Content uploaded by Jesper Brandt
Author content
All content in this area was uploaded by Jesper Brandt
Content may be subject to copyright.
Landscape Ecology vol.
1
no.
4
pp
227
-
240
(1988)
SPB Academic Publishing bv, The Hague
Dynamics
of
small biotopes in Danish agricultural landscapes
Peder Agger’ and Jesper Brandt2
‘Ministry
of
the Environment, The National Forest and Nature Agency, Monitoring Section,
Slotsmarken
13,
DK
-
2970
Hoersholm; 21nstitute
of
Geography, Socio
-
Economic Analysis
and Computer Science Roskilde University Centre,
P.
0.
Box
260, DK
-
4000
Denmark
Keywords: agricultural landscape, land use, landscape pattern, landscape elements, patch, network,
boundaries, connectivity, buffering zones
Abstract
The contemporary pattern of small biotopes
-
small uncultivated areas in the agricultural landscape, its
historical development and recent trends have been studied in Eastern Denmark. The study is based on or
-
dinance maps dating back to
1885,
aerial photographs, and field studies (1981 and 1986). The development
of
these areas is seen in changes in agricultural production. A general decrease in the overall density, and
especially in the density
of
the smallest, wet biotopes, is demonstrated. This change, however, is hiding a more
dynamic process
of
removal, establishment and change within the individual biotopes. The consequences on
research, monitoring and management of the agricultural landscape are discussed.
Introduction
Industrialization of agricultural production in Den
-
mark since the second world war has changed the
structure of the agricultural landscape. Agricultur
-
al land covers 65%
of
the national territory. Severe
environmental problems and problems for recrea
-
tional uses have followed. In order to understand
and eventually solve some of these problems, sever
-
al research programs have been started. One of
these involved the study of the dynamic patterns of
small uncultivated habitats (Biotopgruppen
1986),
and the results of this investigation are presented
here.
Hedges, roadside verges, drainage ditches, small
brooks, bogs, marl pits, natural ponds, thickets,
prehistoric barrows and other small uncultivated
areas laying within and between the fields in the
Danish terminology are named ‘small biotopes’
.
Conceptually they correspond to the ‘network’ that
is embedded in a ‘matrix’ of cultivated fields as de
-
fined by Forman and Godron (1986). They can also
be described as ‘ecotopes’, the smallest unit to be
studied in the landscape (Naveh 1984). Other
authors use such terms as ‘small scale landscape ele
-
ments’ (Ruthsatz and Haber 1982), ‘interstitial
habitats’ (Moore
1977), ‘remnant biotopes’ (Erics-
son et
al.
1987) and ‘biotopes in the rural landscape’
(Ihse 1987).
These structures play a decisive role for all the
species that are in any way dependent on fresh sur
-
face water, trees, shrubs, and perennial herbs. Fur
-
ther, small biotopes are important for the amenity,
cultural value, and accessibility of the landscape
and, hence, for its recreational potential. Finally,
agriculture can be dependent on the pattern of
small biotopes (Schmel and Englmaier 1982).
The general trend in agriculture has been a
change from mixed farming toward more special
-
ized and concentrated production. Formerly, aver
-
age size, mixed farms of 15 ha have decreased to
about one
-
quarter
of
all farms. The number
of
228
farms has also decreased from
203,000
in
1950
to
90,000
in
1986.
Increasingly regional concentration
of production has occurred, with cattle raising
mostly in western, and cereals and seeds mostly in
eastern Denmark.
In contrast to the majority
of
other European
countries, a profound reallotment
of
agricultural
land was carried out in Denmark at the end of the
18th century. Since this study covers a period from
1885
to the present, this means that for the whole
study period all the areas attached to a single farm
are clearly defined. The farmstead itself is usually
situated in the middle of its holdings. This land
-
scape is quite different from, and probably easier to
study, than the former landscapes with an atomized
field pattern and extended commons with indistinct
boundaries. Landscape patterns, at that time,
seemed to correspond more with function than ter
-
ritorial ownership.
Today, the appearance of the landscape is basi
-
cally derived from this reallotment. In the young
moraine dominated landscapes of Eastern Den
-
mark the main events influencing the landscape
pattern have been:
(1)
in the beginning, the con
-
struction
of
dikes and drainage ditches and their
later removal;
(2)
in the second half of the 19th cen
-
tury marl digging which left a multitude
of
small
marl pits in nearly every field, and
(3)
more con
-
tinuously throughout the period, the reclamation
of
wetland areas through drainage, road building, and
extraction of gravel. But still, the basic structures
seen today were formed around the year
1800.
The aim of this study is to understand the in
-
fluences
of
agriculture on the landscape and on un
-
cultivated habitats for wildlife. Therefore, we have
concentrated attention on those aspects of the land
-
scape which provide permanent or semipermanent
habitat for wild species. Biotopes in
non-agricul-
tural areas and biotopes which persist for only one
year were excluded from the study. The focus
of
the
study is on what we have defined as ‘small biotopes
in agricultural areas’ (Danish: ‘smaabiotop’). A
small biotope is ‘an uncultivated area that
isperma-
nently covered with vegetation (or water) and situ
-
ated within an agricultural area. Further, the size of
the area should be more than
10
m2 and less than
20,000
m2
(2
ha) for patch
-
biotopes, or
a
length of
more than
10
m and a width of more than
0.1
m for
line
-
biotopes.
Methods
The total area of Eastern Denmark (Funen and
eastward excluding the island of Bornholm) has
been studied at three levels:
1.
On the newest four cm maps, the signature for
small biotopes have been counted and measured ac
-
cording to type in each of
249
evenly distributed one
km2 agricultural areas (see Fig.
1).
2.
