www.elsevier.com/locate/foreco

Forest Ecology and Management 232 (2006) 19–25

Mounding site preparation for forest restoration: Survival

and short term growth response in Quercus robur L. seedlings

Magnus Lof a,*, D. Rydberg b,1, Andreas Bolte c,2

a Swedish University of Agricultural Sciences, Southern Swedish Forest Research Centre,

P.O. Box 49, S-230 53 Alnarp, Swedenb Swedish Forest Agency, P.O. Box 234, S-291 23 Kristianstad, Sweden

c University of Gottingen, Institute of Silviculture, Department I, Buesgenweg 1. D-370 77 Goettingen, Germany

Received 19 December 2005; received in revised form 18 April 2006; accepted 1 May 2006

Abstract

Mounding site preparation for forest restoration: survival and short term growth response in Quercus robur L. seedlings control of natural

vegetation during afforestation and reforestation is necessary to avoid economical losses through growth reduction and mortality in seedlings. The

present field experiment, carried out on a ground-water influenced site with planted oak, included three site preparation treatments and undisturbed

control (C). The treatments were: repeated herbicide application (H), mounding site preparation (MSP), and mounding site preparation in

combination with repeated herbicide application (MSP + H). The mounds were 25 m long, 2 m wide and 20 cm high inverted mounds on humus,

and this method is sometimes also called bedding. Seedlings were monitored for 3 years and at the end of the experiment 90% of oak seedlings

survived in the various site preparation methods compared with 58% in the undisturbed control. The best growth was obtained when mounding site

preparation was combined with repeated herbicide treatment, producing five times greater seedling biomass compared with the control. Mounding

site preparation resulted in equal growth of seedlings compared with repeated herbicide application. Interference from vegetation had a strong

negative effect on seedling growth while mounding site preparation itself resulted in a positive seedling growth response. We conclude that

mounding site preparation is an efficient tool for forest managers in establishing oak stands and is a good alternative to herbicide treatment on

certain sites. On the other hand, a relatively large disturbance area and deep soil disturbances may impair recreational values and destroy

archaeological remains.

# 2006 Elsevier B.V. All rights reserved.

Keywords: Conversion; Regeneration; Seedlings; Soil preparation; Weed competition

1. Introduction

European temperate broadleaved forest types previously

covered much larger areas than they do today (Hannah et al.,

1995; Bradshaw and Lindbladh, 2005) and restoration of these

forests is believed to be a step towards sustainable forestry (e.g.

Spiecker et al., 2004; Stanturf and Madsen, 2005). Since the

beginning of the nineteenth century, many original broadleaved

woodlands have been replaced by conifer plantations (Kenk and

Guehne, 2001). Although economically attractive for the forest

* Corresponding author. Tel.: +46 40 41 51 19; fax: +46 40 46 23 25.

E-mail addresses: magnus.lof@ess.slu.se (M. Lof),

dan.rydberg@skogsstyrelsen.se (D. Rydberg),

abolte@bfh-inst7.fh-eberswalde.de (A. Bolte).1 Tel.: +46 44 18 67 36; fax: +46 44 10 97 61.2 Tel.: +49 551 39 36 50; fax: +49 551 39 32 70.

0378-1127/$ – see front matter # 2006 Elsevier B.V. All rights reserved.

doi:10.1016/j.foreco.2006.05.003

owner, these plantations have often proven to be unstable and

are subject to windthrow, drought damage and ultimately forest

decline (Oleskog and Lof, 2005). Other negative effects

associated with conifer plantations include enhanced accumu-

lation of organic material, abundant grass colonization of the

forest floor and decreased humus quality (Bolte, 1999; Prietzel,

2004; Jansen et al., 2005).

This study, conducted on a temperate forest site in the

southernmost part of Sweden, aims to increase the under-

standing of efficient establishment of oak stands. At the site, the

former forest stands of beech (Fagus sylvatica L.) and hybrid

Larch (Larix � Europlepis Henry) have repeatedly been storm-

felled. The oak species used in the present study, however, is

among the most wind stable tree species in European forestry

(Lupke von and Spellmann, 1999).

