FL-08-05
11/13/06
6:57 PM
Page 1399
The Application of Lidar in Woodland Bird
Ecology: Climate, Canopy Structure, and
Habitat Quality
Shelley A. Hinsley, Ross A. Hill, Paul E. Bellamy, and Heiko Balzter
Abstract
Habitat quality is fundamental in ecology, but is difficult
to quantify. Vegetation structure is a key characteristic of
avian habitat, and can play a significant role in influencing
habitat quality. Airborne lidar provides a means of measuring vegetation structure, supplying accurate data at high
post-spacing and on a landscape-scale, which is impossible
to achieve with field-based methods. We investigated how
climate affected habitat quality using great tits (Parus major)
breeding in woodland in eastern England. Mean chick body
mass was used as a measure of habitat quality. Mean
canopy height, calculated from a lidar digital canopy height
model, was used as a measure of habitat structure. The
influence of canopy height on body mass was examined
for seven years during which weather conditions varied.
The slopes and correlation coefficients of the mass/height
relationships were related linearly to the warmth sum, an
index of spring warmth, such that chick mass declined
with canopy height in cold, late springs, but increased
with height in warm, early springs. The parameters of the
mass/height relationships, and the warmth sum, were also
related linearly to the winter North Atlantic Oscillation
index, but with a time lag of one year. Within the same
wood, the structure conferring “best” habitat quality differed
between years depending on weather conditions.
Introduction
Habitat quality varies with both biotic (e.g., Fretwell and
Lucas, 1970; Pulliam and Danielson, 1991; Bourski and
Forstmeier, 2000) and abiotic factors (e.g., Murcia, 1995;
Rathcke, 2000; Martin, 2001). For breeding birds, proximate
weather conditions can be crucial and especially so for
species such as tits (Parus) where reproductive success
depends on the relatively brief availability of an abundant
food supply (Perrins, 1970). Mis-timing breeding in relation
to peak caterpillar abundance can reduce both breeding
success and adult survival (Thomas et al., 2001). Other
aspects of poor weather conditions, that affect both adult
and chick nutrition, have similar effects (e.g., Siikamäki,
1995; Sanz, 1996; Pasinelli, 2001). Habitats which are
acceptable, or even good, under certain environmental
conditions, may be poor under others (Lõhmus, 2003).
Defining habitat quality in terms meaningful to the
organism or population in question is difficult and time-
Shelley A. Hinsley, Ross A. Hill, and Paul E. Bellamy are
with CEH Monks Wood, Abbots Ripton, Huntingdon,
Cambridgeshire PE28 2LS, U.K. ([email protected]).
Heiko Balzter is with the Department of Geography, University of Leicester, University Road, Leicester, LE1 7RH, UK.
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
consuming (Dias, 1996; Pasinelli, 2001). For woodland,
these difficulties are compounded by the three-dimensional
volume and complexity of the habitat. Many studies have
therefore taken the retrospective approach of using bird
reproductive performance and/or territory occupancy as
measures of quality (e.g., van Balen, 1973; Matthysen,
1990; Newton, 1991; Muller et al., 1997). This works well,
but often limits comparison to broad categories of habitat
types, such as woodland versus farmland (Riddington and
Gosler, 1995) or small versus large woods (Hinsley et al.,
1999), and may not identify specific attributes comprising
“quality.” Furthermore, temporal variation in bird performance, in conjunction with the limited time span of most
studies, increases the difficulty of defining quality. When
weather conditions for breeding are good, the influence of
habitat quality on success may be less crucial, and therefore, also less detectable, than when weather conditions
are poor. This may be especially true for species, including
tits, which utilize a super-abundant food supply under
good conditions.
In 2001, breeding conditions were poor for tits in
Cambridgeshire, and nationally in the U.K. (Balmer and
Milne, 2002). For great tits (Parus major) breeding in
nestboxes in a deciduous woodland in eastern England in
2001, we found a strong negative relationship between mean
tree canopy height around the nest site and mean chick
body mass (r2 0.80, P 0.001, n 11, Figure 1a) (Hinsley
et al., 2002; Hill et al., 2004). In contrast, blue tits (Cyanistes
caeruleus) showed a positive relation (r2 0.61, P 0.014,
n 8), but we have insufficient data for other years to
include blue tit in this current work. For great tits, the
direction of the relationship indicated that parents reared
heavier chicks as the mean height of the tree/shrub canopy
around the nestbox decreased. This result seemed counterintuitive given that great tits feed their young chiefly on treedwelling lepidopteran larvae (Perrins, 1979).