In
13
two by two km sampling areas, all small
biotopes were recorded in
1981
and
1986,
and their
owners interviewed about the production on the
farm and the use and plans for the small biotopes.
3.
In five of the
13
sampling areas, the biotopes
have been followed back in time through two series
of aerial photographs and four series of maps to the
oldest ordinance maps from about
1885.
The only criteria for the choice of the
249
one
km2 areas was that they should be chosen imme
-
diately northeast of the point where the UTM-grid-
net lines with numbers ending with
0
or
5
crossed
(which covers
4%
of the total area) unless more
than
25%
were covered with a non
-
agricultural
area, in which case, one of the neighboring squares
was chosen.
The
13
two by two km sampling areas were select
-
ed using a stratified sampling scheme related to
a
previous regional study (see below). The five areas
chosen for historical analysis were selected
so
that
the major landscape types were represented.
Since there is no detailed map of special land
-
scape types in Denmark,
a
biotope relevant regional
description was made to economize the limited
resources for fieldwork. First, the development
of
agricultural technology and structure was seen as
the main force behind the development
of
small
biotopes in the agricultural landscape. Second, this
was applied to
a
concrete landscape modified by
physical geographic components. And third, the
pattern was influenced by urbanization. Within
these three groups
of
conditions, the following
statistical data were selected to classify
115
munici
-
palities (having an average size of
110
km2).
229
GEN
1
:
1.500.000
Fig.
1.
The location
of
the
13
four km* field survey areas (marked with black and a number), and the
249
one
km'
map
-
based test areas
in
the Eastern part
of
Denmark.
Agricultural data:
1.
Cultivated area in
Yo
of total area.
2.
Area of farms
>
50
ha in
070
of total area cul-
3.
Area of permanent grassland in
Yo
of total area
4.
Number
of
cattle per
100
ha.
Physical geographical data:
5.
Area
of
lighter soils in
Yo
of total area.
6.
Area
of
clay soils in
070
of total area.
7.
Area of organic soils in
Yo
of area.
8.
Density of brooks in m per ha.
Urban data:
9. Inhabitants per
100
ha.
10.
Urban population in
Yo
of total population.
11.
Town area in
Yo
of total area.
A classification of the
115
municipalities was
made by principal component analysis and cluster
analysis. The final classification, based on a projec
-
tion of the plane given by the two first principal
components, is shown on Fig. 2.
Of the eight classes recognized on Fig. 2, four
classes were described as variants of the average for
all
of
the municipalities (average classes) and the
rest represented more marked deviations (diverging
classes). The geographical relevance of the
classifi-
cations was judged to be well
-
founded since a clear
relation between the classification and the
region-
alization could be obtained, as shown in Table
1.
tivated.
cultivated.
230
Second princ
106.9
,
66.8
53.5
var.1
40.1
-I
A
-223
201
227.
235
.447
pal component
var.4 Cattle
var.5 Light soil
var.10 Urban population
-205
var.1 Agricultural area
var.8 Dens.
of
ditches
8,
brooks
var.11 Urban area
var.2 Farms
50
ha
327e.211
217-
.I81
2O8e-207
-
219
221.
225
I
B
461
el55
26.7
-1
-169
13.4 -.367
0.0
-431 e485
-
13.4
-
-
26.7
-
-
40.1
-
359
-
53.5
-
-
66.8
I
I I
I
I
I
I
I
I
First
component
-
principal
average classes:
D
E
G
diverging classes:
A
a
-
with signatures for the corresponding regions (see fig.
3)
Fig.
2.
Classification of municipalities in Eastern Denmark for small biotope related variables, based on PCA
-
vector 1
-
vector 2-plane,
describing
68.6%
of the total variance. The projection of the variables best correlated
(>
0.25) with this plane are indicated on the graph.
Table
1.
Relation between classes and regions of municipalities
in Eastern Denmark concerning small biotope
-
related variables.
The classes refer to Fig. 2, the regions to Fig.
3.
Class A
B
C
D
E
F
G
H
Coverage
(Yo)
Region
1
2
3
4
5
6
I
8
9
1
90
22
1
96
10
100
7
1
86
2 1
24
88
8
100
11
10
84
11
2
13
16
The final regional description is shown in Fig.
3.
For the selection of field study areas, representa
-
tive municipalities were chosen from the diverging
and typical regions. Field study areas were meant to
represent ‘average’ areas for the regions. They were
supplemented with representatives of large scale
landscape types (such as river
-
valley
-
bottoms, em
-
banked areas, and dead
-
ice landscapes) which had
not been included because the selection procedure
was based on regional scale information. Thus, the
field area selection procedure was made with two,
probably conflicting, purposes: first,
to
secure a
sample that was representative of small biotope
in-
23
1
Fig.
3.
Regionalization
of
municipalities in Eastern Denmark with small biotope related variables.
fluencing conditions in Eastern Denmark, and se
-
cond, to insure that the main types of agricultural
landscapes and landscape
-
developments were re
-
presented.
For the
249
one km2 areas, measurements of
length or area were made directly from the maps
us
-
ing a millimeter stick and millimeter transparent
paper. Readings from older maps and aerial photo
-
graphs were made with a zoom
-
transfer
-
scope.
With this instrument, it is possible to compare a
field map with an old map, or an aerial photograph.
The field location, type, size, level, coverage with
trees, herbs, reeds and water and human use were
recorded.
The criteria used for separating the biotopes into
types is
as follows:
A.
The classification should correspond (include)
with the definition of signatures on the maps.
B.
It should, as far as possible, correspond to the
common classification, expressed using the ver
-
nacular names, for the different biotopes in
Denmark.
C.
It should allow the fieldwork to be done with
-
in a reasonable time.
D.
It should be strictly objective and permit repe
-
tition of the fieldwork at a later time.
