites as part of a larger
project, trapping methods were revised. In 1982, 6 x 7
trapping grids were used, with trap stations 10 m apart and
2 Sherman traps per station. One half of the traps were
placed in trails and 35 located in interiors, in order
to check for differential use of the habitat. Each trap
was baited with whole oats, animal fat and anise extract
and equipped with cotton nest material. The June 1982
trapping period was abandoned due to sludge application,
but trapping resumed in July, 2 days after sludging was
completed, and a final period was conducted in mid-August.

Terrestrial salamanders were censused along belt
transects using the method described by Burton and Likens
(1975). On rainy summer nights after litter and under-
story plants are well soaked, terrestrial salamanders such
as the red-backed salamander (Plethodon cinereus) can be
found foraging above ground on litter, plant leaves or
stems up to 2.8 m above ground level. Randomly located
belt transects, 2 m x 95 m, each spanning a plot's in-
teriors and trails, were marked out in advance with string.
0n appropriate rainy nights, beginning 1 hr after sunset,

teams of observers using bright flashlights slowly walked

21

transects recording each salamander observed in the
transect. 'Each team of observers worked a pair of tran-
sects: l transect in a trail-only plot and 1 in a sludge
and trails plot. An equal number of transect pairs began
with observation of a sludge plot and a trail-only plot,

so as to minimize any effect from time or observer fatigue.

Data Analysis

Since the study was designed around 3 types of
treatment plots (the 2 manipulations: trails-only, and
trails and sludge application, and the controls), l-way
analysis of variance was used to compare vegetation data
and identify significant differences among treatments
(Steel and Torrie 1960). In the seedling size classes,
variances often differed enough to be classified as hetero-
geneous when examined with Bartlett's test for homo-
geneity of variance. Heterogeneous data were subjected
to a log transformation, which often resulted in homo-
scedasticity, thus making analysis of variance the appro-
priate test for significance. In the few instances when
transformation did not correct heterogeneity, pairs of
treatments were compared using the t-test for unequal
variances (Steel and Torrie 1960).

Percent cover data, not expected to follow a normal
distribution, were subjected to the arcsinLNXer' trans-

formation (Steel and Torrie 1960).

22

The stratified random sampling procedure for vegetation
yielded estimates of means for entire plots, in which data
from the 2 strata (trails and interiors) were combined.

In addition, estimates from interiors or trails alone

were available, and were compared to each other and to
controls, again using l-way analysis of variance. Finally,
trail data were compared to interior data, within treat-
ments, using t-tests.

Required sample sizes for all vegetation measures were
calculated using Snedecor's (1956) formula:

S2t2

nS—T
d

in which n = required number of plots or lines
32 a sample variance
t = normal deviate at confidence limit level
(a=-0.10) and appropriate degrees of
freedom

d = margin of error (sample X times designated

accuracy of 20%)

The number of individual mammals captured in August,
1981 was relatively high, while populations were greatly
reduced in 1982. Since 1982 numbers were too low to make
the use of conventional capture-recapture population esti-
mators feasible, enumeration was chosen to estimate all
mammal populations. Beginning at time t on a plot, the I

number of animals caught at time t was summed with the

23

numbers of previously marked animals caught after time t,
but not at time t (Krebs 1966). Since treatment plot sizes
were small, the study was concerned with comparisons of
relative numbers of animals rather than with population
densities, and density estimates were unnecessary. One-
way analysis of variance compared mammal numbers among
treatments on a month-by-mpnth basis. T-tests compared

the number of trail captures to interior captures.

Salamander populations were estimated by direct count,
‘and the data for pairs of transects counted 0n the same
night were evaluated using a paired-comparison t-test.

A chi-squared test compared numbers seen on trails to
numbers seen in interiors.

Foliage height diversity and mammal species diversity
were each estimated with the Shannon-Wiener equation:

H1 = -Zpi£npi (Brower and Zar 1977), where pi is the pro-
portion of the total (vertical cover or mammal species)
which the ith category contributes. One-way analysis of
variance evaluated treatment differences.

Linear correlations were used to test for associations
between mammal species diversity and foliage height di-
versity, and associations between small mammal numbers and
various cover estimates and FHD. In conjunction with analyses
0f variance, specific treatment differences were isolated
using Duncan's new multiple range test (Chew 1976). Dif-
ferences were considered significant at the c1= 0.10 level,

:for all comparisons.

RESULTS

Vegetative Community Composition

When trail and interior data were combined to re-
present an entire plot, neither sludge and trail nor
trail-only treatment had a discernible effect on total
living woody stem densities in 4 of the 5 size classes.

In the large tree size class, >2 m in height and >10 cm
dbh, however, control plots had stem.densities signifi-
cantly greater than plots with trails or plots with both
sludge and trail treatment (P<:0.05) (Table 2).

The density data were also broken down into interior-
only and trail-only comparisons, by species, between the
2 treatments and the controls. These comparisons revealed
several treatment responses which varied according to
species and size class. The general trend was for sludged
interiors and sludge plot trails to have lower stem
densities relative to those in unsludged interiors, controls,
or trails in trail-only plots.

The small seedling class (0-30 cm) followed this

trend. When interior strips were compared between treatments

24

25

.Amo.ou.mv quHmMMHm hauamo

uHMchHm mum mumHuomuomsm umuuoa ucmHmMMHm suH3 30H HmumouHuos mama onu SH madman:

.ucmsummnu Hum muoam m Scum .uouum mummomum nuHs osHm> x

N

 

 

3 Haas Hm Hnfio 3 Huang is so 3. .5.
a? H $~.~ 2.4 H oom.~ . «3 H 2: £6 so By .s N.
a: H 2m; m2 H m2; . can H on: am - a a
3N H mafia 2... H Sam.” .5 H at: 5 .5 on

SW; H 39? 2:; H $0.6m 8%: H 02.2 .8 3-0
muoam mHHmHu muoaa mHHmHH\3 muoam Houucou mmmao oNHm

mam owmaam H

 

.Nme .mamh .Hz .humsoo mucouoaucoz cH SOHumoHHmmm mmmsfim
mam mHHmuu £33 muoam man .9333 sung muon .maouucoo you Hammad mNHm Hum
AmHHmnu mam mHCHHouaHV muoam oHHumo CH mucmam huooz Ham mo Amn\mEmumv kuHmcmn .N manna

26

(Table 3), sludge-treated interiors had fewer total seedlings,
and fewer "combined maple"*, red oak, and "combined other
species" seedlings than did controls or interiors in
trail-only plots. However, the sludge plot interiordensities
«mftotal seedlings and combined maple seedlings were not
significantly lower than control densities, nor were

sludged interior numbers significantly lower than trail-

only plot densities of "combined other" species. White

ash seedling densities proved the consistent exception to

the trend, and in this smallest size class, interiors in

both types of treatment plots had significantly more ash
seedlings than were found in controls.

Sludge was not applied to the trails themselves. In
trail density‘comparisons between treatments, in the 0-30<xn
size class (Table 4) while ash had significantly higher
densities on trails in both treatment types than on con;
trol plots (P < 0.01). As in the interiors, red oak
had fewer seedlings on sludge-plot trails than on other
trails or controls, although “the difference was
Significant only between control plots and sludge plot
trails.

Among larger seedlings (30 cmn~1 m) in interior
Strips, only white ash and total seedling numbers differed

SUgar and red maple data were combined in this size class,
due to difficulty in their identification.

27

.mmoamHnm> Hummus: you

umousu :uH3.uoummu mops .u.m.s .soHumaHommcmHu an umuomuuoocs mommHum> mooomowououonsk
Stay m on “Sammie 328323.13

.mmHMHommm mmHsuonuo mmmHss no.ov.m um mamuommau hausmo

uHchme one mumHHomummsm umuuoa unoHoMMHm suH3.3ou HmuaouHHon 08mm emu :H madnessm

.uaoaumouu Hum uuoam m scum .Houuo mummmmum nuHs osam> Ma

 

 

 

 

 

 

 

 

 

 

292 H5393 _ ~3.2H..ao3.oa 8Q2Hfi2~.2 H38.

_ aw Hpmme Sm HRSHJ «2; Haaomé 3303 mafia emaansoo

2 Hanan can Hana: H3 Haenm; $28 320:3

mac cam

mafia H ON; o? H mafia SN H «S; A8823.“ 23.:

mooBmmmn mmoHuma<

. .. . . . .. 2.: in H gag Aaaaopaam 2:50.33

mg H +n$~ m as m lffim a :3. 823

S H on“ a: H m2; Sm; H 3.; 33033me 23$

nommn :MOHH08<

ante H.392 or; Haida... on: Hfimosdm @3233 .H a 938

N Hmo<v nuance mmmHQEoo

mmH.H o~m.a oom.H mmo.~ «mm.H ~o~.m AmGMHckuH>.mwuumov

ammncnommom

wuoag mHHmuu muoaa mHHmHH\3 muoam Houucoo mmHooam
use owuoam H

 

.NmmH mama .Hz .auqsou mucouoaumoz mH
.mOHumOHHmmw owusam mom mHHuHu nuHs muoam mam .mHHmHu suHs muon .mHouumoo How
"mHOHHmuGH SH mmmHo mNHm EU onto mnu :H mmHommm some mo Ama\m8mumv hufimsmn. .Avagams

28

unmauwouu Mom muOHm m Baum .Houum mummmmum nqu omHm> x

oH.ov.m .ucmHoMMHm hHuamoHMchHmk

.ponHoomm omHshmnuo mmoHas Ho.ou.m um uSmHmMMHm aHusmo
:HmHmem mum mumHHomHmmmm HouumH uanmMMHm nuHB 30H HmumouHHos 08mm onu mH muonasmu

 

 

 

 

 

 

 

 

 

 

..H
mm: H «3...; RNJ H 93.3 03.: H 02.? ~33.
m2 H 2N; m3; H a2: N2; H momé 332; 3:3 BEES
in H mg a? .1902me HS Hump: $23 3338
sac spa

Sm; H 2: SN H as; new H NS; Aaaaopaam 3:5
moo3mmmn :moHHma<

m2 Hnomma Sim Hammad «RH Rpm A2823... 35x33
Na «awn mamas

«S H 3...; SN H 8m; 3m; H SRN Qfiofleafiw 2&3:
Scoop amoHuma<

«35 H 8a.: 3mg H stag an: H 3.3m 33233 .H a 358
Hoo<v mmHmma_mocHnaoo

com H 2;; NS H 3.: «mm H ~85 3:353? H.528
Smoochonmom

quHa mHHdHu mHOHm mHHmww\3 muon Houucoo mmHommm

was owmsHm

H

 

.NmmH mHsh .Hz .mussou mocmuoauaoz cH

.sOHuuoHHmmm owmaHm use mHHmHu nuH3 muon mam .mHHuHu nqu muon .mHOHucou
How "mHHmuu mH mmmHo 0NHm so onto onu sH moHoomm some mo Amn\mamumv h, , m
V/afiflfi a «H n a

29

oH.oo.m .ummumMMHm mHummoHMHmmHms

moHMHoomm omthonuo mmoHss .no.ou.m um usmuommHu mHusmo
uHmHamHm mum mumHHomHomom HouuoH uaouommHu squ sou HausouHHon 08mm on» :H muomascm

.ummaummuu Hum mHOHm m scum .Houuo mumusmum zuHs osHm> x

 

 

 

 

 

 

 

 

 

1H

RN +fioo~ m . e3 33.» a . «3+1.QO N H38.