Low survival and growth of seedlings results in economic

losses through extended rotation times and the cost of

M. Lof et al. / Forest Ecology and Management 232 (2006) 19–2520

Table 1

Chemical soil properties of the organic layer and the rooted mineral soil

Organic layer (Oh horizona)

pH (KCl) 4.2

C/N ratio 17.8

Mineral soil at 0–10 cm (Ah horizona)

pH (KCl) 3.8

C/N-ratio 17.9

CEC (mequiv. g�1) 71

Al (% CEC) 63.7

K + Ca + Mg (% CEC) 17.4

Mineral soil at 10–27 cm (Bv horizona)

pH (KCl) 4.1

CEC (mequiv. g�1) 30

Al (% CEC) 84.9

K + Ca + Mg (% CEC) 8.6

Mineral soil at 27–58 cm (Bv/Go horizona)

pH (KCl) 4.1

CEC (mequiv. g�1) 24

Al (% CEC) 87.0

K + Ca + Mg (% CEC) 7.8

Means from two samples.a Classification according to Ad-hoc-AG Boden (2005); Oh: humified organic

layer containing amourphous organic material; Ah: humified mineral top soil

horizon; Bv: cambic mineral soil horizon; Bv/Go: cambic/gleyic mineral soil

horizon.

replacement of dead seedlings (Margolis and Brand, 1990).

Typically, it is necessary to control the natural ground

vegetation during forestation on open sites and this is usually

achieved by using herbicides, mechanical site preparation,

mulching or prescribed burning (e.g. Orlander et al., 1990).

Numerous studies show a positive correlation between all these

methods of vegetation control and growth and survival of

seedlings (e.g. Davies, 1988; Munson et al., 1993; Brose and

Van Lear, 1998; Lof et al., 2004). In general, forest managers

select herbicides because of their effectiveness and relatively

low cost compared with available alternatives. However,

increased environmental awareness has prompted public and

industry concern over herbicide use in forests (Willoughby,

1999). The European commission aims to make a decision on

many herbicides before the end of 2008 and forest managers

may have to look for alternative management practices.

Since the use of herbicides on forest land is restricted in

Sweden, mechanical site preparation such as soil scarification is

carried out on approx. 150,000 ha each year to control

vegetation and to facilitate manual planting (Anonymous,

2005). Mounding site preparation is an old mechanical site

preparation method that has attracted new attention (Sutton,

1993). It has become an alternative to other methods such as

soil scarification, especially in northern boreal coniferous forest

sites, where low soil temperature, high water table and high

competition are serious problems (Hawkins et al., 1995;

Hallsby and Orlander, 2004; Pennanen et al., 2005). Little

research has been done on mounding site preparation in the

temperate forest zone.

Site preparation influences many environmental conditions

simultaneously and the direct cause of any positive effect on

regeneration is often unknown (Margolis and Brand, 1990).

Moreover, there is a great variation among sites, and the growth

response of seedlings due to mechanical site preparation has

seldom been separated from other effects, such as simulta-

neously altered interference from natural ground vegetation

(Munson et al., 1993) or pine weevil herbivory (Hylobius

abietis L) (Lof, 2000a). Although the positive effect of

mechanical site preparation on survival and growth of conifer

seedlings is well established, the same cannot be said for

broadleaves.

This study investigates the influence of mounding site

preparation on the establishment of young oak seedlings. The

specific objectives of this study were: (1) to evaluate survival and

growth response of oak seedlings following mounding site

preparation and to compare site preparation using repeated

herbicide application with the plantation in undisturbed soil and,

(2) to examine whether any positive growth response following

mounding site preparation was a result of reduced competition

from natural vegetation or from the mounds themselves.

2. Material and methods

2.1. Site description

The study site was located in the Skarhult experimental

forest (558500N/138240E, 90 m above sea level) in the south-

ernmost part of Sweden, where a 32-year-old hybrid Larch

(Larix � Europlepis Henry) stand of 4.3 ha had been wind-

thrown in December 1999. The Larch stand had been planted in

1970 following a European beech (F. sylvatica L.) stand that

was wind-thrown in 1967. The parent soil material at the site

was formed by loose moraine sediments and the soil type can be

described as gleyic cambisol (Gleyic B) according to FAO

(1988). The lower soil levels from approx. forty to 60 cm soil

depth are intermittently water saturated and the soil texture is

sandy loam consisting of 25–43% sand, 40–50% silt and 17–

25% clay content (Ad-hoc-AG Boden, 2005).

In December 2004, two soil pits were dug to ground water

level and samples were taken from each horizon and analyzed

for chemical properties (Table 1). All samples from the organic

layer and the mineral soil were oven-dried at 40 8C, sieved

(2 mm), and pH was measured with a digital pH-meter (glass

electrode, WTW GmbH Weilheim, Germany) in 1 mol L�1

KCl (1:2.5). Sub-samples were milled for total N and C

analysis; for the measurements, an automated C and N analyzer

(Carlo Erba NA 1500, Milan, Italy) was used. The mineral soil

samples were further analysed for cation exchange capacity

(CEC) and exchangeable cations (Ca2+, Na+, K+, Mg2+, Al3+,

Fe2+, Mn2+). The cation exchange capacity (CEC) was

determined by percolating (6 h) 2.5 g of the dried and sieved

soil material with 1N NH4Cl. Cations in the percolate were

determined with an atomic absorption spectrophotometry

(AAS, Varian Spectra 300A, Varian, Darmstadt, Germany).