The biotopes were divided into line
-
biotopes,
which are indicated on the maps with true length
but uncertain width, and patch
-
biotopes. Then they
were divided into wet
-
biotopes, with permanent or
semi
-
permanent surface water, and dry
-
biotopes.
Next, they were divided into those covered with
trees and shrubs and those without such vegetation;
and finally, they were divided by function. The
232
Table
2.
Classification of small biotopes.
Line
-
biotopes Patch
-
biotopes
00
Field road verge 16 Wet marl pit
01 Gravel road verge 17 Other wet pit
02 Paved road verge 18 Artificial pond
03 Avenue verge 19 Bog
04 Field divide 20 Natural pond
or
lake
05
Hedgerow 21 Village pond
06
Stone wall 22 Alder swamp
07 Hedgerow on 06 23 Rain water basin
08
Dyke 24 Dry marl pit
09 Hedgerow on 08 26 Other dry pit
10 Slope 27 Barrow
11
Dry drainage ditch 28 Plantation for game
12 Wet drainage ditch 29 Other plantations
13 Canal 30 Natural thicket
14 Brook 31 Solitary tree
15
River 33 Ruderal area
32 Railway dike 33 Ruderal area
43 Treerow* -14 Highpower el
-
mast
44 Treerow on
11
or
12
45 Treerow on
08
46 Hedgerow on
11
or
12
47 Footpath
*‘Treerow’ is a single row of equidistant planted trees.
types are given in Table
2.
There were several problems involved in making
this classification (Agger and Brandt
1984;
and Bio-
topgruppen
1986).
One problem was to separate
biotopes that were physically interconnected. For
the line
-
biotopes, this problem was solved by treat
-
ing the network of line
-
biotopes as the summed
length and not as a number
of,separate biotopes.
Since the type may change along a line, they were
divided in bits of
20
m. In this way, for instance, a
hedgerow was defined as a
20
m line
-
biotope with
more than
50%
coverage of trees and/or shrubs.
Whether a structure was recorded as
a
single bio-
tope,
i.e.,
a marl pit with surrounding tree covered
shores, or as a conglomerate consisting of
a
marl pit
and a thicket was judged either in the field or later
when studying the old maps. The principle followed
was that one biotope should be defined as the
lar
-
gest stretch
or part of a biotope which has had
a
unit history
from the oldest map to the present.
Another problem was the validity of maps and
photographs. There are three problems with using
old maps. These are:
A. The choice of signatures. How have the
bio-
topes been classified?
B. The power of dissolution of the maps: How
large should a biotope be to be recorded?
C.
The obsolescence of the maps: How outdated
was the map at the time of issue?
The instructions for drawing the old large scale
topographical maps
(1
:20
000)
have been examined
and found useful, if one remembers the predomi
-
nantly military purposes for the maps. For in
-
stance, behind the modern definition of dykes as
more than
75 cm high line structures in the land
-
scape, was a functional definition bound to what
was necessary to hide a soldier with his pack. The
main problem has been the character of areas indi
-
cated with signatures for bogs, leys and wet
meadows and the distinction between permanent
pastures and cultivated grassland. On some occa
-
sions, these signatures change back and forth over
the
100
year period in a way which indicates
changes in classification rather than in the land
-
scape.
The transition from the old topographical maps
(1 :20
000)
to the modern topographical maps
(1:25
000),
has meant changes in the definition of
types and increase of minimum size criteria for the
mapping of landscape elements. Neither drainage
ditches along roads, nor field borders are recorded
(unless they are higher than
3/4
m above field
level).
Obsolescence is always present, but is more im
-
portant with the older maps where new issues often
were revised only for features such as roads and
buildings and not for structures of minor impor
-
tance such as many of the small biotopes. With the
modern photogrammatric methods of mapping,
this problem can be considered
of
minor im
-
portance.
The overall validity of the maps was tested for the
five areas that were used for historical analysis.
Supposing
a
constant rate of change, it is possible
to estimate how many different types
of
biotopes
should have been on the maps that might have been
or were issued in
1954,1968
and
1974
by comparing
with aerial photographs in these years with
field-
studies in
1981.
The results are given in Table
3.
‘Other roads’, hedgerows and dykes nearly al-
23
3
Table3.
The percentage
of
the different types
of
small biotopes
that can be estimated to be included
on
maps issued in
1954,
1968
and
1974.
ways are indicated on the maps, but
1/6
of the field
roads and wet linear biotopes are missing, as are
1/3
of
the wet and
1/10
of the dry patch
-
biotopes.
These relations seems rather constant, with the ex
-
ception of dry patch biotopes which have increased
from
66%
in
1954
to
91%
in
1974.
Loss
of
the
smaller part of the biotopes, along with coloniza
-
tion with trees and shrubs of the remaining area,
which make recording more obvious, may be the
explanation. To these should be added the modest
field borders, the ditches along roads and the
1974
88
99
97
87
67
91
__
1954
Field roads
80
Other roads
104
Hedgerows and dykes
100
Wet linear
-
biotopes
86
Wet patch
-
biotopes
67
Dry patch
-
biotopes
66
1968
86
97
100
85
68
75
Table
4.
Distribution
of
line
-
biotopes in the
13
areas.
Total length
070 Average density
Area covered Average width
km
m/kmz
@lo
m
Field road verges
Gravel road verges
Paved road verges
Avenue verges
Field divides
Hedgerows
Dykes
Drainage ditches
Canals
Brooks
Rivers
Slopes
Railway embankments
31.3
19.0
39.2
2.5
73.5
68.6
5.5
31.2
3.9
6.5
4.0
1.4
1.9
10.8
6.6
13.6
.9
25.5
23.8
1.9
10.8
1.4
2.3
1.4
.5
.7
645
391
808
51
1516
1412
114
644
80
134
83
29
40
9.5
3.5
15.3
1.4
11.5
29.2
2.1
10.2
3.3
4.7
5.2
1.3
2.6
2.7
1.6
3.5
4.9
1.4
3.8
3.1
2.8
8.0
6.4
11.5
7.8
11.6
Line
-
biotopes
288.5 100 5947 100 3.1
Total area
of
line
-
biotopes:
88.5
ha or
1.82%
of
the agricultural area.