3 H SH o3 H m3 . 2 H «3 mafia“; panda 3338

A H S a H 3 an H i 3.2.3 3338

xao sum

2 H 8 S H 2: 2 H 3. @5335 3:3

. moo3mmmn mmoHHoa<

2N Hfimmm 3m H.384 2: Had; Aacaoapmaa 35x38

N sum muHsz

mm H man 2 H 8m mm H is Aafioflsafiw 335

Scoop cmoHHoE<

SN H 26; Non H m2; 5 H 2: @3283 p33

mHmma_Hmw5m

3 H mm a H am 3 H .5 A528 “003

oHamE mom

8 H New EN H A: N2 H S: 355mg? «538

Emonmuosmom

muOHm mHHmHu muon mHHmnaxs HmHOHm Honumoo mmHoomm
mam mwmmHm

 

. .NmmH thh .Hz .mussoo monouoaumoz :H .:0Humo
uHHamo ommsHm mam «HHmHu nuHs muOHa use .mHHmuu squ muon .mHouucoo mom

"mHOHHoumH :H @930 oNHm E H :80 on 93 SH @9323 Sumo mo Amaqmamumv %//%%1V§ Home
.n m

30

significantly among treatments (Table 5). In both cases,
densities were significantly higher on interiors in trail-
only plots than on controls, with sludges interior densities
at nonsignificant intermediate levels.

Trail densities had nearly the reverse pattern in
large seedlings (Table 6). Control plots had more hophorn-
beam, beech, and "combined-other-species" seedlings than
did trails in either type of plot with trails. However,
in the case of hophornbeam and "combined other species,"
control densities were not significantly greater than
trail-only plot densities.

The final treatment influences on stem.densities
appeared in the interior strips in the 1m- 2m size
class (Table 7). White ash and American basswood both
had significantly higher (P<0.05) densities in trails-
only plot interiors than in either sludged interiors or
controls.

Trails did not contain trees or shrubs greater than
I'm tall. Pole size sapling (>»2u1 tall,:>10 cm dbh) and
large tree ( >21n tall, >10 cm.dbh) densities did not
differ significantly between control plots and treatment
Plot interiors (Tables 8 and 9).

When trail densities were compared to interior den-
Sities within the same treatment, a consistent pattern
emerged with interiors containing similar or higher

densities than found in trails. In the 0-30 cm size class,

31

mo.ov.m .ucmHmMMHm hHummoHMHmem k

.mmHMHommm omHsuomuo mmoHcs 0H.ouvm um uanmmeu

hHucmoHMchHm mum mumHHomHomnm HmuuoH HamHoMMHm nqu 30H HmucouHHon m SH mumnamcm
unmaumouu Hum mHOHm m Baum .HOHHm uHmmcmum nuH3 msHm> NH

 

 

 

 

 

 

 

 

 

.mNsH 2m; ommH RAJ in H 33 :38.
mm Hnmm mHHHnmme mH Hmqu moHomam Honuo mmmHnBoo
NH_H mq OH.H mH mm.H an Amundu mSUHmpdv

xmo mom

mu H_mn m_H oq NH H.Hn AmcmoHHmEm MHHHHV
moopmmmn mmoHHma<

o: H 9% 9.3 H 3: S: H Sm 350203 83x35
5mm manz

2 Ham: 3 Has: an H $3 328%:an Swat
k momma chHHmE<

NmH.H moo NwN.H cam nm_H own Aamumnoomm Hmu<v
. mHamB_Hmwmm

S H an 3 H 2 3 H .5 A533 p83
onma_uom

as HBSN HS H mom «2 H net 35333? $28
a N amonmuosmom

muOHm mHHmHu muOHm memHH\B muon Houucoo mmHoomm

mam mwmaHm

.NmmH mHsh .Hz..zu:ooo hoaouoaumox CH .moHumo
uHHmmm owusHm mam mHHuuu nqu muOHm use .mHHmHu £DH3_muon .mHouumoo How

"mHHmHu pH mmmHo mNHm a H..Eo o u a mo comm
on a a a .13... mo GEARS «fr/£6 a .013

32

mHucmoHMstHm ohm mumHHomHmmsm HmuumH uanoHMHu nuHa 30H HmucouHHon m aH madman:

umoEumoHu Hum mHOHm m_Bouw .Houuo mummamum nuHs osHm> x

.Amo.o..mv unopmmmae
N

 

 

 

 

 

 

 

 

 

HHN H Hmo.N HHH + Ham.~ .omm H saw.a Hmuoe
oH.H mm mm H mma mm H mm mmaumAm

Honuo umcHnaou

o.H o m_H o o.H o Amundh machomdv

xmo mom

m Ham m.Hn¢H m Hum AnamoHHmEm MHHHHV

moo3mmmn cmoHHoa<

mm HuNNH qu_Hmmm¢ mN H m¢o AnamoHHmSm mmameHmv

, N can muHsz

«NH H mas o H own owa H awn Amaaomaeaaaw mswamv

momma mmoHHoE<

SHH.H ewe mm H com Noam H mos Asapaauoam pmoav
mHmma Hmwmm

o.H o m H NH on H on Asapnnp pupae

. mHmmE mom
am_H son HNH H_mmo Ham H Has .Amaaasawpa> «Noumea

Emoncuonmom

.muOHa mHHmHu muoHa mHHmHH\B muon Houumoo moHommm

mam mwmnHm

 

.Nme mHmh .Hz..%ucboo monouoaucoz CH

.SOHHHOHHamw mwmmHm mam mHHmHu £HH3 muOHm mam .mHHmHu nqu muon .mHouucoo How

"mHOHumucH 5 mafia muHm EN 18H on“. 5 $3QO comm Ho $53833 EH23

.N mHan

. .mmocMHHMP Hmpvmfifi How umwuu nu £HH3 vmummu H553
mocmHmwme unmoHMchHm om .cOHumaHommcmHu hp monomuuoomp .mmocmHHm> msomcmwouoummN

.umoaumouu Hum mHOHm m Scum .Houum mumuamum :UHa msHm> x

 

33

 

 

 

 

 

 

 

 

Em H ~85 2: H n86 «8 H ~35 . :38.

m H m 3 H H3 an H 2 1.38%

Honuo ummHmaoo

m H m o H o o H o €58, 330:8

. . xmo mom

m H.HH m H_m m_H mN . AnamoHHoam MHHHHV

mooBmmmn amoHHmE<

0N H.N¢ «N.H 00H m.“ o AnamoHHoam mamemHmv

3mm manz

m: H 03 S H SH SH H «S Aaflomaeafiw Swab

momma amoHHma<

NNq H noN.N mHH_H omm.H MNN H omo.H . Aapnmnoomm Hoo<v

onma Hawsw

m_H m H_H mN mm H wON . Aabnnnu Hoomv

NmHmmE mom

co “.mON mNH H_Hoq NNH H NNN AmmchwaH> mwmumov

amonmuonaom

quHa mHHmHu . muon,mHHmHH\3 muOHm Houuzoo mmHommm
mom mwmmHm . H .

 

.Nme hHsh .Hz Laumpoo moamhoaucoz :H .GOHumoHHmmm owmsHm
can mHHmHu squ mHOHm was mHHmHu nuHB muOHm .mHopuaoo HomumHOHHoucH cH
wwwHo oNHm saw so OHv .HHmu 8 NA onu :H mmHoomm some mo Aw:\mamumv HuHmmma .m mHAmH

34

.ucmaummuu Hum quHm m Eoum .HOHHo mumuamum nuHB msHm> x

 

 

 

 

 

 

 

 

 

H

mm_H com He_H Hoa om_H mam HmuoH
on H N¢ oq H HOH mN.H oq mmHommm

_ Honuo moaHQEoo

mm H.m~ m_H Hm e_H mm Ampasp msopmncv

xao use

Hm.H mHH OH H HOH mH_H OHH HacaoHpmsm mHHHev

mooBmmmn amOHHmE<

«H H mH H.H eH q H.0H AmcmoHHmEm mamemHmv

. Ham moHaz

m~.H so HH H_Ho Hm.H am HmHHoHHeaapw assume

momma mmoHHmE<

mH.H Ham NH H.Hnm He H mom Assumaooam “moav

mHmmE uanm

H H H o H aH as H «a Haspnap Hmoav

mHmmE mom

H H mm N_H HH N.H a AaemHeHwHH> ammumov

amonmnonmom

.muon mHHmHu muOHm mHHoMH\3 muOHm Houuaou moHoomm

mam mammHm H

 

.Nme aHDH .Hz .hucpou hocmHoEucoz CH .GOHumOHHmmm ommmHm
mam mHHmHu £HH3 muon mam .mHHmHu SHH3 muOHm .mHOHuaoo How mHOHHouaH GH
mmmHo oNHm nan Bo OHA .Hku 8 NA mnu aH moHommm come we Aw:\m8mumv zuHmama .o anmH

35

red Oak, combined species of maple, and total seedling
densities were significantly higher in trail-only plot
interiors than in trails in the same plots (P<:0.10); no
other significant differences occurred (Tables 10 and 11).
In the large seedling size class (30 cm- 1m) the pattern
was not so well defined, although interiors still had
similar to higher densities. Here, hophornbeam and total
seedling densities in sludged interiors exceeded those

in trails in the same plots (Table 11). Trail-only plot
interiors had significantly more beech, basswood and total

seedlings than were found in the trails (Table 10).