To investigate the role of weather conditions on this
phenomenon in more detail, we extended these analyses
over a range of years, during which conditions varied. In
this paper, we use the relationship between canopy height
(measured using lidar) and chick mass for breeding great tits
to demonstrate how weather conditions affected both habitat
quality and our ability to detect the effects of habitat quality
on breeding success.
Photogrammetric Engineering & Remote Sensing
Vol. 72, No. 12, December 2006, pp. 1399–1406.
0099-1112/06/7212–1399/$3.00/0
© 2006 American Society for Photogrammetry
and Remote Sensing
D e c e m b e r 2 0 0 6 1399
FL-08-05
11/13/06
6:57 PM
Page 1400
Figure 1. Relationships between mean chick body mass and mean canopy height for great tit broods in
Monks Wood in 1997 to 2003. The years are arranged in order of increasingly warm springs as measured
by the warmth sum (a) 2001, warmth sum 568, (b) 2000, warmth sum 648, (c) 2003, warmth
sum 660, (d) 1998, warmth sum 661, (e) 1999, warmth sum 709, (f) 2002, warmth sum 750, (g) 1997, warmth sum 759, and (h) warmth sum 771. The results for 2003 are shown twice,
once in graph (h), positioned using the standard method of calculating the warmth sum, and again at (c)
where the warmth sum was calculated using an alternative method based on the winter NAO index. When
placed at (c), the results for 2003 fit the trend of the other six years for the slopes of the relationships
to shift from negative to positive as springs become warmer. The apparently anomalous results for 2003
shown in (h) are explained in the text. Regression equations are given in Table 1.
FL-08-05
11/13/06
6:57 PM
Page 1401
Study Area
Monks Wood National Nature Reserve comprises 157 ha
of deciduous woodland in Cambridgeshire (52°24 N, 0°14
W) in eastern England (Steele and Welch, 1973). It occupies
a north facing slope of maximum angle 14.5° and elevational
range of 6 to 46 m. In order of abundance, the three dominant tree species are common ash (Fraxinus excelsior),
English oak (Quercus robur), and field maple (Acer campestre);
other species include small-leaved elm (Ulmus minor), silver
birch (Betula pendula), and aspen (Populus tremula). The
main shrub species are hawthorn (Crataegus spp.), blackthorn (Prunus spinosa), and common hazel (Corylus avellana). The field layer is dominated by grasses and sedge
(Carex pendula), partly due to deer grazing (mostly muntjac,
Muntiacus reevesi). Apart from a number of well-defined
blackthorn thickets, and some stands of elm, species distribution is heterogeneous (Hill and Thompson, 2005).
Methods
Lidar data for the study area were acquired in June 2000
using an Optech ALTM 1210 scanner (10 kHz laser pulse
repetition frequency, 10° scan angle, approximate 0.25 m
footprint size). First and last return elevation data were
recorded per laser pulse, with an average post-spacing of
1 laser hit per 4.8 m2. Both the first and last return data
were interpolated into Digital Surface Models (DSMs) at 1 m
spatial resolution. From these a Digital Terrain Model
(DTM) and a Digital Canopy Height Model (DCHM) were
derived by a process of adaptive morphological filtering,
thin-plate spline interpolation, and surface subtraction.
This process is explained more fully in Hill et al. (2003),
while the validation and subsequent calibration of the
DCHM is outlined in Gaveau and Hill (2003) and Patenaude
et al. (2004).
We used great tit breeding data for the years 1997 to 2003,
i.e., three years either side of the year 2000 in which the lidar
data were collected because canopy height was unlikely
to have changed significantly in this period. Tree-dwelling
lepidopteran larvae comprise the main component of diet for
great tit chicks, and hence the mean canopy height around the
nest site was used as a direct measure of structure likely to
influence food supplies. For example, a taller canopy, indicating older, more mature trees should be associated with better
feeding conditions (Perrins, 1979). Mean canopy height was
calculated for a sample area of 54 m 54 m centered on each
occupied nestbox (Hinsley et al., 2002). Without knowing the
actual foraging locations of the birds, this sample area was
assumed to be representative of at least the core of the birds’
territories. Mean height for each sample area was calculated
from the 2916 (i.e., 54 m 54 m) 1 m2 pixels of the DCHM.