The H+ saturation of the exchange complex was calculated

from the difference of the percolate pH before and after

percolation. The results were corrected by the hydrolyse

reaction of aluminium cations according to Meiwes et al.

(1984), depending on the Al3+ concentration. The CEC results

M. Lof et al. / Forest Ecology and Management 232 (2006) 19–25 21

from the concentration sum of H+ and all measured cations

expressed in units of microequivalents per gram of dry soil

(meq g�1). The CEC saturation of the base cations K+, Ca2+,

Mg2+ and the acid cation Al3+ are given as percentage values

(Table 1).

The soil was acidic, occurring in the aluminium (Al) buffer

range (Ulrich, 1983) with a moderate to low CEC and moderate

CEC saturation of the base cations K, Mg and Ca (Table 1).

With respect to this and the medium status of C/N ratio (AK

Standortskartierung, 2003), overall site nutrition of the topsoil

within the range of seedlings’ roots can be characterised as

moderate.

The mean monthly air temperature and precipitation

measured in January and July at the climate station in Lund,

located 15 km west of the experimental site, averaged�0.3 and

17.9 8C and 67 and 87 mm, respectively, over the 3 years

(Anonymous, 2002–2004). The mean air temperature was high

in July 2003 and in August 2002–2004 (Table 2) and

precipitation was normal following planting in 2002. Pre-

cipitation was low in August 2002, May 2004 and in September

2002–2004 and was high in June 2002 and July 2004.

2.2. Experimental design

The experimental design was randomised blocks with four

treatments in each of the four blocks. The treatments were:

herbicide (H); mounding site preparation (MSP); mounding site

preparation plus herbicide (MSP + H); and untreated control

(C). Each treatment consisted of three randomly placed rows of

bare-root oak seedlings (Quercus robur L., 1/0, 15–25 cm,

Visingso-Herrangen, Sweden) in each block. Each row

consisted of 25 seedlings with a distance of 1 m between

seedlings. The distance between rows/treatments was approx.

2 m. Seedlings were planted manually in the end of April 2002

using a planting spade. Inverted mounds on humus about 25 m

long, 2 m wide, and 20 cm high were made by excavator in

early April 2002, an operation sometimes called bedding

(Sutton, 1993). The herbicide treatments were applied in early

June and in the middle of July each growing season during

2002–2004. The herbicide (glyphosate, 0.3 g m�2 active

Table 2

Monthly average air temperature (8C) and monthly precipitation (mm) during

the 2002, 2003 and 2004 growing seasons at Lund climatic station located

approx. 15 km south-west of the experimental site (Anonymous, 2002–2004)

Month 2002 2003 2004 30-year-mean

Air temperature

May 13.5 12.9 12.0 11.5

June 16.7 16.9 14.3 15.4

July 18.6 19.3 15.7 16.8

August 20.6 18.5 18.4 16.5

September 14.7 14.3 13.9 13.1

Precipitation

May 55 59 17 45

June 96 60 83 56

July 69 65 128 70

August 21 49 76 65

September 21 36 40 64

gradient) was spread in approx. 25 m long and 1 m wide

strips along the planting rows with the seedlings in the middle

of the strips. A shield was used to protect the seedlings. The size

of each block was 27 m � 25 m (0.067 ha) and the distance

between blocks was 4 m. All blocks were located approx. 20 m

from the nearest forest edge. The site was fenced to exclude

larger herbivores.

2.3. Data collection

The seedling diameter (at ground level) and height

(stretched distance from ground level to highest living bud)

were measured on all living seedlings at the beginning of April

in 2003, and 2004, and in the end of September 2004. In April

2004, seedlings were difficult to find in the control treatment

due to abundant ground vegetation. Seedling diameter at

ground level was only measured for the first 10 seedlings in

each row in September 2004. In September 2004, the

percentage of natural vegetation cover was estimated in

40 cm fixed radius plots close to the living seedlings, and

assigned to one of 11 categories, where 0 = 0%, 1 = 1–11%,

2 = 12–21%, 3 = 22–31%. . .10 = 91–100% vegetation cover.

Following growth measurements and estimation of vegetation

cover in September 2004, the percentage of seedling foliage not

shaded by natural vegetation was estimated. A similar scale was

used, where 0 = 0%, 1 = 1–11%, 2 = 12–21%, 3 = 22–

31%. . .10 = 91–100% non-shaded foliage that overtopped

the natural vegetation.

Soil water potential at 10 cm soil depth was measured in

August and September 2004 with gypsum blocks (5201

Soilmoisture Blocks, Soilmoisture Equipment Corp., CA,

USA) at one position in each treatment. The gypsum blocks

were buried down at random selected positions near the oak

seedlings within each treatment.