Herbaceous Species Frequency

Herbaceous understory species tended to occur in.a
varied distribution from plot to plot, making it difficult
to characterize the forest floor's herbaceous composition
in other than broad terms. However, some patterns emerged
(Fig. 4). .

There were 7 plants common to nearly all 9 of the
study plots: starflower (Trentalis borealis), Canada
mayflower (Maianthemum canadense), sedge (Céggg spp.),
bedstraw (Galium aparine), violet (Viglg spp.), heart-

leaved aster (Aster cordifolius), and brambles (Rubus spp.)

 

Their distributions closely resembled each other in ..con-
trol. plots and traileonly plot interiors (Fig. 4). These
same herbaceous species had a somewhat different occurrence

profile in the trails. Trails had similar profiles in

36

Amo.ouvmv huHmcom HHmuu swan Hoummuw hHuamonchHmsk
AcH.ouvmv zuHmcou HHmHu mmnu Houmoum HHummonchHm«
muOHm unmaumouu m o£u Scum .Houuo unmmcmum nqu osHm> NH

 

 

 

 

 

 

 

 

 

 

 

 

ommHHHH.~ oHnHaosa.e om~.NHaam.ms .HHn.mHHH«ooa.oa Hades

mHH+mmH omHHama am¢.HHomm.~ meHHHm.~ mmHomam
1 1 1 1 “mamm emaHasou
OH+mH a+aH Hms+omH.H emm+ mmm.~ Hannah msopmnov
1 1 1 m xmo sum
m+ws wo+ SOH Hw~+moo.H oHH+meH.H HacaoHposm mHHHev
I «H I I moosmmmn amomuosm
mm~+smH mmm+omo.H oeH.m+o~m.mH oeo.m+-~.HH HaaaoHHmsa macHxapmv
1 1 1 1 saw muHaz
oq+oeH ~H+ cam HH~+mom.H mmH+mmH.H HmHHoHHecapm mammmv
kt noomn cMUHHmE<
11 11 can.mHoHH.mH oHs.aH*pHHH.He mmHaaa emaHaaoo
NwNqum NcmHnMH.H II II Hasnmnoomm Hoo<v
I I II II «Hana Hmwmm
wH+mH m+m~ Asapnsp Hmo<v
1 1 1 1 39...s wax
HaH+Hom aH~+HHH HmH+mms.H cam+mmo.~ HmamHanHH> ammumov
Emmnmnonmom

mHHmHH mHOHHoucH mHHmHB HmHOHHmucH

E H 1 So on So om1o moHoomm

 

.mmaH HHaa .Hz

.hussoo monouoaucoz .HH:01mHHmuu nuH3 muon How .mHHmHu mam mHOHHmunH cH
.mommmHo ouHm 8 H160 on was Eu OM10 emu :H moHoomm sumo mo Am£\mamumv >uHmamn .oH MHQHH

37

Amo.ouvmv auHmcom HHMHH cmnu Houmoum hHucmonchHm¥¥
216 v5 .3386 :85 can”. 830% >353chme a
muOHm ummEummHu m emu Scum .HOHHo mummcmum nuH3 osHm> x

 

 

 

 

 

 

 

 

 

 

 

 

IH
aHsHmHm.H HHNH«OOH.H m~m.~HmsH.sm mme.oHHoee.ss Hades
HHHHH asHHHH HHHHHH~.H aNHmme mmHomam

I . I I I Hmauo mmmHnfiou
NH+mq s+aH «NH+mo¢ mm+mom Ampoou mounmoov
I I I I xmo mom
mN+mN MH+oo nem.H+cHH.m mNN.N+mHN.¢ AnamoHHoEm mHHHBV
I I I I woozmmmn :moHHoE<
oNH+aqm mNN+mmw mmm+¢mm.m mmo.H+moN.m AnamoHHmEm msameHmv
I I I I .mmm manz
~m+mHH -+ amm Nam+mmn.H Hm+emH HaHHoHHeaan mswmmv
* I I noose :moHHmE<
1| I 30633.3 35:30.? 339: 3:358
NmHHmoo OHNHomoH l1 1| Assumzoomm .803
I. I oHamE Hmwsm
OH+Nm c¢+mm II II Asounsu Hoo<v
1 1 1 1 0Hams wmm
m¢+ooN om+ssN¢o oom+mqo.H mmH+on.H AmcmHanuH> mmuumov
amoncuonmom

mHHmHH mHOHHoucH mHHmHH HmHOHHousH
a H 1 so on :8 07o moHomam

 

.NmmH NAH3. .HE .Humsou Hocmuoaucoz mH

.HOHumoHHmmw owmsHm mam mHHmHu :uH3 muOHm How .mHHmHu mam mHOHHmumH mH

.mmmmmHo ouHm EH 150 on mam So onio 93 SH mmHommm some mo Am£\m50umv .SHmama

.HH aHHHH

 

100

50
U)
l'?
m
H
H'I
p—A
O
s
ID
'1

100

59
U!
H
93
H
F?!
...:
O
s
m
'1

Figure 4.

U.)
(..:

INTERIORS sludge & trans I'll-

 

EC) ACO on < >
ID no It I". ‘D
‘45 mo. o. o n
H19) H00 CD F" I’D
HQ- mm H (D '1
OD) x H r?
S m
m g 2
H v
v
TRAILS
i
I
l
I
50 Am a: < >
9393 CID I‘D H- In
<5 DID- D.- O f"?
H593 I100 m I--' G)
I-‘D- mm H I‘D '1
29: >4 3 n
m g s
"‘ 1?
V

(°dds annu)

(-dds annu)

w/trails ——

 

control _

 

3 :<
5 a
v H
H m
m H
m :1
m

H E
3 I:
c m
H H
m o
(n

Frequency of commonly occurring herbs and shrubs
(Z of quadrats in which the plant occurred) in
control plots, plots with trails, and plots
with trails and sludge application, Montmorency

County, MI, July 1982 (X with SE).

39

both types of plots with trails. A third, intermediate
pattern emerged from herbaceous plant occurrence in sludged
interiors.

As. the profiles indicate, 9513935, violet, bedstraw and
heart-leaved aster occurred with similar frequency in-
all situations. Starflower and Canada mayflower were the _
most common plants in controls and trail-only plot interiors,
and were among the least frequent in the trails and in
sludged interiors. §_11_b_u§_ had a relatively low occurrence
in controls and trail-only plot interiors, was relatively
more common in sludged interiors, and became the most
frequently found herbaceous plant in the trails. An ad-
ditional species, mullein (Verbascum spp.), appeared
commonly in the trails and did not. occur elsewhere.

A list of all plant species identified on the study

area indicates the tremendous variety present (Appendix 1).

Vertical Vegetative Cover

When plots were considered in their entirety by com-
bining interior and trail percent cover data, percent
vertical cover did not differ significantly between controls
a.l‘ld/or treatments in stratum 1, the 0-10 cm height class
(Fig. 5) . In the 10-30 cm stratum (stratum 2), sludged
Plots had significantly less vertical cover than did
trail-only plots. Sludged plots and trail-only plots both
had less cover than did control plots in the 30 cm-'2m

and the >2 m height classes (strata 3 and 4). This reduction

4:>2m

 

30cm-2m

...
s

3:

311mm
2: I0-30.cm

w/tralls IIIIIIIIIIIII

1: O-IO cm

control-

 

Figure 5. Percent vertical vegetative cover in 4 height
strata in entire plots: for control plots,
plots with trails, and plots with trails and
sludge application, Montmorency County, MI,
July 1982 (X with SE).

1'Bars in the same stratum with different letter superscripts
are significantly different (P < 0.05).

*Significantly different, P < 0.10.

41

in cover was not statistically different from control plot
cover, however, in trail-only plots for stratum 3 (Fig.
5).

Considering the data for interior and trail cover
individually revealed a breakdown of the treatment effects.
When only interior cover was compared (Fig. 6), sludged
.interiors had significantly less vertical cover (P<:0.05)
in the 10-30 cm stratum.than did trail-only plot interiors.
This parallels the lower seedling densities in the same
height range. In stratum.4, control plot and sludged
interiors had significantly more cover than trail-only
plot interiors (Fig. 6). However, when interior cover was
separated into 2, 5-m edge segments and a S-m.inner-
interior segment, the cover differed significantly between
treatment interiors and controls only in the edge segments,
adjacent to the trails.

Trail cover comparisons (Fig. 7) indicated the most
pronounced treatment differences in cover. As with
entire plot and interior-only data, trail cover in stratum
1 (0-10 cm) was highly variable and no significant differences
between treatments emerged. In stratum 2 (10-30 cm),
sludge-plot trails had significantly less cover than did
control plots. Strata 3 and 4, together representing all
cover above 30 cm in height, had significantly less (P< 0.01)
vertical cover in trails of both treatment types than in

control plots.

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S 100l‘ ...-I8
:5 575’ sass
25-
. E
‘2' 8.
g 15* é E
5'3 10. E
s 5' E 55.
S E 3 E E
E 530 §
§ 駰- ;
' “g4. é:
ab E
2. g b
9' E
c': 5'
"- ,. g E
.. s

 

Figure 6. Percent vertical vegetative cover in 4 height
strata in forest interiors: for control plots,
plots with trails, and plots with trails and
sludge application, Montmorency County, MI,
July 1982 (X with SE)

1Bars in the same stratum.with different letter superscripts
are significantly different (P<:0.05).

 

 

 

 

 

 

 

 

43
100- +3
E
N
A 75.
if
50!-
25 =3
E h
? I!
o
5 E
g 15- I“
6':
10-
E z-—:
05' .'=_'.
0 b "='
s El- 5 ML. —
0:2 :g
OH! to
€786 a
o I
to -E 3
N 4.
1 a
is
2' .—
5 Eb
E — U
0
o ‘
T e- g
° 5
4'. 4_ ‘, °
2-

 

 

 

 

Figure 7. Percent vertical vegetative cover in 4 height
strata in application trails: for control plots,
plots with trails, and plots with trails and
sludge application, Montmorency County, MI,

July 1982 (X with SE).