Table
5.
Distribution of small patch
-
biotopes in the
13
areas.
Total number
vo
Number/km’ Area/km2
@lo
Average size
m’ m’
Bogs
Small lakes
Marl pits
Gravel pits
Thickets
Barrows
Solitary trees
High power pylons
Other patches
127
41
120
11
126
16
31
7
25
25.2 2.6
8.1
.8
23.8 2.5
2.2
.2
25
.O
2.6
3.2
.3
6.2 .6
1.4 .1
5
.O
.5
5356
1977
942
383
6970
74
29
4
1508
31.0
11.4
5.4
2.2
40.5
.4
.2
8.8
-
2048
2334
387
1691
2683
23 1
46
29
2944
All patches
504
100
10.4 17243
100 1662
Total area
of
patch biotopes:
83.6
ha or
1.72%
of
the agricultural area.
234
Table
6.
Densities
of
signatures for small biotopes
on
maps from the late
1970s
expressed as the percentage
of
the density
of
the signatures
on
maps for the same areas in the mid
1880s.
Area number
2 9 10
11
13
Average excluding
area no.
9
Roadside verges
96 177
55
76 63 72
Ditches, brooks,
canals and rivers
Hedgerows and dykes
Bogs in numbers
Bogs in area
Ponds* in numbers
Ponds* in area
Thickets in number
Thickets in area
All
patch-b.number
All patch-b.area
67 190
35
40 16 40
109 190
5
52 69 59
57
0
0
20 82 40
104
0
0
16 87 52
41
11
26 23
35
31
73 80
27
111
140 88
**
**
278 400
**
89 45
39 63 52 61
125 310
39 84
113
90
220
**
1034
** **
144 2625
*
Ponds include lakes, marl pits and gravel pits.
**
Percentages are not calculable as the type did not exist
on
the oldest maps.
Area numbers refer to the numbers in Fig.
1.
smallest biotopes, that are excluded in the mapping
procedure.
Aerial photographs give
a
picture much closer to
reality, but the interpretation may be very difficult
unless maps and/or field observations are avail
-
able. But having these, all biotopes can be properly
localized and measured. Only a more detailed clas
-
sification might give some problems, for instance,
the tree and shrub coverage often will be underesti
-
mated on photographs. Agger and Jensen (1982)
have estimated this bias to be about 12%.
main types of biotopes recorded in the field in 1981
are given in Tables 4 and
5.
Small biotopes covered
172.1 ha or
3.5%
of
the agricultural areas inves-
tigated. These densities varied considerably among
the
13
areas. The total density
of
line
-
biotopes
varied from
3.4
to
11.3
per km2 and patch
-
biotopes
from 4.2 to 26.9 per
km2 depending on geomorpho-
logical and agricultural conditions in the area.
Historical development
In the historical analysis, the starting point was
the oldest map. On this map, each biotope was
identified. From later maps, changes in type, size,
or existence of biotopes were recorded for each lo
-
cation. These could be ‘unchanged’, ‘changed’,
‘just disappeared’, ‘disappeared by amalgamation
with
a
neighboring biotope’, ‘disappeared by being
included in non
-
agricultural area’ or ‘disappeared
and the area now being urbanized’. Any new
bio-
tope appearing on a map was recorded. The proce
-
dure was repeated on the maps or photographs for
each period.
Results
The biotope pattern contemporary densities
The overall changes from
1884
to 1978 are given in
Table
6,
where the density of signatures on the
latest map is expressed as the percentage
of
the den
-
sity in exactly the same areas as in 1884. Four con
-
clusions can be drawn from Table 6: (1) A tendency
for the total density to decrease; (2) large differ
-
ences among the different types of biotopes;
(3)
large differences among the different areas; (4)
Patch
-
biotopes decreased more in number than in
area. Generally, it
is
the smaller and the wetter bio-
topes that have decreased. In contrast, thickets in
-
creased considerably, but not enough to compen
-
sate for the decrease in the other types.
Area number nine deviates strongly from the
others in the development of line
-
biotopes. This
could have been
forseen as it is situated in an area
The total lengths, areas and average densities of the
where a former fiord was reclaimed. The increase
235
comes from building farms and roads, planting
shelterbelts on the light marine soils and construc
-
tion of canals and drainage ditches. Since it is atypi
-
cal, this area has been excluded from the averages
given in Table
6.
Still, this column should be inter
-
preted with caution as the development
of
the other
four areas is strikingly different. They represent
four types of development rather than general
trends in Eastern Denmark.
This method of calculation hides a dynamic
process where former herb covered field borders
may be colonized, hedgerows removed, ponds
drained
so
they become bogs
or
thickets etc. The
overall net changes also hide the fact that some
bio-
topes are being removed in the same period as
others of the same type are being established. In
order to identify these types
of
changes, records
have been kept for each spot that in any of the seven
registrations had carried some type of biotope.
Their dynamics were designated by the terms un
-
changed, changed type, newcomers, disappeared or
intermediate, where the biotope neither appeared
on the first map nor in the field registration.
Z.e.,
a 'lability
-
index' (LI) was computed in order to give
a single expression of the relative stability of the
different types of biotopes. The index is defined as
the number of biotopes of a type that have been
missing in the first, and/or the last registration
divided by the number
of
all biotopes of that type.