Photosynthetic photon flux density (PPFD) was measured

(LI-190SA, LiCor Inc., Lincoln NE, USA) at the seedling level

and above the herbaceous vegetation (0.3 and 1.5 m), in early

August 2004 on 2 randomly selected locations per treatment

and block. Measurements were made between 11.00 and

13.00 ha on a clear day.

In early December 2004, three randomly selected seedlings

from each treatment and block (seedling no 5, 6 and 7 in the first

row in each block) were carefully extracted from the soil.

Thereafter, the seedlings were washed under running water and

the dry mass of seedling components was determined after drying

at 70 8C for 72 h. Earlier the same year, all leaves from the same

seedlings had been sampled in the end of September. To

determine leaf area, a sub-sample of 10 leaves from each seedling

was photocopied and measured with a computer image system

(Image access, Micro Macro Bildanalys AB, Sweden). They

were then oven dried at 70 8C for 48 h to determine dry mass.

2.4. Calculations and statistical analysis

The leaf area per seedling (LA) was calculated using total leaf

dry mass and leaf dry mass to leaf area ratio of the sub-sample. To

account for size-related variations following 2003 and 2004

M. Lof et al. / Forest Ecology and Management 232 (2006) 19–2522

Table 3

Mean environmental characteristics near the oak seedlings in four treatments and 3 years

Treatment Vegetation

cover (%)

Overtopping of vegetation

(%) of seedling leaves

Soil water potential

(MPa) (August 7)

Soil water potential

(MPa) (September 17)

Relative

light (%)

C 95.9 39.8 �0.09 �0.24 28.8

H 12.4 91.5 �0.05 �0.05 75.8

MSP 95.9 56.1 �0.04 �0.09 23.8

MSP + H 6.7 97.5 �0.05 �0.06 75.1

Undisturbed control (C), herbicide (H), mounding site preparation (MSP), and mounding site preparation plus herbicide (MSP + H). Light was measured 30 cm above

ground near the seedlings and soil moisture was measured at a depth of 10 cm. For further description of measurements see text.

Fig. 1. Mean seedling survival in the four treatments during 3 years. Open

symbols denote treatments with herbicide and filled symbols denote treatments

without herbicide. Treatments followed by different letters are significantly

different ( p < 0.05). Mean � S.E. For descriptions of treatments see Table 3

and text.

growing seasons, the mean relative growth rate in diameter (RD,

mm mm�1 and year�1) was calculated using the formula:

RD ¼ ðlnðD2Þ � lnðD1ÞÞ=ðt2 � t1Þ (1)

where D1 and D2 denote root collar diameter in the end of each

growing season and t2 � t1 is 1 year.

The general linear model (GLM) procedure for the analysis

of variance was used to perform statistical tests on seedling

survival and growth after calculating treatment averages (SAS

Institute Inc., Cary, NC, USA). Before the survival rates of

seedlings were analyzed, frequencies were transformed

according to Zar (1984) using the formula:

p0 ¼ 1=2ðarcsinðX=ðnþ 1ÞÞ1=2 þ arcsinððX þ 1Þ=ðnþ 1ÞÞ1=2Þ(2)

where p0 is the transformed frequency, X denotes the number of

living seedlings at the end of the experiment, and n is the

number of planted seedlings at the beginning of the experiment.

Where significant F values occurred, Tukey’s multiple range

test was done for further analysis. In the comparisons, p < 0.05

was considered significant.

3. Results

Three years following start of experiment, the natural

vegetation cover was similar in the control and the MSP—

treatment (Table 3). Three growing seasons after planting in

April 2002, the vegetation cover near the seedlings was 96%. In

contrast, in the H- and MSP + H-treatments, the cover of

vegetation was only 12 and 7%, respectively. Even though the

vegetation cover was high in the control and MSP-treatment,

40% and 56% of seedling foliage overtopped the competitive

vegetation in the two treatments. In the H- and MSP + H-

treatments, most leaves were not covered by surrounding

vegetation.

Soil water potential values were high, except for control in

September (Table 3). The PPFD at seedling level (0.3 m) was

similar in the control and MSP-treatment and in the H- and

MSP + H-treatments and corresponded to 29%, 24% and 76%

and 75% of full light (PPFD at 1.5 m), respectively.

Three growing seasons following planting in April 2002,

there was a strong positive ( p < 0.001) effect of all site

preparation treatments on seedling survival (Fig. 1). In H-,

MSP-, and MSP + H site preparation treatments the survival of

seedlings was approx. 90%, whereas only 58% of seedlings

survived in the control (C). In the end of the second growing

season, there was a trend of reduced survival in both the control

(C) and MSP-treatment. Furthermore, we observed that many

seedlings were damaged by voles, especially in the control (C)

following the first growing season. Stem bark was stripped and

gnawing had been evident on the lower part of stems.