1Bars in the same stratum with different letter superscripts

are significantly different at P< 0.01 unless otherwise
specified.

*significantly different P< 0.10.

44

Foliage Height Diversity

Foliage height diversity (FHD) was significantly lower
in sludged plots than in trail-only or control plots,
both when entire plots were compared and when interiors
alone were compared (Table 12). Control plots had sig-
nificantly greater FHD than did FHD for trails in sludged

plots.

Plant Annual Production

For most species the current season's primary pro-
duction below 2 m did not differ between treatment and/or
control plots. When differences did occur, the trend
was for greater production on trail-only plots than on
controls, with sludge plot production at intermediate
levels.

When entire plot data were compared (Table 13), trail-
only plots had nearly 3 times as much sugar maple current
annual production as did control plots. Total annual
production on trail-only plots exceeded that on control
plots by 261%.

Interior-only comparisons yielded only 1 treatment
difference: sugar maple annual production was twice
as great on trail-only interiors as on controls (Table 14).
Trails in trail-only plots had 4 times the amount of
sugar maple prodpction and 2.7 times as much total

annual ‘vegetative production as did control plots

45

oa.ouvm .ucmummwan spasmoflmaawams

.powwwommm oma3honuo mamas: mo.o m ucmummmwm
hauamuamwcmam mum muawuomuomsm umuuoa ucouwmmwm £ua3.3ou Hmuaoufiuon m CH mumnabn

 

 

N

udofiumouu Hon muoam m Eonm .Houum pudendum nua3 mdam> Ma

oqmc.o.Hn¢m¢N.o «oHH.o.Hnmunoq.o moao.o_HsmmmNm.o Sago mHHmHH

mNoo.o Hnmmmd.o memo.o_HoomNm.o mmao.o H mmmmm.o Saao muowumucH

amao.o.un¢ewm.o mamo.o Hda¢m¢.o mmso.o_wmmn-m.c muosa usausm

muon mafiwuu muoam maamHBx3 muon Houucoo macawummaoo
mam ompsam H

 

.mwma hash .Hz .xuqsou hocmuosucoz

pH .cofiumoaamao owndam mam mawmuu £uw3 muoam mam .mawmuu £uw3.muoam
.muoam Houuaoo How AmoSHm> NopSH Hmcowz cossmsmv hufimnm>wm uswwmn owmwaom .NH mafima

46

.Amo.ov.mv econommap
mauaoowmaawfim ohm muawuomuomnm Houuoa uooummmwp nuw3 30H Hmuaouwuon m CH madness

 

 

 

 

 

 

 

N

.ucoEummHu Hon muoam m Boum .Houuo pumpnmum nuHS osam> Ma
«H.m +noom.~¢ mn.m~ +noo.me mm.H +doo.o~ Hmuoy
85 H and S.~ u. :3 on; H 24 2533:? 5.323%
show cmxomum

mm.~H 8.: $4..“ 86 stews; $3.2;
hpoos.uonuo

3a Hammofi m: Hawtom 5.0 Hugs A5283 ~83
. magma Hmwom

84 H 35 .35 H 25 85 ...r 3.0 3538.8 against
saw moans

24 H 3g 3s H SS .35 H and Aafioflecmmw Swab
scoop cmownma<
85 H 86 :4 H o: a: H £5 A2353? assume
amonouonmom
mo.~ H om.o -.¢ H.mo.~a Nm.o.H m~.N moaomam msooomnum:
wuoam owpoam muon mHHoHH\3 muoaa Houucoo mowoomm
mam mHHmHH a .

 

.Nme umamn< .Hs
.huaooo hocmnoaucoz ca coaumofiaamm mwpaam mam magnum sues muoaa pom .maamuu
:33 $on .3938”. you usmv 3535 53260.3 H358 um: 2:53 «>22 .2 383.

47

.AoH.ov.mv uaoHoMMHm

haunmonHame mum muaHHomHomom Houuoa ucoummep SHH3 30H HmHGONHHos o :H mumnas:

 

 

 

 

 

 

 

 

N
.uamaumonu Hon muoam m Scum .Houuo pudendum :HH3.o:Hm> xH
35 H 8.: ~05 H 3.3 a: H 2.2 H38
85 H $5 2: H a: 84 H 24 Engine... 53333
show coxomum

an; H Ed 24 H 2.0 84 H 36 $32;
SUOOB Hmnuo

mm.o.Hnomm.MH mw.m.Hnmn.oH mm.o H mmm.m Assumsoowm Hoo<v
N magma Howdm

Na.o.H m~.m -.m H mm.n mH.o H.Ho.a AnamoHHmam chmeHmv
3mm oanz

34 H 35 SH H 8H 35 H SS QEoEecfim $me
50003 cmoHHmE<

No.o H_oo.o mm.H H.mm.o mn.a H ~H.¢ AmachHwHH> oNHumov
amonchonaom

~¢.H H n¢.¢ q~.m_H om.¢a mm.a H mq.m .mom moooomnuom
muoHQ mHHmHu muOHQ mHHmHH\B muoam HOHufiOU moaommm

mam mwpsam H
I .NmmH umsws¢ .Hz .kuasoo zocmuoaucoz
. aH COHuoOHHaam owppam mam mHHoHu nuHa muoaa pom .mHHouu nuHB muoan

maouucoo HomanoHHmuaH Eoum :amv Ao£\wxv :OHuoavoum Honcao um: vasoum o>on< .qa manna

48

(Table 15). In contrast, beech annual production was sig-
nificantly greater on control plots than on trails in either

type of plot with trails.

Small Mammal Community

In August, 1981, a total of 4 small mammal species
was captured on the study area. Eastern chipmunks (Tamias
striatus), white-footed mice (Peromyscus leucopus) and
woodland deer mice (Peromyscus maniculatus gracilis) occurred
on all plots, and a single boreal red-backed vole
(Clethrionomys gapperi) was captured (Table 16). This pre-
treatment trapping period yielded significantly higher
Peromyscus (P<=0.01) and total (P‘<0.05) average per
plot numbers of animals known to be alive on controls than
were on either type of plot scheduled for treatment (Table
17).

In 1982, an additional species was captured on all,
3 types of plots: the woodland jumping mouse (Napaeozapus

insignis) (Table 18). There were incidental captures of

 

a masked shrew (Sorex cinereus) and a short-tailed shrew
(Blarina brevicauda), but no red-back voles were captured.
In contrast to the 1981 capture data, the 1982 average
per—plot numbers of individual animals known to be alive
did not differ significantly between treatments and/or
controls for any month, either pre-sludge (May) or post-
sludge application (July and August) (Table 19). When the

major species were compared separately, however, there

Aoa.ou.mv ucoHoMMHu
mauamoHMchHm mum mumHuomHmQSm Houuma ucmHoMMHn zuHs 30H HmHGONHHon o pH muonssaN

ucoaumouu you muoam m Scum .HOHHo pumpcmum nuHa osam> x

 

49

 

 

 

 

 

 

la
3.: Hoists.» 3.2 Hagan 34 Hui: H38.
:5 H :5 85 H 85 84 H 3A 953353 sauefiflmv
snow aoxomum

$5 H :5 .2: H 2; 34 H 85 83.2;
Hpooz Hmnuo

mag Hammmda $6 HnoHNm Nod Hmood 33233 80$
magma Hmwnm

mm.H H wo.m <m.o H wN.HH mo.o H Nm.o AdamoHHmEm mscmeHhv
5mm manz

and H 84 .2”; Hanna 3.0 H some Qfioflcafim .3me
n _ N nooon amoHHma<

:4 H 9m. 84 H 34 mm; H 2.4 A2333? matte
. EmonGHosaom

me.o H oo.NH HN.NH H qq.mN on.o H mq.N mmHoomm mscoomnuo:
muon mHHmHu muon HHmHH\3 muoflm Houucou mmHummm

mam owmnam H .

 

. .NmmH unswa< .Hz .Huaaou honouoaucoz
. :H coauonqum oupoam new mHHmHu squ muoaa cam .mHHwHu :uHs muo a
maouucoo HomumHHmuu Boum.s_Nv Am£\wxv GOHuoswoua Amanda um: wcsouw w>ow< .mH

manna

50

 

 

on He. mm
o H o
mm. mm an
S m m

Hmuoe

 

AHHmmmmw mNEOaOHunumHov
mHo> xomnvmu HmmHom

 

.amw mSUmHEoumm

 

AmdumHHum mmHEmHv
xcnamHno aumummm

 

muoam mHHmuu
wan mwvaam

muoam mHHmHH\3

muoam Houucoo

mmHommm

 

.HZ .hucbou hoCmHoaucoz CH COHumUHHQQm

wwwnam van mHHmHu How vmumcmHmmw muoam wcm .mHHmuu m>m£ ou wmumcmemw

muon .mHouucoo cH .HmmH umnm=< :H m>HHm an ou czocx mHmEEmB.HHmEm Hmuoa

.oH «Hams

51

HaucmoHMchHm mum mumHHomHmaam Hmuuma unmumMMHv nuH3_30H HmucouHHon m :H mHmAEDC

Ano.ov.mv ucmHmMMHv hauamoHMchHm¥

.vaMHommm omH3Hmnuo mmchs AHo.ouvmv ucmHmMMHw

N
.ucmaummuu

.Hmm muoam m Scum .Houum vumvcmum nuH3 .HOHQ Hum wounummo mHman>chH mo Hmnasa M

 

 

o~.H H Ho.om H©.H H He.o~ mm.o H who.5~
n n ¥

oo.H H oc.- o~.~ H Ho.mH oo.H H oo.m~
A n «m

mm.o H Ho.m mm.o H Ho.H o~.H H Ho.~

Hmuoe

 

.mam mSUmHEoumm

 

AmsumHHum mmHEmHv
xcnamH£o_cumummm

 

 

 

.muoaa mHHmHu

muon mHHmHH\3
csm mwusam

H

muoHa Houuaoo

mmHommm

 