These indices will be zero where no changes have
appeared, and one where everything has changed.
The results of this analysis are given in Table
7.
A
comparison between Tables
6
and
7
demon
-
strates that net changes in the content of biotopes
are hiding a far more dynamic process
of
biotope
removal and creation. The number of
patchbio-
topes in the late
1970s
was
61
070
of what it was in the
mid
1880s,
but of these only
71
(or
16%)
have re
-
mained unchanged. The corresponding percentage
for line
-
biotopes is
6%.
Figure
4
shows, for the present stock of biotopes,
the fraction that were indicated on the oldest maps.
Generally, wet biotopes, together with roads and
barrows, form the conservative elements in the
landscape, whereas the dry patch
-
biotopes (planta
-
tions and gravel pits) are new.
Table
7
also unveils part
of
the methodological
UNCHANGED
CHANGED
100
50
0
50
100
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
LIj
Barrows
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
Dikes
I
I
I
I
a
100
50
0
50
100
Fig.
4.
Stability
of
the different types of small biotopes in order
of
decreasing stability showing what fraction
of
each biotope
has remained the same
on
all seven registrations since
1885
(to
the left of the zero
-
line).
problems in this analysis. For barrows, it is seen
that two should have appeared during the period
studied. This, for obvious reasons, cannot be
the case. The explanation is that some
of
the bar
-
rows were missing on two consecutive maps and,
therefore, were recorded as disappeared and later
Table
7.
The fate of small biotopes during the period
1884
to
1981
for different biotopes grouped by their type in
1981
or where
non-
existant that year, its type at first registration.
(For further explanation see the text above)
Total number Unchanged Changed Newcomers Disappeared Intermediate
LI
Roads
Hedgerows
Dikes
Ditches
Canals
+
brooks
and rivers
Footpaths
Other line
-
b
Bogs and
thickets*
Ponds**
Barrows
Other dry
patch-b.***
Total line
-
b
Total patch
-
b
221
188
199
324
16
19
150
88
282
20
59
1117
449
49
2
7
5
6
0
2
13
45
11
2
71
71
14
124
13
34
8
0
4
38
22
1
7
197
68
25
51
6
41
0
0
6
11
8
2
42
129
63
68
2
140
152
1
13
121
21
186
3
2
497
212
65
9
33
92
1
6
17
5
21
3
6
223
35
.71
.33
.90
.87
.12
1
.oo
.96
.42
.76
.40
.85
.76
.69
*
Only self
-
grown thickets.
**
Ponds include here lakes, marl pits and gravel pits.
***
Including planted thickets.
appeared as newcomers.
In spite
of
these methodological problems, the
indicated rank order
of increasing lability certainly
is valid. Because
of
the time needed for
a
natural
colonization with plants and the following succes
-
sion and immigration of the related fauna (often to
be counted in hundreds of years), the analysis can
be said to have demonstrated nearly chaotic living
conditions for anything other than pioneer plant
and animal species. Populations of plants and
animals in the small biotopes often are hit by catas
-
trophic changes, such as pollution, that cannot be
determined from maps and aerial photographs.
Clearly, we need to create more stable conditions
for these biotopes.
Recent trends
Changes in small biotopes in
1981
and
1986
are
given in Table
8.
Generally, the same trends as from
1884
to the late
1970s
are visible in Table
8.
It is the
smallest and wettest biotopes that are disappearing
most rapidly. In the latest period, however, the
decline has slowed for all biotopes except ponds.
They are still disappearing at a high rate.
A
new
trend is the rapid growth of the group, ‘other
patch-
biotopes’, comprized mostly of small abandoned
areas.
The development is not the same in all areas. In
some, the development in the biotope pattern has
stopped, whereas in others (the most intensively
cultivated) it has accelerated. In this way, the
differentiation between agricultural landscapes is
increasing. Further, in
1986
spontaneous margin
-
alization
of
agricultural land, with nine abandoned
fields, was observed. Five years earlier there were
none.
Determining factors
In the fertile and arable Danish landscapes, where
agriculture has existed for more than
5000
years,
the anthropological influence is overwhelming.
More than three
-
fourths
of
the patterns of small
biotopes can be traced to having been established
with a definite agricultural purpose, and the rest
237
Table&
Development of the total content of line
-
biotopes in km
and patch
-
biotopes in number registered on aerial photographs
and in the field
(1981
and
1986)
for
5
x
4
km*.
1954 1968 1981 1986
Line
-
biotopes
147.4
135.0 99.6 97.1
Wet patch
-
biotopes
169 147 106
96
Barrows 17
16
14 14
Thickets*
60 66
69 70
Other patch
-
biotopes
-
-
17
24
All
patch
-
biotopes
246 229 206 204
Rate of change
1954
-
68 1968
-
81 1981
-
86
in
@lo
per year
Line
-
biotopes
-
0.6
-
2.3
-
0.5
Wet patch
-
biotopes
-
1.0
-
2.5
-
2.0
Barrows
-
0.4
-
1.0
0.0
Thickets*
+0.7
+0.3 +0.3
-
-k
6.7
Other patch
-
biotopes
-
All patch
-
biotopes
-
0.5
-
0.8
-
0.2
*Including solitary trees.
have been used, permanently or from time to time,
for grazing, peat cutting, timber and firewood
production. Hence, the history of agriculture is the
key to understanding the history
of
the biotopes.
Agriculture is, however, dependant on the natur
-
al potential of the landscape, which influences field
size and pattern, type
of
crops, the installations
needed, and often indirectly the general farm size
and type of specialization that develops. Therefore,
it is almost impossible to separate which
anthropo-
genic and non
-
anthropogenic factors are influenc
-
ing development. Even where we have registered
several thousand biotopes, any simple objective
statistical test for correlation is bound to fail be
-
cause
of
the number of possible ways in which the
different conditions have influenced the observed
pattern.