Chick body mass data were available from records of
great tits breeding in a total of 22 nestboxes in Monks Wood
(Hinsley et al., 1999; Hinsley, unpublished data, 2004).
Mean chick body mass for each brood was used as an
indicator of foraging conditions in the territory, and hence
as an indirect measure of territory quality, because it was
influenced by the effects of food abundance and distribution
and by the adults’ abilities to find prey and deliver it to the
nest (Hinsley et al., 2002). Furthermore, several studies
have shown that heavier fledglings are those most likely
to survive to enter the breeding population (Both et al.,
1999; Perrins and McCleery, 2001; Monrós et al., 2002).
Chick body masses were measured at 11 days of age (day of
hatching 0) with a Pesola spring balance. Mean body mass
was calculated for each brood, excluding runts (runts were
defined as chicks too small to ring at 11 days and were
rare). The number of broods measured each year ranged
from eight to twelve (Table 1). Territory quality will also
affect other components of breeding success including
clutch size and the numbers of chicks fledged. However,
when producing eggs, females are free to move more widely
in search of resources than when they are feeding young in
the nest, making it more difficult to define a core territory
area. Similarly, a chick in too poor a condition to survive
may still be capable of leaving the nestbox and thus may
bias estimates of success based on numbers fledged.
For each of the seven years, we calculated the relationship between mean chick body mass and mean canopy height
around the nestbox using linear regression analysis (Minitab™
Release 13) (Figure 1). We then used regression analysis to
investigate how the directions (slope) and strengths (r) of
these seven relationships were affected by both local weather
conditions during the breeding season (using the “warmth
sum” – see below) and by larger-scale climate processes
operating over a longer time-scale (using the winter index of
the North Atlantic Oscillation). In general, breeding success
in tits is better in early, warm springs than in late, cold ones
(Slagsvold, 1976; Hinsley et al., 1999), and therefore the
effects of local climate were summarized using the warmth
sum (McCleery and Perrins, 1998). The warmth sum is the
sum of the maximum daily temperature from 01 March to
25 April and has been shown to be the most useful of several
tested indices of spring temperature in relation to great tit
breeding parameters (Perrins and McCleery, 1989). Thus,
the slopes and correlation coefficients of the seven body
mass/height relationships were plotted directly against the
warmth sum to examine how local climate affected the
relationship in different years. The warmth sum for Monks
Wood for each of the seven years was calculated from daily
records collected by a Meteorological Office weather station
located within 100 m of the southern boundary of the wood.
TABLE 1. RELATIONSHIPS BETWEEN MEAN CHICK BODY MASS (G) AND MEAN CANOPY HEIGHT (M) FOR GREAT TIT BROODS IN MONKS WOODS IN 1997 TO
2003. YEARS ARE LISTED IN ORDER OF INCREASINGLY WARM SPRINGS AS MEASURED BY THE WARMTH SUM. VALUES OF THE WINTER NAO INDEX ARE SHOWN
AS USED IN THE ANALYSIS, E.G., VALUE FOR 1997 IS THAT FOR THE WINTER OF 1995 TO 1996. RESULTS FOR 2003 IN ITALICS ARE LISTED USING THE VALUE
OF THE WARMTH SUM CALCULATED FROM THE WINTER NAO INDEX (SEE DISCUSSION). TIMING OF BREEDING IS EXPRESSED AS THE ANNUAL MEAN FIRST EGG
DATE WHERE 01 APRIL 1. STANDARD ERRORS OF REGRESSION COEFFICIENTS ARE GIVEN IN PARENTHESIS
Linear Regression
Equation
Year
2001
2000
2003
1998
1999
2002
1997
2003
mass
mass
mass
mass
mass
mass
mass
mass
21.4
18.7
18.0
20.1
16.2
11.6
12.9
18.0
(0.54)
(0.66)
(0.44)
(2.06)
(2.56)
(4.25)
(1.96)
(0.44)
0.260
0.056
0.047
0.140
0.127
0.385
0.323
0.047
(0.04)
(0.05)
(0.03)
(0.16)
(0.18)
(0.32)
(0.15)
(0.03)
height
height
height
height
height
height
height
height
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
r2
P
No. of
Broods
Warmth
Sum
NAO Winter
Index
1st Egg
Date
0.82
0.16
0.15
0.07
0.05
0.15
0.45
0.15
0.001
0.281
0.209
0.417
0.492
0.262
0.069
0.209
11
9
12
11
11
10
8
12
568
648
660
661
709
750
759
771
2.80
1.70
0.76
0.20
0.72
1.89
3.78
0.76
27
20
22
17
11
9
9
22
D e c e m b e r 2 0 0 6 1401
FL-08-05
11/13/06
6:57 PM
Page 1402
The North Atlantic Oscillation (NAO) is a major mode
of atmospheric circulation variability in the northern
hemisphere and is known to have an important impact
on climate in the northern hemisphere, particularly in
winter (Hurrell and van Loon, 1997; Hurrell et al., 2001).