JHS .HHCbou .hucmuosucoz fiH sOHumoHHaaw «mus

muoHa van .mHHmnu m>m£ ou wmumcmeov muoHa .mHouuaoo CH

“came .HmmH .um:m=< :H m>HHm ma ou caoax mamasma Hawam mo H

Hm wan mHHmHu How wmumcmemw

.Hfico mmHomqm
mQESfi mwmhm><

.NH GHQMH

52

 

mm 0H m on Hm m m¢ om o HMHOH

 

 

 

 

 

 

 

H o o o o o o o o HmwsmoH>mHH maHHmHmv
3mH£m cmHHmuuuHonm

o o o o o o H o o 3:883 3.83
3mnnm vmxmmz

m H H m o N NH q o AchmecH mammuommmmzv
mmnoa wchssn vcmeooz

Hm o N m m m m . N o Amnnooan msommfioummv

manoa nmuoomumans

¢H m H m o H m H o AmHHHumHm
mnumHaoHamE anomhaoummv
mmnoa Hmmv vameooz

 

 

 

mH o H ¢ OH H mH mH o HmsumHHHm mmHamHv
. xcaamHso cumummm

¢H\m NH\B NH\m ¢H\m NH\~ NH\m ¢H\w NH\N NH\m
muoHHH mHHmuu muOHa mHHmHH\3 muon Honucoo mMHommm

can mchHm

 

.Hz .zucsoo mocmuce :o c
:oHumoHHamm mmwsHm vcm mHHmHu nuH3 muon van .mHHmHu zqu muoHa .mwouwwow

:H_.mmHommm ma .Nme .HmBESm :H m>HHm mp ou 230:3 mHmHHHEHHE HHmEm Hmuoe .wH w
HHwH

53

.ucmaummuu Hmm mHOHm m aoum .Houuo cumvcmum suHs mDHm> xmvcH huHmHm>Hw M

Hum muon m Eoum .Houum uumucmum nuHB_.uon Hon 60H9ummo mHmnoH>chH mo Hmaanc_m

N
.ucmEummHu

H

 

 

-wam wwwaHm

NrH

HH.o m mo:H o~.o m Ho.H o¢.o m HH.o mnHm> HHHmHm>Ha
mm.m + H©.¢H mm.H + oo.oH mm.m + mm.¢H mHmsama HHmam
Nme umdws<

m~.o m oo.o mH.o m om.o o~.o m om.o osHmp HHHmHm>HH
m¢.H + mm.m mm.H + oo.H ww.o + Hc.o mHmaamaAHHmam
NmaH HHsa

Hm.o m mn.o Hm.o m mm.o o m o gHg HHHmHm>Ha
mH.H + oo.~ mq.H + mm.~ o + o mHmaama HHmEm
Nme am:

mHOHm mHHwHu muon mHHwHB\3 muon Houucoo mevGH HmSMHSacoaamnmv

muHmHm>Hw mmHommm Hmaama
mHmaams HmchmmHH Ho maHe

 

HHm .NmmH .Hmasam :H m>HHm mg on caocx mHmEEmE HHmam mo Hanan: mwmum><

.Hz .xucsoo moamuoaungz SH :OHumoHHmam mwvsHm cam
mHHmuu :HH3.muoHa van .mHHmHu nuHa muon .mHouucoo :H .wmcHnEoo mmHoQO

.mH anwH

54

were significantly higher numbers of Peromyscus captured
on sludged plots than on either control or trail-only
plots (Table 20).

No significant differences were found among compari-
sons of captures in traps placed in trails with traps
located in interiors.

Mammal species diversity did not correlate well with
FHD (r-0.l4). Peromyscus numbers showed a strong negative
correlation (r-=-O.835, P‘<0.01) with FHD; and also cor-
related negatively with percent vertical cover in the

10-30 cm stratum (rm-‘0. 63, P < 0.05) and percent cover in

the 2 m stratum (r--0.59, P<0.10).

Salamander Populations

The only salamander species found on the study area
was the red—backed salamander, Plethodon cinereus. This
terrestrial species did not demonstrate a detectable popu-
lation response to sludge application when its numbers
were compared between sludged plots and trail-only plots.
There were an estimated 1484 salamanders per ha 1J1 trail-
only Plots, and 1447 per ha in sludged plots. A chi-squared
31‘3le18 comparing numbers of salamanders observed in
trails to numbers seen in interiors showed a significantly
greater (P < 0.01) number of salamanders in interiors

compared to very few animals in trails.

55

.God v 8 HESEHHHU
hHucmonchHm mum mumHHomHmmsm HouumH uamuommHv nuH3_3ou HmucoNHuoH m GH muonadsN

unmaumouu
Hon muon m Eouw .HoHHm wumocmum nuHB .uon Hon mounummo mHmapH>HucH mo Hogans x

 

 

.om.H.HHmm.mH H Hmoo.q mq.H H «Hc.o .aam anomaaopmm

N
I AmsumHHum mmHamav

Hm.m H oc.m mm.o H oo.m H.¢ + mm.o HanaaHno aumummm
NmmH Hm=m=<

 

 

 

 

 

 

oo.H H oo.m mq.H + >.m mm.o + o.H .mmm mnommfioumm
I I I AmaumHHum mMHEmHv
oo.~ + oo.~ mm.o + mm.m mm.H + mm.¢ HonaaHso cumummm
NmmH HHan
mm.o H o.H ww.o H m.H o H o .mam odommaoumm
I I I AmnumHHum mmHEmHv
mm.c + mm.o mm.o + mm.o o + o HGDBQHno auoummm
NmmH Hm:
.muoHa mHHmuu muoHo wHHwHH\3 muon Houucoo moHommw
wow owvsHm H maHaomuu mo oaHH

 

.Hz..%u::oo monmuoaucoz GH aoHumoHHmmw owcsHm
paw memHu ¢uH3 muoHa can .mHHmHu nuHa muoHa .mHoHucoo GH .hHao onooam
Momma NmmH HmESSm :H m>HHm on ou :Bozx mHmaams HHmam mo Hmnabc owmuo>¢ .ON oHan

DISCUSSION

Vegetative Community Response

Both the forest thinning to create application trails
and the sewage sludge application to forest interiors prompted
notable first-year changes in the hardwood forest plant
cxmmnnaity. Composition was altered in tree seedling size
classes by increased density of some species, such as white
aSh, and.decreased density of others (e.g. red oak, beech,
and the maples). Some understory species, such as Canada
mayflower, were suppressed, while post-disturbance plants
such as brambles and mullein were favored. Structural
changes included loss of cover from middle and upper strata
in trails and from lower strata in sludged interiors,
which.1ed.to reduced foliage height diversity.

Because the 2 largest stem density size classes comprised

trees that were sapling-size and larger, they would not

be eXpected to respond to either treatment by increased
densixy'in the first year following treatment. Therefore
density differences in these size classes would logically

be attributable to actual tree removal or to inherent
differences in plant community composition. Forest in-
terior and control densities did not differ, in these

large size classes, either by species or in total,

56

57

indicating that the tree community composition'was homo-
geneous among all plots. Not surprisingly, total large
tree stem densities in entire plots (interior and trail
data combined) were reduced in both types of plots subjected
to trail cutting (i.e. tree removal). Sapling size trees
(> 2 m tall, <10 cm dbh), even though also eliminated

from trails, showed a nonsignificant trend of greater
contrrfl.plot total densities. Since the variances were
particularly high among treatment and control plots in
this size class, there appeared to be inherently high
variation in sapling densities throughout the study area.
Large variances coupled.with the low replication of 3 plots
per treatment, affected the power of tests for treatment
differences .

In order to differentiate plant community changes
resulting from the 2 treatments, vegetation measures in
trails and interior areas were also considered separately.
In the smallest seedling size class (0-30 cm), the reduced
numbers of red oak, combined maple, combined-other species
and total seedlings on sludged interiors when compared to
controls and/or to trail-only plot interiors suggest a
negative response to the sludge itself. Shortly after
treatment many withered seedlings and herbaceous plants
were observed, heavily coated with sludge, indicating
sludge induced leaf and plant mortality. This mortality
may have been a function of applying sludge during the

growing season, since a related experiment involving sludge

58

application the previous fall in a mixed-oak forest pro-
duced no forest interior seedling density differences
(Haufler et al. unpubl. data).

In contrast, the dramatic increases in white ash
seedling numbers on both types of plots with trails,, in
both the interiors and in the trails, suggest a not un-
expected, positive response to forest thinning. White ash is
known to regenerate heavily (via seedlings) in recently cut
northern hardwood forests (Marquis 1965, Borman etal. .1970,
Leak and Solomon 1975). White ash regeneration appeared to
have occurred in this study, either by increased germination or
survival, or both. Other ash seedlings presumably present
before either treatment may have benefited from release
by forest thinning (i.e. trail construction), thereby
gaining sufficient increased growth and/ or survival to
account for the observed greater densities of tall seedlings
(30 cm - 2 m) in the forest interiors of both types of
plots with trails. Bicknell (1982) monitored seedling
growth following the clearcut of a northern hardwood
forest, and her data for first-year post-cut growth by
white ash seedlings substantiate this possibility. In
reality, forest thinning such as that caused by trail
construction may have provided the optimal release con-
ditions for both ash and basswood, which also had greater
tall seedling density in trail-only plots. White ash and
basswood are considered "gap phase" species, which normally

Persist in climax forests by colonizing gaps in the canopy.

59

Such species tend to be able to make more extension growth
in a season than, for instance, sugar maple or beech, and
are best adapted to competing in conditions intermediate
between large disturbed areas and a closed canopy (Marks
1975).

The trails themselves did not receive sludge treatment.
However, they were subjected to considerable disturbance
both by tree removal and later by the sludge applicator.
Therefore, differences in stem densities should reflect
either the effects of trail construction (e.g. trauma during
logging, loss of soil or leaf litter, microclimate
changes, etc.) and/or, in the case of sludged plots, 4
possible physical trauma from the sludge vehicle's tires.
The data exhibit vegetation changes in response both to
characteristics of the trails or their construction and
to trauma during sludging. Small (0—30 cm) red oak seed—
lings, and large (30 cm- 1 m) hophornbeam and combined-
other-species seedlings had significantly reduced densities
only in sludge plot trails, and appear to have suffered
damage by the vehicle. In addition, red oak is another
gap phase species and would seem unlikely to react nega-
tively to light forest thinning such as this study's trail
construction, unless the procedure itself was damaging.
0n the other hand, trails in both types of treatment plots
had fewer large beech seedlings, and this species appears

to have been more vulnerable to trail construction.