Instead our approach has been to see whether
ecological and agricultural conditions could be cor
-
related with the observed densities.
Ecological conditions
Geomorphological conditions are known to in
-
fluence agriculture in Denmark; slopes may prevent
soil cultivation, soil gives varying conditions for
growing crops and hydrological conditions deter
-
mine the need for drainage and irrigation.
A measurement of the effect of slope was ob
-
tained by counting the number of times the borders
of each square kilometer were crossed by elevation
contours on the map. This sum has been correlated
with the densities of the main types of biotopes
within each area. The results showed marked nega
-
tive correlations between relief and the densities of
wet linebiotopes, and marked positive correlations
with wet patchbiotopes. The dry types did not show
any significant correlations.
Based on existing soil maps, higher densities
of
dry linebiotopes (especially the shelterbelts) occur
on lighter soils. Wet linear biotopes were also unex
-
pectedly common on these soils, which can be ex
-
plained by the many canals and drainage ditches in
an area that was situated in
a
reclaimed fiord. The
organic soil types showed a high density
of
wet line-
biotopes, as expected. Wet patch
-
biotopes were sel
-
dom found on light soils, but are often on clay
(marl pits) and organic soils (bogs). The dry
patch-
biotopes were more often found on lighter soils.
The need for draining may have been the most
important factor in the development of agriculture
in Denmark. More than
50%
of the total agricultur
-
al area is drained. Correlation of the areas of organ
-
ic soils with the pattern of wet biotopes indicate that
the percentage of peat soil areas carrying wet patch
biotopes declined from
74%
in
1884
to
37%
in
1981.
This decline illustrates the increased agricul
-
tural independance from the conditions set by na
-
ture over this period.
Agricultural conditions
Nowadays, material input from the biotopes to
agricultural production are few and relatively
unimportant. Excavation of raw materials are con
-
centrated in
a
few, rather big pits. Water for irri
-
gation and watering of cattle is almost entirely
dependant on pumped groundwater. Only in the
production of tree products is there some limited
dependence on the biotope.
23
8
The drainage of surplus water is still important
for the existence
of
wet line
-
biotopes. Most of the
biotopes of this type received water from drain
-
pipes. Solid waste coming from households, agri
-
cultural production and stones removed from fields
was observed in many biotopes. This waste was es
-
pecially abundant among the wet patch
-
biotopes.
At first, it may cause eutrophication and/or oxygen
deficit. In the long run, it leads to their disappear
-
ance.
Today, the most important function
of
the small
biotopes in Eastern Denmark seems to be for mark
-
ing boundaries of fields and estates. In the 13 areas,
there were 208 km of estate boundaries. Of these
88%
carried small biotopes. This percentage varied
among the 13 areas (from 52% to 100%) being
lowest in the most intensively cultivated and most
fertile areas and highest in the newly reclaimed area
(area IX). Almost 2/3
of
the line
-
biotopes were
found in estate boundaries, this was
so
for only
35% of the patch
-
biotopes. This dependence makes
the pattern of small biotopes highly sensitive to fu
-
ture development within the agricultural structure.
Game production is the most important recrea
-
tional function of the small biotopes. The density
of game and hunters in Denmark is extremely high
(averaging four hunters per
km2), and about 95%
of the agricultural area was used for hunting. Game
production lead to both preservation and manage
-
ment of existing biotopes and to creation
of
new
ones. This is especially true for patch
-
biotopes
which are planted around existing ponds and
thickets.
Discussion
Classifcation
of
the data
The classification
of
small landscape elements into
different types of small biotopes has not been pos
-
sible on strictly objective criteria. Rather, it is a
compromise between the restrictions of a unique
combination of the locality and the time period
studied. Only the procedure, but not the types
themselves, may be useful in other landscapes.
The information on biotope structures was rather
crude. Only locality, type, and size have been dis
-
cussed above. We attempted to analyze another
element, pattern, into classes of different combina
-
tions of biotope types. Two types of auto-classifi-
cation were tried: Cluster Analysis and Principal
Component analysis, based on the 13
4km2 and the
249 evenly distributed 1
km2 areas. The analyses
were carried out on two levels: One where the
bio-
topes were grouped into 18 main types, and one
where they were grouped into only four (Brandt
1986).
None of the classifications gave clear results. The
reason for these negative results could be of techni
-
cal origin. It is important to analyze how changes
in the size of the squares could influence the results.
Detailed tests would presuppose a large
-
scaled
landscape survey of chorological units, and such
surveys have not been carried out in Denmark. An
inspiration for this type of mapping could be the
comprehensive standardization described by Haase
et al.
(1985).
Information on plant and animal species in the
small biotopes is as important for nature conserva
-
tion as information on the densities of small bio-
topes. Obtaining biological information is far more
time consuming than mere geographical registra
-
tion
of
habitats. In such ecological studies, space is
only one element among many. Tvevad (1987) gives
a comprehensive review (in Danish) of the literature
on the biology of small biotopes in Danish land
-
scapes. Additional information is in Arnold (1983),
Beebee (1981), Zwoelfer (1984), Mader
et al.
(1986), Dowdeswell(l987) and ten Houte de Lange
(1984).
In the study, species of trees and shrubs were
recorded (Agger and Jensen 1982, 1983 and 1984,
Biotopgruppen 1986). For each second meter, cross
sections were laid along line
-
biotopes and along
lines diagonally crossing the patch
-
biotopes and the
species intersected recorded. This was done in each
tenth linebiotope and each fifth patch
-
biotope.
The analysis demonstrated considerable differ
-
ences in the species composition among the differ
-
ent biotope types.