Many recent studies concerned with the effects of climate
change have demonstrated links between the winter NAO
index and the phenology and ecology of birds and other taxa
(e.g., Beebee, 1995; Crick et al., 1997; Forchhammer et al.,
1998; Brown et al., 1999; Sanz, 2002; Sanz, 2003). Therefore,
we investigated the influence of this larger-scale climate
phenomenon on the body mass/height relationships by
plotting the slopes and correlation coefficients against the
winter NAO index. We also plotted the relationship between
the winter NAO index and our measure of local climate, the
warmth sum. Both these relationships with the winter NAO
index involved a time lag of one year; this is explained in
more detail in the results and discussion. We used additional data from 2004 (outside of the seven year period
spanning 2000, the year the lidar data were collected) to
show how the winter NAO index could out perform the
warmth sum as a predictor of the body mass/height relationships under certain conditions and time scales. The winter
NAO index was obtained from the data of Hurrell (2005).
Results
Despite the strong negative relationship between mean chick
mass and mean canopy height found in 2001 (Figure 1a),
results for the other six years appeared inconsistent on
first inspection, being largely non-significant and having
both positive and negative slopes (Table 1). However, the
two relationships with the highest statistical significance
occurred in a particularly warm (1997) and a particularly
cold (2001) spring which corresponded to years in which
breeding started early and late, respectively. This influence
of local climate on the mass/height relationship was clearly
demonstrated by the variation in the direction and strength
of the relationship with warmth sum (Figure 2a and 2b). In
cold springs (small warmth sum), the slope of the relationship between mass and height was negative, whereas in
warm springs (large warmth sum) it was positive, and the
change between these two extremes was linearly related to
the warmth sum (Figure 2a). Similarly, the strength of the
relationship (correlation coefficient, r) was highest at the
two extremes of cold and warm springs and, although the
individual relationships for intermediate springs were not
significant, overall, r was linearly related to the warmth sum
(Figure 2b, Table 1). The results for 2003 did not fit this
pattern because the warmth sum, calculated as the sum of
the maximum daily temperatures from 01 March to 25 April,
appeared to be an over-estimate. However, an alternative
value for the warmth sum, calculated using its relationship
with the winter NAO index (Figure 3, Table 1), gave a result
consistent with the trend for the other six years (Figure 2).
This difference, dependent on the method of calculation of
the warmth sum, may have been due to a shift in the timing
of breeding in contrast to the fixed calculation period of the
warmth sum and is explored further in the discussion.
Both the slopes and the correlation coefficients of the
mass/height relationships were also linearly related to the
winter NAO index (Figure 4), but in both cases, the results for
2003 no longer appeared to be outliers (as in the relationships with warmth sum, Figure 2). Without the results for
2003, the warmth sum was the better predictor of both slopes
and correlation coefficients, but when 2003 was included,
the winter NAO index was the better predictor. The apparent
influence of the winter NAO index on both the warmth sum
and the mass/height relationships is demonstrated using
1402 D e c e m b e r 2 0 0 6
Figure 2. Influence of local temperature conditions, as
summarized by the warmth sum, on (a) the direction
and (b) the strength of the relationships between mean
chick body mass and mean canopy height for great tit
broods in Monks Wood in the years 1997 to 2003.
Results for 2003 are indicated by arrows. Open symbols
show results for 2003 plotted using a value for the
warmth sum calculated from the winter NAO index – see
discussion. Excluding 2003: (a) slope 2.28(0.37) 0.00343(0.00054) warmth sum, r2 0.91, P 0.003,
n 6; (b) correlation coefficient 5.52(0.42) 0.00801(0.00061) warmth sum, r2 0.98, P 0.001,
n 6. Standard errors of coefficients are given in
parenthesis. Including 2003 as represented by the solid
symbols: (a) slope 1.67(0.66) 0.00247(0.00095)
warmth sum, r2 0.58, P 0.048, n 7; (b) correlation coefficient 3.97(1.54) 0.00557(0.0022)
warmth sum, r2 0.56, P 0.053, n 7.
observed and predicted results for 2004. Using the relationships in Figure 4, and the NAO index for the winter of
2002/03, i.e., a value of 0.20 (Hurrell, 2005), the predicted
values for 2004 for the warmth sum and the parameters of
the mass/height relationship were: warmth sum 675, slope
0.0308, r 0.138. The actual values were: warmth sum
694, slope 0.0394, r 0.146 (Figures 3 and 4).