60

In density comparisons of trails to interiors within
treatments, the trend for interiors to have similar or
greater seedling densities than trails also suggests losses
due to trail construction or changes in microenvironment.

In trail-only plots, interior densities of red oak, beech,
basswood and total seedlings did not exceed those on con-
trols, but were greater than those in trails. A similar
situation occurred in sludged plots, with sludged interior
densities of hophornbeam, beech, and total large seedlings
exceeding trail densities in the same plots, but not in
controls. Theoretically, the interior vs. trail density
differences could be due to gains in interiors, to losses
in trails, or to a combination of both (e.g. slight gains,
perhaps in response to release or fertilization, when juxta-
posed against moderate losses would together constitute a
significant difference). ,Since in the 2 cases mentioned
above the interior densities were not significantly greater
than in controls, the data favor either of the latter
choices. Particularly in the sludged plots, however, there
is evidence of sludge damage, and the short h-time span (2-

3 weeks) between sludge application and vegetation measures
would not seem long enough for fertilizer-induced growth to
have exerted much measurable effect in the interiors. This
argues against widespread density gains in sludged interiors.
Some apsects of the trails, whether construction, micro-
environment, or use seem to have reduced stem densities.

In the oak forest which received sludge in the fall of 1981

 

61

as a parallel to this study, 1982 trail seedling densities
(0-1 m) were: over 2 times greater in plots that did not re-
ceive sludge (Haufler et al., unpubl. data), which further
supports the hypothesis that sludge applicator destruction
‘was responsible for some of the reduced trail densities.
This effect should be short-term and trail densities should
recover, although not without initial competition from
typically post-disturbance species (such as brambles, pin
cherry, Prunus pensylvanica, and aspen, Populus tremuloides)
(Auchmoody 1979, Bicknell 1982).

Final composition changes were produced, primarily
by trail construction and to a lesser degree by sludge
damage, in the herb-shrub segment of the plant community.
Trails and sludged interiors possessed relatively sparser
distributions of the typically deep woods herbs starflower
and Canada mayflower, presumably due to microenviroment
changes in the trails (perhaps increased light or tempera-
ture, or litter loss) and to sludge-induced damage or en-
‘viornmental changes in the interiors. The several-fold
increase in occurrence of brambles in the trails was expected,
since 53223 spp. characteristically invade after recent
disturbances. EEEEE seeds are usually present in the
forest floor and are quick to take hold in an area when-
ever conditions become favorable (Stearns 1951). Common
Inullein, an early successional inhabitant of old fields,
roadsides and waste places (U.S. Dept. of Agriculture

1971). was an anticipated invading species.

 

62

Plant damage by sludge application, both from the
sludge itself and.from vehicle use of the trails, along with
mechanical removal of larger trees, also produced structural
changes in the forest plant community. ‘When data were
combined to represent entire plots, large tree removal from
the trails resulted in reduced cover above 2 m.in both
types of plots with trails, although these plots retained
'more than 90% of total available cover. Tree removal reduced
trail canopy cover itself to about 65-75%, but trails made
up only 25% of each plot's area. The 16-fold decrease
in trail cover in the 30 cm - 2 m level can also be attri-
buted to trail construction, presumably by physical
damage, since it occurred in all trails regardless of
treatment type. This trail effect accounts for the observed
entire plot cover decrease in sludge plots. In the 10-

30 cm.cover stratum, which had low (less than 5%) cover
throughout the forest, only sludge plot trails and interiors
lost vertical cover. This reduced forest interior low
level cover parallels the decreased woody stem density in
the corresponding size class, and reflects sludge-induced
leaf and plant mortality in the herb-shrub vegetation.
Ultimately, cover reduction in sludge plots frmm the 3
strata above 10 cm was great enough in combination to
drop foliage height diversity indices for sludge plot

trails, interiors, and plots considered as a whole.

63
Small Mammal Response

Small mammal populations, as indicated by the August
1981 trapping data, were moderately high on the study
site. The 2 species of Peromyscus and the eastern.chipmunk
‘were apparently the site's only common small mammal in-
habitants, although the bait used may not have been
attractive to shrews. The final day of trapping produced
from 20% to 50% newly captured individuals, indicating that
1 trap per station was inadequate for capturing a majority
of the site's inhabitants. The fact that Peromyscus
numbers were 1/3 higher on controls plots than on either
set of plots designated for treatment suggested inherent
differences in pre-treatment mouse populations. In 1981,'
however, no habitat data were available for the area, so
it was impossible to explain the mouse capture differences
by any habitat discrepancies. No age or sex differences
'were found between mouse populations.

Trails were cut through both types of treatment plots
in September, 1981. ‘When mammal trapping resumed in
IMay, 1982, numbers had equalized between the 3 types of
plots but were exremely low; The drastically reduced
1982 populations followed an unusually severe winter, with
record-breaking low temperatures coupled with heavier
than normal snows in December and January (NOAA 1982).
The extent to which this weather was responsible in re-

ducing small mammal populations is unknown, but it is

64

expected to have played a major part. The responsible
factors appear to have been widespread, as southern Michigan
Peromyscus populations also suffered a precipitous decline
between 1981 and 1982 (Haigh, MSU personal comm.).

The July trapping period followed sludge application
by less than 1 week, and much of the litter was still wet,
with frequent pooling of sludge. Nevertheless, the treat-
ment had no discernible effect on small mammal population
'numbers or their use of the area. Mammal numbers rose
equally from their May figures on all 3 types of plots.
There was no differential preference for trail or interior
use of the habitat, as indicated by equal captures in both
trap locations. In mid-August, however, sludged plots
yielded significantly greater Peromyscus numbers (twice as
large) than found in Control or trail-only plots. .There'
was insufficient time between sludge application and the
August census (4-5 weeks) for this increase to reflect aug-
mented nmmmml production in response to the treatment.
The gestation period for both Peromyscus species is 21 days,
but new mace do not usually leave their nests for the first
20 days <3f life and therefore would not be in the trapped
population for at least 40 days following conception.
Neither sex nor age ratios differed significantly between
the 3 types of plots, although there were so few adults
that such comparisons are suspect. Thus the increased mouse

captures on.sludged plots were presumably due to differential

65

survival, or to a behavioral response by the animals to
some change in their environment. An increase in survival
might be attributed, for instance, to increased food avail-

ability. Peromyscus spp. are fairly strongly insectivorous

 

(Van Horne 1982), and forest fertilization may result in
increased litter invertebrate populations (Weetman and Hill
1973) .

Neither species of Peromyscus demonstrated a greater
population on a given type of plot than did the other,
nor did either show a greater increase between trapping
periods. Juvenile and subadult fractions of the trapped
populations were slightly higher on sludged plots, but not
significantly so. This argues against attributing the
increase to differential habitat use patterns by young mice,
as a result of niche displacement, which Van Horne (1982)
described in P. maniculatus in the northwest. Trapped
populations of all animals in 1982 comprised over 90% re-
captures for all plots by the final day of trapping, indi-
cating that the estimated populations were reliable for
comparative purposes.

Most studies have failed to correlated P. leucopus
and P. maniculatus spring and summer breeding with any
identifiable, exclusively food-associated parameters, such
as plant species diversity (Verts 1957, M'Colskey and Lajoie
1975). Jameson (1955) asserted that good mast years supported

continued reproduction in Peromyscus through fall and into

66

winter, which led to high populations in the following year.
This would not, however, account for a differential increase
in midsummer on some parts of the study site. Specifically
food-associated variables other than net above-ground
vegetative production, woody species density and herb-shrub
frequency were not examined in this study. Density, fre-
quency, and annual production each failed to demonstrate
a positive response (e.g. increased production) confined
exclusively to sludged plots. For the primarily granivorous-
insectivorous Peromyscus species occupying this habitat,
the measured first year vegetative changes occurring on the
area, (sludged or otherwise) such as increased ash seedling
density, would not be expected to provide significant addi-
tionaly food resources (the invading Egbgg did not produce
fruit in the 1982 season).

Several studies have looked for associations between

Peromyscus numbers and habitat structural diversity or

 

available vegetative cover at various heights; results pro-
vide conflicting evidence. Peromyscus leucopus numbers

have correlated positively with vegetation density below

7.6 cm but not above it (M'Closkey and Lajoie 1975), in-
dicating that the herb-shrub profile may provide an important
component in habitat utilization. Drickamer (Unpubl. data)
found a positive association between structural complexity
canopy l m or more above traps, and g, leucopus numbers,

While 3. maniculatus numbers did not show any correlation

67

with those variables. Verts (1957) found no association
between P, maniculatus populations and cover or lack of it,
while Lobue and Durnell (1959) and Miller and Gretz (1977)
reported negative P. maniculatus population responses to
herb cover in their habitat. Van Horne (1982) separated
cover into several categories (e.g. tree cover less than
1.5 m, seedling cover less than 1.5 m, shrub cover less

than 25 cm), and found positive correlations between adult

 

P, maniculatus densities and all forms of low cover except
seedling cover, which had a negative association with mouse
densities. Juvenile densities correlated negatively with
the cover variables measured. There is, then some evidence
linking higher P, maniculatus numbers, in particular, to

a paucity of low-level vegetative cover. In this study,
‘both.species of Peromyscus appeared to increase equally in
trumbers on sludged plots, and their combined numbers showed
negative correlations with foliage height diversity and
percent vertical cover in the 10-30 cm and greater than 2 m
strata. This suggests that Peromyscus numbers may have been
determined largely by the structure of their vegetative
cover profile or a correlate of it. The strength of such
an argument is somewhat weakened, however, by the 1981
mouse P0pulation figures. Any differences in habitat variables
that might account for higher 1981 Peromzscus numbers on
control plots than on plots designated to receive treatment
were undocumented. All indications from 1982 vegetation

measures suggest that for the vegetation variables measured,

68

the study site was homogeneous prior to treatment. Dice
(1931) and Fitch (1979) maintain that Peromyscus distribu-
tion is influenced by a behavioral habitat selection in
reaction to visual and other stimuli associated with forests
and prairies, and that a suite of interacting factors are
probably responsible. Van Horne's (1982) and Drickamer's
(1979; unpubl. data) data support this. The data from this
study are not comprehensive enough over time or habitat
measures to implicate any causal relationships between

Peromyscus numbers and habitat characteristics. The results

 

do indicate, however, that neither trail nor sludge treatment
has a negative impact, in the senseof lowering a population
even immediately post-treatment, on small mammal population
:numbers or distribution. This was consistent with the
findings of other forest fertilization studies (WOod and
Simpson 1973, Bierei et al. 1975, West et al. 1981, Woodyard
1982).