There was also a strong positive effect ( p < 0.0001) of the

different site preparation treatments on seedling growth of stem

base diameter (Fig. 2A). The difference was obvious as early as

the end of the first growing season. In September 2004, the

mean stem base diameter of seedlings in the MSP + H-

treatment was approx. twice as large as in the control (C). There

was also a positive effect of both MSP- and MSP + H-

treatments on seedling length compared with the control (C)

(Fig. 2B).

For seedling dry mass, there was a strong significant effect of

treatment in the end of 2004 ( p < 0.001) (Fig. 3A). The dry mass

of seedlings from the MSP + H-treatment was approx. twice as

high compared with the H- and MSP-treatments and five times

higher than in the control (C). The same trend was found for

M. Lof et al. / Forest Ecology and Management 232 (2006) 19–25 23

Fig. 2. Mean seedling stem base diameter (A) and height (B) in four treatments

and 3 years. Open symbols denote treatments with herbicide and filled symbols

denote treatments without herbicide. Treatments in the same box followed by

different letters resulted in statistically significant differences ( p < 0.05).

Mean � S.E. For descriptions of treatments see Table 3 and text.

Fig. 3. Mean seedling dry weight in 2004 (A), mean seedling leaf area in 2004

(B), seedling stem base relative growth rate from 2002 to 2003 (white) and

2003–2004 (grey) (C) and root:shoot dry weight ratio in 2004 (D) in four

treatments. Treatments in the same box followed by different letters resulted in

statistically significant differences ( p < 0.05). Mean � S.E. For descriptions of

treatments see Table 3 and text.

seedling leaf area (Fig. 3B). For relative diameter growth rate

there was no significant difference between treatments in 2004,

although there was a trend of higher relative growth rate in the

various site preparation treatments compared with the control (C)

(Fig. 3C). In 2003, the MSP + H-treatment had a higher

( p < 0.05) relative growth rate than the control (C) and MSP-

treatments (no letters signifying differences shown) (Fig. 3C).

Moreover, although there was tendency for lower root:shoot

ratios in MSP- and MSP + H-treatments, no significant

difference was found between treatments at the end of 2004

growing season (Fig. 3D).

4. Discussion

Mounding site preparation proved to be an efficient

reforestation method. In this study, seedling survival was high

(approx. 90%) when bare-root oak seedlings were planted in the

various site preparation treatment sites that included mounding

site preparation. In comparison, lower survival was found in the

undisturbed control and the result was clear from the first

growing season. This might have been a result of less

transplanting stress for seedlings planted following site

preparation. In this study, seedlings in the control were planted

into already existing natural ground vegetation having developed

over two growing seasons and abundant vegetation may decrease

soil water content (Lof et al., 1998) which might have enhanced

the internal water stress that seedlings undergo following

planting (Kozlowski and Davies, 1975). However, in this case,

after the seedlings were planted a period of high precipitation

followed. Furthermore, we observed much vole damage on

seedlings, especially in the undisturbed control, possibly causing

many to be missed from the first inventory. This agrees with

earlier research on broadleaved seedling survival and establish-

ment in herbaceous environments, where a substantial number of

the established seedlings can be damaged or consumed by rodent

herbivores, and thus disappear over time (Lupke von, 1987;

Manson et al., 2001; Lof et al., 2004; Hytonen and Jylha, 2005).

In 2003, lower survival was found in the MSP-treatment and

in the control (C) than in 2004. This was probably an effect of

when the inventory took place. Survival and growth during the

2002 and 2003 growing seasons were measured in early spring

2003 and 2004. In April 2004, natural vegetation in these two

treatments was abundant and some seedlings may have been too

difficult to find. In September 2004, when leaves were present

on the seedlings, more seedlings were found.

Interference from vegetation had a negative effect on the

growth of oak seedlings. This is similar to the results of numerous

M. Lof et al. / Forest Ecology and Management 232 (2006) 19–2524

other studies (e.g. Davies, 1988; Cogliastro et al., 1990; Munson

et al., 1993; Lof, 2000b). Comparison of RD values in H- and

MSP + H treatments with the control (C) and MSP-treatment

shows a trend of reduced seedling growth for 2 or 3 years after

planting. However, the differences were not always significant.

Furthermore, interpreting relative growth rates may be difficult

since larger seedlings normally exhibit lower values of relative

growth rate than smaller ones (Brand et al., 1987).