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
FL-08-05
11/13/06
6:57 PM
Page 1403
Figure 3. Relationship between the warmth sum, calculated from Monks Wood Meteorological Station data, and
the winter NAO index, with a time lag of one year in the
winter NAO index, i.e., warmth sum for 1997 plotted against
winter NAO index for the winter of 1995/96, etc. Data are
for the years 1997 to 2003. Results for 2003 are indicated by an arrow. The open triangle shows the value for
2004 predicted from the winter NAO index of 2002/03, the
open circle shows the actual value for 2004. All years:
warmth sum 696(21) 23.8(10) wNAOi, r2 0.52,
P 0.069, n 7. Excluding 2003: warmth sum 680(15)
26.6(6.8) wNAOi, r2 0.79, P 0.017, n 6.
However, all the relationships with the winter NAO
index showed a one year time lag. Both of the relationships
shown in Figure 4 use the winter NAO index for the winter
before the one immediately preceding breeding, i.e., slope
and correlation coefficient for the 1997 mass/height relationship are plotted against the winter NAO index for the winter
of 1995–96, etc. There was no relationship between either
the slopes or the correlation coefficients and the winter
NAO index for the immediately preceding winter (P 0.631
and P 0.573, respectively). The warmth sum and the
winter NAO index, with a one-year time lag, were negatively
correlated (Figure 3), but there was no relationship between
the warmth sum and the winter NAO index for the immediately preceding winter. Positive values of the winter NAO
index are usually associated with warmer temperatures in
western Europe (Hurrell et al., 2001; Sanz, 2002) and thus,
without a time lag, a positive relationship between the
winter index and the warmth sum would be expected.
Discussion
There are numerous field-based methods of assessing structural territory quality for woodland birds, but they require
direct access, are laborious and time consuming, and often
difficult to integrate into an estimate of habitat quality. Most
especially, they can usually only be applied on a small scale
relative to the geographical extent and three-dimensional
complexity of woodland. In contrast, lidar supplies accurate
data at high post-spacing and on a landscape-scale, allowing
woodlands to be mapped in their entirety. The success of
lidar in this study to supply a meaningful measure of habitat
quality was probably due to canopy height acting as a surrogate for tree age/maturation and hence canopy volume.
Previous work has shown a good relationship between fieldbased canopy density estimates and lidar-measured canopy
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
Figure 4. Influence of the winter NAO index on (a) the
direction and (b) the strength of the relationships between
mean chick body mass and mean canopy height for great
tits breeding in Monks Wood in the years 1997 to 2003.
Results for 2003 are indicated by arrows. Open triangles
show the values for 2004 predicted from the winter NAO
index for 2002/03, open circles show the actual values
for 2004 (symbols jittered slightly for clarity). (a) slope
0.0489(0.054) 0.0906(0.026) wNAOi, r2 0.71,
P 0.018, n 7; (b) correlation coefficient 0.094(0.10)
0.221(0.049) wNAOi, r2 0.80, P 0.006, n 7.
height (Hinsley et al., 2002). In Monks Wood, tree height
derived from the DCHM has also been used successfully as
a surrogate for the presence of a well developed shrub layer
beneath the top canopy (Broughton et al., 2006), but such use
will be site specific because tall, closed canopy woodland
may also have little or no shrub layer. However, shrub density
is related, at least in part, to canopy density and the degree
of canopy closure, and thus, characteristics such as laser
penetration between first and last return and gap fraction, in
D e c e m b e r 2 0 0 6 1403
FL-08-05
11/13/06
6:57 PM
Page 1404
combination with height, may supply more information on
interior structure. Similarly, lidar data collected for deciduous
woodland in winter may assist in assessing presence and/or
densities of shrub and ground layer vegetation.