The appearance of jumping mice among captures in 1982
was not limited to either of the treatments or the controls,
and therefore cannot be attributed to any known isolated
habitat change. Studies of this animal's food habits
(Whitaker 1963, Vickery 1978) indicate that they are simi-
lar to those ‘of woodland deer mice, although jumping mice
consume relatively fewer arthropods and more fungus. There-
fore, there was no reason to suspect that the bait used in
1981 W0111d not have attracted jumping mice, had they been

present. The study site was altered in general character,

69

however, by the creation of application trails and the wide,
east-west access road between the northern study plots.
Jumping mice are known to colonize recently cut, open wood-
lands (Krull 1970, Kirkland 1977), suggesting that this
species may have immigrated onto the study area after it

was opened up by trail construction.

Salamander Response

The red-backed salamender is an entirely terrestrial
salamander, and as such was the only species expected to
occur in workably high densities throughout the study area.
In a similar forest in New Hampshire, it comprised 93.5%
of total salamander biomass, while remaining species occurred
only along streams (Burton and Likens 1975). In the present
study, red-backed salamander populations demonstrated no
response to sludge treatment. Post-sludge censusing began
3 weeks after sludge treatment, after much of the sludge
had dried sufficiently to produce a surface crust, but large
pools still remain. Unfortunately, census procedures were
too labor-intensive to allow for censusing control plot
populations in order to investigate trail treatment effects
on salaumnder populations. The disproportionately high
number of salamanders seen in forest interiors compared to
those seen in trails suggested that application trails do
affect aboverground, rainy night foraging distribution.

Jagger (1978) found that surface population densities of

70

red-backed salamanders (on or above soil level) remain in

a steady state, even when the litter is dry, and are not
correlated with temperature, soil depth or litter depth.
However, the percent of the surface population in the litter
(rather than under rocks and logs) is positively and sig-
nificantly correlated with rainfall. Studies have found
that foraging success was high in wet periods, and that
during rainfall the salamanders shifted their microhabitat
use upward from the litter-soil interface and foraged in-
stead on top of the litter and on vegetation (Burton and
Likens 1975, Burton 1976, Jaeger 1978). Although the present
study did not attempt to analyze salamander habitat compo-
nents, it was noted that many of the salamanders observed

in interiors were crawling on the lower (below 1 meter)
portions of tree trunks. Since trails lacked any vegetation
above 1 m, and all tree trunks were cut as close to ground
level as possible, it may be that vegetation and tree
removal, in particular, rendered application trails sub-

optimal above-grournd foraging habitat for red-backed

salamanders.

SUMMARY AND RECOMMENDATIONS

Sewage sludge disposal on forest lands offers an
attractive solution to a serious waste management problem.
It allows significant nutrient reclamation while potentially

stimulating increased production of both forest and wild-

life resources. The response of a forest ecosystem to sludge

amendment appears to depend heavily on the age and type of
forest, the method of application, and the timing of appli-
cation.

This study examined the first-year response of a
mature northern hardwoods forest to a single application of
sludge, at commercial fertilizer loading levels, using a
heavy equipment agricultural application technique. The
technique required construction of a series of application
trails throughout the forest, and this aspect of the treat-
ment had significant effects on the forest that were inde-
pendent of sludge effects. In the first growing season
following trail construction, the trails appeared to alter
plant community composition. They stimulated tree seedling
and.stump sprout regeneration, favored some post-disturbance
understory species, and suppressed other typically deep

forest understory species. Application trails, by their

71

72

nature, altered forest structure by reducing canopy cover.
Trails did not appear to affect small mammal population

size or use of the area, but did appear to provide suboptimal
foraging habitat for terrestrial salamanders.

Sludge applied during the growing season reduced vege-
tative vertical cover in lower strata and decreased seedling
densities, through leaf and plant mortality. This appeared
to be a short term response. Red-backed salamander popu-

lations showed no effect from sludge application.‘ Peromyscus

 

mice increased in number on sludged plots a month after
sludge application, apparently due to a behavioral response;
other small mammal populations showed no effects from.the
sludge treatment. Sludge and trail construction together
reduced vegetative structural diversity, but neither treatment
affected small mammal species diversity.

Other studies (Campa 1982, Woodyard 1982) have reported

that sludge application to a young jack pine (Pinus banksiana)

 

clearcut improved nutritive quality, digestibility, and
productivity of wildlife forages, increased (small) mammal
diversity, and may have accelerated ecosystem succession.
Thus sludge amendment of forest soils can help achieve a
variety of common wildlife management objectives.

The present study identified some immediate effects
of sludge and its application using a specific technique,
in a mature northern hardwoods forest. The application
technique required cutting trails through the forest, which

set back succession and favored plant species uncommon to

73

woodlands in the narrow, localized strips. In addition,
it opened the forest and ”apparently stimulated strong
regenerative efforts by some dominant climax tree species
while others suffered seedling density losses. Certain of
the favored species, such as brambles and sugar maple stump
sprouts, are highly preferred foods for browsing herbivores.

The immediate effects of sludge on vegetation were
primarily destructive, causing understory stem density and
cover losses, and combining with trail cutting to reduce
vegetative structural complexity. Such damage to existing
plants by the sludge itself may be avoided by applying sludge
when plants are dormant, such as autumn or early spring.
Over time, forest fertilization may lead to increased net
primary production, encouraged in addition by forest thinning
such as that achieved by the sludge application trails.
In a mature forest such as this one, fertilization and/or
thinning may not change timber production appreciably,
(Ellis 1979, Koterba et al. 1979, Sopper and Kardos 1979,
Stone et al. 1982), but fruit and seed.production may in-
crease (Weetman and Hill 1973, Daniel et a1. 1979) and some
understory plants may flourish (Anthony and WOod 1979,
Auchmoody 1979, Koterba et a1. 1979). Such changes may
alter animal community composition, and/or support its in-
creased production.

Any potential wildlife and forest resource benefits

from these changes must be evaluated in light of the costs

I

74

associated with the application technique, which is costly
both in terms of equipment and site preparation. Future re-
search should compare the relative costs and benefits of
this application procedure with a more economical, less
radical one, such as one using heavy irrigation spray
equipment.

The potential bioaccumulation and toxicity and food
chain transfer of sludge constituents in forest flora and.
fauna, including litter-dwelling invertebrates and herpeto-
fauna, are virtually unknown. These topics need study before
the environmental soundness of widespread sludge application
to forests can be evaluated.

Forest land application of sewage sludge is a feasible
habitat management and waste disposal technique. However,
mature forests present technical challenges to application
which need further experimentation and assessment. The
benefits derived from sludge fertilization and forest thinning
may not outweigh the costs of the procedure investigated
in this study, for most wildlife management objectives.
Sludge is, however, a relatively inexpensive, readily avail-
able fertilizer which presents a potentially useful habitat
management tool. It may be used, in various situations,
to increase forage production and nutritive quality, to
alter ecosystem succession, and to change plant and animal

community composition and structure.

LITERATURE CITED

LITERATURE CITED

Anthony, R. G. and G. W. WOod. 1979. Effects of municipal

wastewater irrigation on wildlife and wildlife
habitat. p. 213-223 .23 W. E. Sopper and S. N.
Kerr, eds. Utilization of municipal sewage and
sludge on forest and disturbed land. Penn. State
University Press, University Park. 537 pp.

Auchmoody, L. R. 1979. Nitrogen fertilization stimulates

germination of dormant pin cherry seed. ' Can; J.

Egg. Reg. 9: 514-516.

Bicknell, S. H. 1982. Development of canopy stratification

during early succession in northern hardwoods.
Forest Ecol. and Manage. 4:41-51.

Bierei, G. R., G. W. WOod and R. G. Anthony. 1975. POpula-

tion response and heavy metal concentrations in
cottontail rabbits and small mammals in wastewater
irrigated habitat. pp. 1-9 '12 G. W. Wood et al.,
eds. Faunal response to spray irrigation of
chlorinated sewage effluent. Institute for Research
on Land and Water Resources, Pub. no. 87. Penn.
State Univ. Press, University Park. 89 pp.

Bormann, F. H. et a1. 1970. The Hubbard Brook Ecosystem

Study: Composition and dynamics of the tree stratum.
Ecol. Monographs. 40: 373-388. '

Bormann, F. H. and G. E. Likens. 1979. Pattern and process

in a forested ecosystem. Springer-Verlag, New York,
N.Y. 253 pp.

Brower, J. E. and J. H. Zar. 1977. Field and laboratory

methods for general ecology. Wm. C. Brown, Dubuque,
Iowa. 194 pp.

Burton, T. M. 1976. An analysis of the feeding ecology of

the salamanders (Amphibia, Urodela) of the Hubbard

Brook Experimental Forest, New Hampshire.‘ J. Herp.
10: 187-204. 7

75

76

Burton, T. M., and G. E. Likens. 1975. Salamander popu-
lations and biomass in the Hubbard Brook Experimental
Forest, New Hampshire. Copeia. 1975(3): 541-546.

Campa, H., III. 1982. Nutritional response of wildlife
forages to municipal sludge application. M.S.
Thesis. Michigan State University, East Lansing,
Michigan. 88 pp. .

Chaney, R. L. 1973. Crop and food chain effects of toxic
elements in sludges and effluents. pp. 129-141.
In Recycling municipal _sludges and effluents on
Iand. Nat. Assoc. State Univ. and Land-Grant Colleges,
Washington, D.C.

Chew, V. 1976. Comparing treatment means: A compendium.
Hortscience. 11: 348-357.