Interference from vegetation was mainly below ground in

this study. For seedling foliage that was below competing

vegetation, the relative light level was approx. 24–29% of full

light. This is a light level that results in growth reduction of oak

(Oleskog and Lof, 2005). However, a large proportion of

seedling foliage in the control (C) and in the MSP-treatment

was above competing vegetation and thus had access to full

light. Hence, light was not the primary competitive constraint

between herbaceous vegetation and seedlings.

In accordance with earlier research on conifer seedlings,

mounding site preparation had a positive influence on the

growth of oak seedlings compared with the undisturbed control

(Sutton, 1993; Nilsson and Orlander, 1999; Hallsby and

Orlander, 2004; Pennanen et al., 2005). Stem base diameter

growth, seedling biomass and seedling leaf area were approx.

the same in the H- and MSP-treatments. In the present study, the

herbicide treated strips were only 1 m wide and root

competition may have occurred from the vegetation on both

sides. A much stronger effect of repeated herbicide application

is expected to be found when larger-sized experimental plots

are used (Davies, 1988; Munson et al., 1993; Lof, 2000b).

Seedling height was greater in MSP-compared with the other

treatments including H-treatment. However, it is well known

that biomass partitioning of seedlings shaded by trees and other

competing vegetation may change, i.e. that taller oak seedling

are found where relative light levels decrease (Ammer, 2003).

There was a positive growth response of oak seedlings

following mounding site preparation itself. Although the

vegetation cover in year 3 in the control (C) and in the MSP-

treatment was the same and resulted in the same light level at

30 cm from ground, seedling growth was greater in the MSP-

treatment than in the control (C). This growth effect can

somewhat be attributed to less competition and transplanting

stress for seedlings in the MSP-treatment since competitive

vegetation was not present at the time of planting, as it was in

the control (C). Furthermore, although the same vegetation and

light pattern were found in the H- and MSP + H-treatments,

seedling growth was higher in the MSP + H-treatment than in

the H-treatment. This is in contrast to findings of Sutherland

and Foreman (2000) where planting in mounds resulted in less

growth of Picea Mariana (Mill.) BSP seedlings compared with

those in a repeated herbicide treatment. Munson et al. (1993)

did not find any substantial effect of mechanical site preparation

on conifer seedling growth compared with a repeated herbicide

treatment but they used blade scarification and not mounding.

According to Carlquist (2000), working in boreal forest sites,

mounding creates a positive microclimate for planted seedlings

by increasing heat absorption, which results in better root growth

of planted seedlings and a longer growing season. He also stated

the importance of fine soil texture for good water storage in the

mound. In the present study, the soil consisted of a large fraction

of silt and clay, which is why mounding was a good choice of

method. Such sites are prone to surface frost which is negative for

seedlings and elevated planting positions may have reduced the

influence from frost in this study. Others have stated that

mounding improves the short-term nutrient supply to seedlings

due to a stimulation of humus mineralization by mechanical

disturbance (Orlander et al., 1990). However, there is a risk that in

a later stage of stand development, growth may be retarded due to

the fact that the leached elements had previously been lost from

the ecosystem. Furthermore, on ground-water influenced sites,

mounding increases the part of the rooting zone without

groundwater influence and associated reduced conditions.

Hallsby and Orlander (2004) have pointed out that elevated

planting positions may cause water stress in seedlings during

periods with little precipitation. We experienced such periods

during the study but did not find low water potential in the MSP-

treatment. Soil water potential was not measured continuously

during the three growing seasons and therefore an influence of

soil drought on seedling growth cannot be excluded. However,

we found a trend of lower root:shoot ratio following mounding

site preparation which is in contrast to expected biomass

partitioning following drought conditions and perhaps soil

drought was not important in the mounds in this study.

5. Conclusions

We conclude that mounding site preparation is an efficient

tool for forest managers in establishing oak stands and is a good

alternative to herbicide treatment on ground-water influenced

sites. In the present study, survival and growth of seedlings

following mounding site preparation were equally as high as in

the repeated herbicide treatment over three growing seasons.

However, such high effort in herbicide application is seldom put

into practice. Mounding site preparation results in a relatively

large disturbance area and deep soil disturbance, which is

negative for recreational values and sustainable nutrient

management. In addition, mechanical site preparation such

as mounding may destroy hidden archaeological remains and

therefore, careful site selection is needed.

Acknowledgements

We appreciate the assistance of Ulf Johansson at Tonnersjo-

hedens and Skarhults experimental forests. Our thanks also to

Georg Andersson, Rolf Overgaard and Oriana Pfister (SLU,

Southern Swedish Forest Research Centre) as well as to Tomasz

Czajkowski and Heiko Rubbert (Gottingen University, Institute

of Silviculture) for their help with field work, and to Michelle

Slaney for linguistic improvements. We thank R.F. Sutton and

one anonymous reviewer for constructive comments. Financial

support was received from the Lidellska foundation, the Swedish

Research Council for Environment, Agricultural Sciences and

Spatial Planning and the research program Sustainable manage-

ment in hardwood forests. The root and soil studies were

M. Lof et al. / Forest Ecology and Management 232 (2006) 19–25 25

supported by the German Federal Ministry of Science and

Technology (Junior Professorship A. Bolte).