Syringa vulgaris
and
Prunus
spinosa
were typical for hedgerows. Typical for wet
patch
-
biotopes were
Salix viminalis
and
Salix
cinerea,
for planted patch
-
biotopes,
Picea abies
239
and
Pinus
sp., and for other dry patch
-
biotopes,
Acer pseudoplatanus, Fraxinus excelcior
and
Fagus
sylvatica.
A
considerable number of species were
found in more than one of the types.
Sambucus
nigra
and
Crataegus monogyna
were the two spe
-
cies found most abundantly in all biotope types.
Perspectives for management
Small biotopes are characterized by a continuing
decrease in their density, especially among the wet
biotopes, and a high instability in their pattern,
together with frequent occurrances of pollution and
disturbance. Protection against removal and pollu
-
tion is needed. However, such measures of protec
-
tion must be seen in a broader economic
-
political
context. Declining agricultural prices within the
EEC, in combination with growing pressure for
non
-
agricultural use of the open land for forestry,
nature conservation and recreation, make it likely
that there will be a change in the use of open land
within the next years. Officially, it has been esti
-
mated that the marginalization of agricultural land
will amount to
15%
of
the total national territory
within the next one or two decades.
In fact, two different directions of change can be
seen. Biotope structure may be more stabilized in
areas with poorer agricultural conditions, due to
less pressure for the land use and more environmen
-
tal and recreational protection. Areas with better
agricultural conditions show an intensification of
land use, with a reduction in the number and area
of small biotopes (Brandt and Agger
1988). This
difference calls for different management strate
-
gies.
In areas with spontaneous marginalization, the
biotope structure can be preserved mostly through
voluntary arrangements. The main problem is ac
-
cessibility for the public on private land. In areas
with good agricultural conditions, more strict regu
-
lations are needed. Expropriative conservation can
be used, but it is extremely expensive, and is used
rarely in areas other than those
of
special natural or
cultural interests. Such areas are seldom concen
-
trated on good agricultural land.
In Denmark, the most important single means
of
regulation is paragraph 43 in the Nature Conserva
-
tion Act. It states that changes in the beds of open
watercourses, of lakes, bogs, moors, heaths, salt
meadows and salt marshes of a certain size require
permission from the nature conservation authori
-
ties. Since 1972, the range of biotope types covered
by this paragraph has been broadened several times
and the minimum size has been lowered (Koester
1980 and 1984). Today, it forms a useful instrument
for conservation planning of the rural landscape. It
enables the authorities to practice a diverse ad
-
ministration, restrictive in landscapes where small
biotopes have high priority and more liberal in
other areas. Together with other limited means for
establishing and management of private and public
owned small biotopes, paragraph
43
can be used as
a tool for the improvement of dispersal
of
wildlife
in the fragmented landscape.
Since roadside verges are one of the only expand
-
ing elements in the pattern
of
small biotopes, and
estate boundaries still have a high density
of
small
biotopes, two models are suggested (Agger
et al.
1987; Brandt and Agger 1988):
The road structure
model,
that indicates where marginalization of
fields might satisfy recreational needs, and the
the
boundary model,
which states that all boundaries
around borroughs, parishes and estates shall carry
some small biotopes. Both models will form a
general barrier against further fragmentation of the
habitats for wildlife, and at the same time they can
be used as a skeleton for the creation of a new land
-
scape as marginalization proceeds.
To some degree, this type of development has
received support from an unexpected source. The
severe cases of fish death in the Danish seas since
1981 caused by oxygen deficit induced by fertilizer
runoff from agriculture have made serious counter
-
actions necessary. In the packet of legislative mea
-
sures that are now under discussion in the parlia
-
ment, creation of uncultivated buffer zones along
watercourses probably will come into practice. Be
-
sides limiting the leaching of fertilizers, these zones
will serve as dispersal corridors for wildlife.
Finally, it must be stressed, that a close intercon
-
nection with agricultural development and plan
-
ning is a precondition for successful implementa
-
tion of any of these models, especially within the
240
more intensively used agricultural landscape.
If
general price and market regulations continue to
dominate agricultural policy, as has been the case
within the
EEC,
such models are deemed to fail.
References
Agger, P. and Brandt,
J.
1984.
Registration methods for study
-
ing the development of small
-
scale biotope structures in rural
Denmark.
In
Proceedings of The First International Seminar
on
Methodology in Landscape Ecological Research and Plan
-
ning. Vol. 11, pp.
61
-
72.
Edited by J. Brandt and
P.
Agger.
Agger,
P.
and Brandt, J.
1987.
Smaabiotoper og marginaljorder
(Small biotopes and marginal soils). Miljoeministeriets
projektundersoegelser
1986.
Teknikerrapport nr.
35.
Skov
-
og Naturstyrelsen.
Agger,
P.,
Andersen,
S.S.,
Brandt, J., Nielsen,
T.S.,
Pedersen,
S.
and Tvevad, A.
1987.
Moraenelandets marginaljorder.
(The marginal soils
of
the morainic landscape). Miljoe-
ministeriets projektundersoegelser
1986.
Samlerapport nr.
VIIIa. Skov
-
og Naturstyrelsen.
Agger, P. and Jensen, S.L.
1982.
Hegn og skel i Roskilde amt.
(Hedges and field divides the district of Roskilde).
For-
skningsrapport nr.
32.
Institut for Geografi. Samfundsana-
lyse og Datalogi. Roskilde University Centre.
Agger,
P.and Jensen, S.L.
1983.
Traeer og buske i hegn og skel.
(Trees and bushes in hedges and field divides). pp.
427
-
432.
Ugeskrift for Jordbrug.
23.
Agger,
P.
and Jensen, S.L.
1984.
A method for quantitative in
-
vestigation of the vegetation.