Daniel, T. W}, J. A. Helms and F. S. Baker. 1979. Princi-
ples of silviculture. 2nd Ed. ‘McGraw-Hill, New
York, N.Y. 500 pp.

Dice, L. R. 1931. The relation of mammalian distribution
to vegetation type. Scientific MOnthly, 33: 312-
317.

Douglass, R. J. 1977. Effects of a winter road on small
mammals. J, Appl. Ecol., 14: 827-734.

Drickamer, L. C. (Williams College, Massachusetts) Unpubl.
data. 1981. Seminar on Peromyscus leucopus and
P, maniculatus gracilis. Michigan State University
Dept. of ZooIOgy.

Ellis, R. C. 1979. Response of crop trees of sugar maple,
white ash, and black cherry to release and ferti-
lization. Can J, For. Res. 9:179-188.

Fitch, J. H. 1979. Patterns of habitat selection and occur-
rence in the deermouse, Peromyscus maniculatus.
Michigan State University, Publications of the Museum
Biological Series.

Gysel, L. W. and L. J. Lyon. 1980. Habitat analysis and
evaluation pp. 305-327 In S. D. Schemnitz, ed.
Wildlife management techniques manual. 4th ed.
The Wildlife Soc., Washington, D.C. 686 pp.

Haigh, G. (Michigan State University, Dept. of Zoology),
personal communication.

Haufler, J. B. et a1. Unpubl. data. 1982. Influences on
wildlife populations of the application of sewage

sludge to upland forest types; annual report: Oct.
1981-Oct. 1082. 37 pp.

77

Jaeger. R. G. 1978. Plant climbing by salamanders: perio-
dic availability of plant-dwelling prey. Copeia.
1978(4): 686-691.

Jameson, E. W. 1955. Some factors affecting fluctuations
of Microtus and Peromyscus. J. Mammal., 36: 206-
209. .

Kirkland, G. L. 1977. Responses of small mammals to the
clearcutting of northern Applachian forests. J.
Mammal., 58: 600-609. _

Koterba, Mr T., J. W. Hornbeck, and R. S. Pierce. 1979.
Effects of sludge applications on soil water solution

and vegetation in a northern hardwood stand. J.
Environ. anl., 8: 72-78.

 

Krebs, C. J. 1966. DemOgraphic changes in fluctuating

populations of Microtus califOrnicus. Ecol. Mbnog.
36: 240-273.

 

 

Krull, J. N. 1970. Small mammal populations in cut and
uncut northern hardwood forests. New York Fish and
Game Journal, 17: 128-130.

 

 

Leak, W. B. and D. S. Solomon. 1975. Influence of residual
stand density on regeneration of northern hardwoods.
USDA Forest Serv. Res. Pap. NE-310. 7 pp.

Lobue, J. and R. M. Durnell. 1959. Effect of habitat dis-
burance on a small mammal population. J, Mammal.
40: 425-437.

Marks, P. L. 1975. On the relation between extension growth
and successional status of deciduous trees of the
northeastern United States. Bull. of the Torrey

'Bot. Club. 102: 172-177.

Marquis, D. A. 1965. Regeneration of birch and associated
hardwoods after patch cutting. USDA Forest Service
Research Paper NE-32.

M'Closkey, R. T. and D. T. Lajoie. 1975. Determinants of
local distribution and abundance in white-footed
mice. Ecology 56: 467-472.

Michigan State University. Department of Forestry. Unpubl.
data. 1982. Ecological monitoring of sludge ferti-
lization on state forest lands in northern lower
Michigan; annual report, 1982. 52 pp.

78

Miller, D. H. and L. L. Getz. 1977. Factors influencing
local distribution and species diversity of forest
small mammals in New England. Can. J. Zool. 55:
806-814.

National Academy of Sciences. 1978. Municipal sludge
management: Environmental factors. NTIS Report
PB 277 622. Washington, D.C. 152 pp.

NOAA (National Oceanographic and Atmospheric Assn. ). 1981.
Climatological data, annual summary, Michigan.
Vol. 96.

NOAA. 1981-1982. Climatological data, August 1981-August
1982 (by month), Vols. 96 and 97.

Safford, L. O. 1973. Fertilization increases diameter
growth of birch-beech-maple trees in New Hampshire.
USDA-F8 Research Note NE-182, Broomall, PA. 4 pp.

Schmid, J. , D. Pennington and J. iMcCormick. 1975. Eco-i
logical impact of the disposal of municipal sludge
onto the land. pp. 156- 168 In National Conf. on
Municipal Sludge Management and Disposal, 2d,
Anaheim,, California, 1975. Proceedings. Rockville,
MD, Information Transfer, Inc. 237 pp.

Snedecor, G. W. 1956. Statistical methods. 5th ed.
Iowa State Press, Ames, IA. 534 pp.

Sopper, W. E. 1975. Wastewater recycling on forest lands.
In Forest soils and forest land management. Pro-
ceedings of the 4th North American Forest Soils
Conference, Laval University, 1973. B. Bernier and
C. H. Winget eds. Quebec, Les Presses de l' Universite
Laval. 675 pp.

Stearns, F. 1951. The composition of the sugar maple-
hemlock-yellow birch association in northern
Wisconsin. Ecology. 32: 245-265.

Steel, R. G. and J. H. Torrie. 1960. Principles and pro-
cedures of statistics. McGraw-Hill, New York, NY.
481 pp.

Stone, D. M., S. G. Shetron and J. Peryam. 1982. Ferti-
lization fails to increase diameter growth of sawlog-
size northern hardwoods in upper Michigan. Forestry
Chronicle. 58: 207-210.

'Turkq A., R. C. Haring and R. W. Okey. Odor control
technology. Env. Science and Tech. 6:602.

 

79

U.S. Dept. of Agriculture. 1971. 'Common‘weeds of the
United States. Dover Publications, New York,
NY“ 463 pp.

Van Horne, B. 1982. Niches of adult and juvenile deer
mice (P, maniculatus) in seral stages of conifer
forests. Ecology 63: 992-1003.

Verts, B. J. 1957. The population and distribution of 2
species of Peromyscus in some Illinois strip-
mined land. J: Mammal. 38: 53-59.

' Vickery, W. L. 1979. Food consumption and preferences
in wild populations of Clethrionomys a eri and
Napaeozapus insignis. Can. J, Zool. 57 I536-
1542.

Weetman, G. F. and S. B. Hill. 1973. General environmental
and biological concerns in relation to forest
fertilization, p. 19-35 In Forest fertilization
symposium proceedings. USDA Forest Service Gen.
Tech. Rep. NE-3. 246 pp.

West, S. D., B. D. Taber and D. A. Anderson. 1981. Wild-
life in sludge-treated plantations. Pages 115-122
Jg C.S. Bledsoe, ed. Municipal sludge application
to Pacific northwest forest lands. Univ. of
Washington, Seattle, WA. 432 pp.

Whitaker, J. O. 1963. Food, habitat and parasites of the

woodland jumping mouse in central New York. J.
Mammal. 44: 316-321.

Wollum, A. G. and C. B. Davey. 1975. Nitrogen accumulation,
transformation, and transport in forest soils.
pp. 67-108 In Forest soils and forest land manage-
‘ment, Proc. 65 the 4th North American Forest Soils
Conf., Laval University, 1973. B. Bernier and
C. H. Winget, eds. Quebec, Les Presses de l'Uni-
versité Laval. 675 pp.

Wood, C. W. and D. W. Simpson. 1973. Effects of spray
irrigation of treated sewage effluent on wildlife.
Trans. N.E. Fish and Wildlife Conf. 29: 84-90.

Woodwell, G. M. 1977. Recycling sewage through plant
communities. American Scientist. 65: 556-562.

Woodyard, D. K. 1982. Wildlife response to sludge appli-
cation. M.S. Thesis. Michigan State University,
East Lansing, Michigan. 64 pp.

APPENDIX

APPENDIX

List of vascular plants on the northern hardwoods study
site,'Montmorency Co., MI, summer 1982.

 

Common Name

Scientific Name

 

Indian cucumber-root
Wild sarsaparilla
Spikenard

Bellwort

Hooked crowfoot
Sweet cicely
Trillium

Starflower

Bedstraw

Woodland strawberry
White baneberry
Fringed polygala
Violet

Brambles

Columbine

Small Solomon's seal
False Solomon's seal
Canada‘mayflower
waterleaf

Common mullein
Rattlesnake plantains
Indian pipe

Pyrola

Pipsissewa
'Wintergreen
Partridgeberry
Enchanter's nightshade
White lettuce
Heart-leaved aster
Dogbane

Willow-herb

Hop clover

Troup lily

Sedge

Leather wood

Medeola virginiana
Aralia nudicaulis
Aralia racemosa
Uvularia grandiflora
Ranunculus recurvatus
Osmorhiza claytonii
Trillium grandiflorum
Trientalis borealis
Galium.aparine
Fragaria vesca

Actaea pachypoda
Polygala paucifolia
Viola spp.

Rubus spp.

Aquilegia canadensis
Polygonatum pubescens
Smilacina racemosa
Maianthemum canadense
Hydrophyllum virginianum
Verbascum spp.
Goodyera spp.
MonotrOpa uniflora
Pyrola spp.
Climaphila umbellata
Gaultheria procumbens
Mitchella repens
Circaea quadrisulcata
Prenanthes alba

Aster cordifolius
Apocynum spp.
Epilobium.spp.
Trifolium.agrarium
Eurythronium americanum
Carex spp.

Dirca palustris

 

80

APPENDIX (cont ' d.)

81

 

Common Name

Scientific Name

 

Eastern hophornbeam
American beech
American basswood
Sugar maple ‘

Red maple

Striped maple
White ash

Eastern hemlock
Red oak

Paper birch
Yellow birch
Viburnum
Juneberry

Pin cherry

Aspen

Balsam.fir
Dogwood

Ostrya virginiana
Fagus grandifolia
Tilia americana
Acer saccharum
Acer rubrum

Acer pensylvanicum
Fraxinus americana
Tsuga canadensis
Quercus rubra
Betula papyrifera
Betula lutea
Viburnum spp.
Amelanchier spp.
Prunus pensylvanica
Populus spp.

Abies balsamea
Cornus spp.