References

Ad-hoc-AG Boden, 2005. Bodenkundliche Kartieranleitung, 5th ed. Hannover,

Germany.

Standortskartierung, A.K., 2003. Forstliche Standortsaufnahme, 6th ed. IHW,

Eching, Germany.

Ammer, C., 2003. Growth and biomass partitioning of Fagus sylvatica L. and

Quercus robur L. seedlings in response to shading and small changes in the

R/FR-ratio of radiation. Ann. For. Sci. 60, 163–171.

Anonymous, 2002–2004. Swedish meteorological and hydrological institute

(SMHI), Vader och vatten, Norrkoping, Sweden.

Anonymous, 2005. National Board of Forestry. Statistical yearbook of forestry

2005. Official Statistics of Sweden, National Board of Forestry, Jonkoping,

Sweden.

Bolte, A., 1999. Canopy thinning, light climate and distribution of Calama-

grostis epigejos (L.) Roth in a Scots pine forest (Pinus sylvestris L.). In:

Verhandl. d. Ges. f. Okologie 29, Spektrum Akademischer Verlag Heidel-

berg, pp. 9–18.

Bradshaw, R.H.W., Lindbladh, M., 2005. Regional spread and stand-scale

establishment of Fagus sylvatica and Picea abies in Scandinavia. Ecology

86, 1679–1686.

Brand, D.G., Weetman, G.F., Rehsler, P., 1987. Growth analysis of perennial

plants: the relative production rate and its yield components. Ann. Bot. 59,

45–53.

Brose, P.H., Van Lear, D.H., 1998. Responses of hardwood advance regenera-

tion to seasonal prescribed fires in oak-dominated shelterwood stands. Can.

J. For. Res. 28, 331–339.

Carlquist, C.G., 2000. Hogen som markbehandlingsmetod vid skogsodling. Del

I. Report 67, Department of short rotation forestry. Swedish University of

Agricultural Sciences, Uppsala, Sweden, pp. 3–17.

Cogliastro, A., Gagnon, D., Coderre, D., Bhereur, P., 1990. Response of seven

hardwood tree species to herbicide, rototilling, and legume cover at two

southern Quebec plantation sites. Can. J. For. Res. 20, 1172–1182.

Davies, R.J., 1988. Sheet mulching as an aid to broadleaved tree establishment

II. Comparison of various sizes of black polythene mulch and herbicide

treated spots. Forestry 61, 107–124.

FAO, 1988. FAO/Unesco Soil Map of the world. Revised legend, with correc-

tions and updates. World Soil Resources Report 60, FAO, Rome, Italy.

Hallsby, G., Orlander, G., 2004. A comparison of mounding and inverting to

establish Norway spruce on podzolic soils in Sweden. Forestry 77, 107–117.

Hannah, L., Carr, J.L., Lankerani, A., 1995. Human disturbance and natural

habitat: a biome level analysis of a global data set. Biodiv. Conserv. 4, 128–

155.

Hawkins, C., Letchford, T., Krasowski, M., 1995. Artificial regeneration of

spruce on cold, wet soil—10 years along. Water air soil pollut. 82, 115–124.

Hytonen, J., Jylha, P., 2005. Effects of competing vegetation and post-planting

weed control on mortality, growth and vole damages to Betula pendula

planted on former agricultural land. Silva Fennica 39, 365–380.

Jansen, V.M., Chodak, M., Saborowski, J., Beese, F., 2005. Determination of

humus stocks and qualities of forest floors in pure and mixed stands of

spruce and beech. Allg. Forst u. Jgd. -Ztg. 176, 176–186.

Kenk, G., Guehne, A., 2001. Management of transformation in central Europe.

For. Ecol. Manage. 151, 107–119.

Kozlowski, T.T., Davies, W.J., 1975. Control of water balance in transplanted

trees. J. Arboricul. 1, 1–10.

Lof, M., 2000a. Influence of patch scarification and insect herbivory on growth

and survival in Fagus sylvatica L., Picea abies L. Karst. and Quercus robur

L. seedlings following a Norway spruce forest. For. Ecol. Manage. 134,

111–123.

Lof, M., 2000b. Establishment and growth in seedlings of Fagus sylvatica and

Quercus robur: influence of interference from herbaceous vegetation. Can.