In
Proceedings of The First
In
-
ternational Seminar
on
Methodology in Landscape Ecologi
-
cal Research and Planning. Vol. V, pp.
109
-
115.
Edited by
J. Brandt and
P.
Agger.
Arnold, G.W.
1983.
The influence of ditch and hedgerow struc
-
ture, length
of
hedgerows, and the area of woodland and
garden
on
bird numbers
on
farmland. J. App. Ecol.
20:
Beebee, T.J.C.
1981.
Habitats of the British amphibians
(4):
Agricultural lowlands and
a
general discussion of require
-
ments. Biol. Conserv.
21: 111
-
124.
Biotopgruppen (Agger, Brandt, Byrnak, Jensen og Ursin)
1986.
Udviklingen i agerlandets smaabiotoper ioestdanmark. (The
development
of
small biotopes in the agricultural areas
of
Eastern Denmark). Forskningsrapport nr.
48.
Institut for Ge-
ografi, Samfundsanalyse og Datalogi. Roskilde University
Centre.
Brandt,
J.
1986.
Small
-
biotope structures as
a
synthetizing fea
-
ture in agricultural landscapes.
In
Landscape Synthesis
-
Foundations, Classification and Management. Part
I,
Geo-
ecological Foundations. pp.
52
-
62.
Edited by H. Richter and
G.
Schoenfelder.
Martin
-
Luther
-
Universitaet
Halle-Witten-
berg. Wissenschaftliche Beitraege
1986/16.
Halle(Saale),
GDR.
Brandt, J. and Agger,
P.
(ed.)
1984.
The International Associa
-
tion for Landscape Ecology (IALE). Proceedings of the first
international seminar
on
Methodology in Landscape Ecologi
-
cal Research and Planning, Vol. 1
-
V. Roskilde University
Centre, October
1984.
Brandt, J. and Agger,
P.
1988.
The influence of EEC-agri-
731
-
750.
cultural policy on the conditions for development of biotope
structures in rural landscapes
-
some Danish experiences.
In
Connectivity in Landscape Ecology. Proceedings from the
2nd
International Seminar of IALE. Muenster, FRG, July
1987.
Edited by K.
-
F. Schreiber (Forthcoming).
Dowdeswell, W.H.
1987.
Hedgerows and verges. Allen
&
Un-
win. Boston.
Ericsson et al.
1987.
Poster presented at the
2nd
international
seminar of IALE. Muenster, FRG, July
1987.
Edited by K.
-
F.
Schreiber (Forthcoming).
Forman, R.T.T. and Godron, M.
1986.
Landscape Ecology.
John Wiley
&
Sons,
New York.
Haase,
G.
et al.
1985.
Richtlinie fur die Bildung und Kennzeich-
nung der Kartierungseinheiten der ‘Naturraum
-
typen der
DDR
im mittleren Masstab’. Institut
f.
Geographie u. Ge-
oekologie d.A.d.W. der DDR. Wissensch. Mitt. Sonderheft
3.
Leipzig.
Ihse, M.
1987.
Disappearing biotopes in the rural landscape:
Changed conditions for man, mammals and plants. Poster
presented at
2nd
international seminar of IALE, Muenster,
FRG.
Koester, V.
1980.
Nordic countries’ legislation
on
the environ
-
ment with special emphasis
on
conservation
-
a
survey. IUCN
Environmental Policy Law Paper, No.
14.
Koester, V.
1984.
Conservation Legislation and General Protec
-
tion
of
Biotopes in an International Perspective. Environ.
Pol. Law.
12(4): 106
-
116.
Mader, H.J. et al.
1986.
Experiments zum Biotopverbund-
system
-
tieroekologische Untersuchungen an einer Anpflan-
zung. Schriftreihe fur Landschaftpflege und Naturschutz.
Heft
27.
Moore, N.W.
1977.
Arable land.
In
Conservation and Agricul
-
ture. pp.
23
-
43.
Edited by
J.
Davidson and R. Lloyd. Wiley
and
Sons,
New York.
Naveh, Z.
1984.
Towards a transdisciplinary conceptual frame
-
work of landscape ecology.
In
Proceedings
of
The First Inter
-
national Seminar
on
Methodology in Landscape Ecological
Research and Planning. Vol. I. pp.
35
-
45.
Edited by
J.
Brandt and
P.
Agger.
Ruthsatz, B. and Haber, W.
1982.
The significance of small-
scale landscape elements in rural areas as refuges for en
-
dangered plant species.
In
Perspectives in Landscape Ecol
-
ogy. pp.
117
-
124.
Edited by S.P. Tjallingi and A.A. de Veer.
Pudoc, Wageningen.
Schmel, H.J. and Englmaier, A.
1982.
Zur Bedeutung natur-
naher Kleinstrukturen fur die Landwirtschaft im Rahmen der
Flurbereinigung.
Z.f. Kulturtechn. Flurber.
23: 75
-
86.
Ten Houte de Lange, S.M.
1984.
Effects of landscape structure
on
animal population dynamics.
In
Proceedings of The First
International Seminar
on
Methodology in Landscape Ecolog
-
ical Research and Planning. Vol.
I.
pp.
19
-
31.
Edited by
J.
Brandt and
P.
Agger.
Tvevad, A.
1987.
Smaabiotopernes betydning for flora og fau
-
na. Marginaljorder og miljoeinteresser. (The importance
of
small biotopes for flora and fauna. Marginal soils and en
-
vironmental interests. Miljoeministeriets projektundersoe-
gelser
1986.
Teknikerrapport nr.
38.
Zwoelfer et al.
1984.
Die tierokologische Bedeutung und Bewer-
tung von Hecken. Berichten der Akademie fur Naturschutz
und Landschaftpflege.
LaufedZalzach.