J. For. Res. 30, 855–864.

Lof, M., Gemmel, P., Nilsson, U., Welander, N.T., 1998. The influence of site

preparation on growth in Quercus robur L. seedlings in a southern Sweden

clear-cut and shelterwood. For. Ecol. Manage. 109, 241–249.

Lof, M., Thomsen, A., Madsen, P., 2004. Sowing and transplanting of broad-

leaves (Fagus sylvatica L., Quercus robur L., Prunus avium L. and

Crataegus monogyna Jacq.) for afforestation of farmland. For. Ecol. Man-

age. 188, 113–123.

Lupke von, B., 1987. Einflusse von Altholzuberschirmung und Bodenvegetation

auf das Wachstum junger Buchen und Traubeneichen. Forstarchiv 58, 18–

24.

Lupke von, B., Spellmann, H., 1999. Aspects of stability, growth and natural

regeneration in mixed Norway spruce-beech stands as a basis of silvicultural

decisions. In: Olsthoorn, A.F.M., Bartelink, H.H., Gardiner, J.J., Pretzsch,

H., Hekhuis, H.J., Franc, A. (Eds.), Management of Mixed-Species Forest:

Silviculture and Economics, vol. 15. IBN Scientific Contributions, Wagen-

ingen, The Netherlands, pp. 245–267.

Manson, R.H., Ostfeld, R.S., Canham, C.D., 2001. Long-term effects of rodent

herbivores on tree invasion dynamics along forest-field edges. Ecology 82,

3320–3329.

Margolis, H.A., Brand, D.G., 1990. An ecophysiological basis for understand-

ing plantation establishment. Can. J. For. Res. 20, 375–390.

Meiwes, K.-J., Konig, N., Khanna, P.K., Prenzel, J., Ulrich, B., 1984. Che-

mische Untersuchungsverfahren fur Mineralboden, Auflagehumus und

Wurzeln zur Charakterisierung und Bewertung der Versauerung von Wald-

boden. Ber. Forschungz. Waldokosysteme/Waldsterben 7, 1–67.

Munson, A.D., Margolis, H.A., Brand, D.G., 1993. Intensive silvicultural

treatment: Impacts on soil fertility and planted conifer response. Soil.

Sci. Soc. Am. J. 57, 246–255.

Nilsson, U., Orlander, G., 1999. Vegetation management on grass-dominated

clearcuts planted with Norway spruce in southern Sweden. Can. J. For. Res.

29, 1015–1026.

Oleskog, G., Lof, M. (Eds.), 2005. The ecological and silvicultural bases for

underplanting beech (Fagus sylvatica L.) below Norway spruce shelterwood

(Picea abies L. Karst.). Schriften aus der Forstlichen Fakultat der Uni-

versitat Gottingen und der Niedersachsischen Forstlichen Versuchsanstalt,

Band139. J.D. Saurlander’s Verlag Frankfurt am Main, 94 pp.

Orlander, G., Gemmel, P., Hunt, J., 1990. Site preparation: a Swedish overview.

BC Ministry of Forests. FRDA Report 105. ISSN 0835-0752. 61 pp.

Pennanen, T., Heiskanen, J., Korkama, T., 2005. Dynamics of ectomycorrhizal

fungi and growth of Norway spruce seedlings after planting on a mounded

forest clearcut. For. Ecol. Manage. 213, 243–252.

Prietzel, J., 2004. Humus changes after introduction of beech and oak into

Scots-pine monocultures. J. Plant Nutr. Soil Sci. 167, 428–438.

Sutherland, B., Foreman, F.F., 2000. Black spruce and vegetation response to

chemical and mechanical site preparation on a boreal mixedwood site. Can.

J. For. Res. 30, 1561–1570.

Sutton, R.F., 1993. Mounding site preparation: A review of European and North

American experience. New For. 7, 151–192.

Spiecker, H., Hansen, J., Klimo, E., Sterba, H., Skovsgaard, J.-P., Teuffel, von,

K., 2004. Norway Spruce Conversion—Options and Consequences. EFI

Reseach Report 18, S. Brill Academic Publishers, Leiden, Boston. 269 pp.

Stanturf, J.A., Madsen, P. (Eds.), 2005. Restoration of Boreal and Temperate

Forests. CRC Press, Boca Raton, Florida, USA, 569 pp.

Ulrich, B., 1983. Soil acidity and its relation to acid deposition. In: Ulrich, B.,

Pakrath, J. (Eds.), Effects of Accumulation of Air Pollutants in Forest

Ecosystems. D. Reidel Publ. Comp, Dordrecht (The Netherlands), pp. 127–

146.

Willoughby, I., 1999. Future alternatives to the use of herbicides in British

forestry. Can. J. For. Res. 29, 866–874.

Zar, J.H., 1984. Biostatistical Analysis. Prentice-Hall Inc., New Jersey, 718 pp.