Previous work on woodland tits has shown that heavier
fledglings are those most likely to survive to enter the breeding population (Both et al., 1999; Perrins and McCleery, 2001;
Monrós et al., 2002), and thus, it was expected that the best
quality territories in Monks Wood should produce the heaviest
fledglings. The relationships between mean chick mass and
canopy height for the seven study years showed both negative
and positive slopes and the r2 values varied from 0.05 to 0.82
(Table 1). This variation in both the direction of the relationship and its ability to function as a measure of habitat quality
appeared to be influenced by the local weather conditions
during each particular breeding season. As conditions changed
from cold and late to warm and early (as summarized by the
warmth sum), the direction of the relationship changed from
negative to positive (Figure 2). In cold, late seasons mean
canopy height was a good predictor of habitat quality, in
warm, early seasons it was also good, but in intermediate
conditions its predictive power was negligible.
This change in the direction of the relationship indicated that the habitat structure providing the best overall
foraging conditions for great tits differed according to the
prevailing weather conditions. In cold, late springs, pairs
with territories with a relatively low mean canopy height
reared the heaviest chicks, whereas in warm, early springs,
the heaviest chicks were produced in territories with a taller
canopy. In average springs, canopy height had little influence on chick mass. Weather conditions can directly affect
both the birds and their caterpillar prey. For example, foraging
in the top canopy will be more difficult for birds exposed to
cold, wet, and windy conditions, and caterpillars are likely
to descend or be washed down to lower levels. In good
weather, maximum prey productivity may be concentrated
in the outer canopy, and although this may not be ideal
foraging substrate for great tits (Lack, 1971; Rytkönen and
Krams, 2003), any disadvantage may be offset by the combination of good conditions and high prey abundance.
That habitat suitability should change according to
prevailing weather conditions is not unusual on a larger
scale. For example, cold weather movements in waders
and wildfowl are well known (e.g., Townsend, 1982) and
productivity of various species of sea and land birds varies
with large-scale shifts in weather patterns such as the El
Niño/Southern Oscillation (ENSO) (e.g., Schreiber and Schreiber,
1989; Nott et al., 2002). On a smaller scale, the breeding
success of tits in sub-optimal habitat such as small woods
and farmland may be more vulnerable to severe weather
than that of birds in larger woods due to greater exposure of
both the birds and their prey (Riddington and Gosler, 1995;
Hinsley et al., 1999). In nuthatches (Sitta europaea), individual territory quality can vary according to annual variation
in beech and hazel nut crops (Matthysen, 1998), and such
effects are likely to be significant for other species. However,
evidence for subtle shifts in territory quality from year to
year in relation to woodland structure (other than that due
to directional processes such as succession and maturation)
appears to be sparse (Lõhmus, 2003).
In addition to the effects of local climate on habitat
selection, weather conditions also affected our ability to
detect an effect of structure on the birds’ breeding performance. That the strongest relationship was found under the
most difficult conditions in 2001 was not really surprising.
Under extreme conditions, differences in quality, of both
habitat and birds (Ferns and Hinsley, 2004), will have
most effect on breeding success and hence also be most
detectable. However, for five of the seven years studied, we
1404 D e c e m b e r 2 0 0 6
were unable to detect a difference in habitat quality using
the parameters of chick mass and canopy height. Although
extreme conditions are relatively rare, they may have a
disproportionate effect on bird survival and population
persistence. Therefore, the ability to identify what constitutes best quality habitat (for both breeding and survival)
under extreme conditions is vital, but given the time-scale of
many ecological studies, such years may be easily missed. It
was of interest that a particularly warm spring, as well as a
cold one, appeared to constitute more extreme conditions.
Drought conditions associated with warm springs may affect
the abundance and phenology of the birds’ caterpillar food
supplies due to an effect on vegetation development (Visser
et al., 1998). Such conditions may become more frequent in
the future if the current trend of increasing global temperatures continues (Huang et al., 2000).
The relationships between the slopes and correlation
coefficients of the mass/height relationships and the winter
NAO index indicated that effects could also operate on larger,
spatial and temporal scales. Many recent studies have
demonstrated links between the winter NAO index and
various bird population parameters, especially in relation to
timing of breeding (e.g., Crick et al., 1997; Forchhammer
et al., 1998; Brown et al., 1999; Nott et al., 2002; Sanz, 2002;
Sanz, 2003). However, these studies generally use the winter
NAO index for the winter immediately preceding the breeding
season(s) in question. For great tits in our study, a time
lag of one year gave significant results (Figure 4) whereas
there were no relationships with the immediately preceding
winter NAO index. In a study of breeding phenology and
climate in birds and amphibians, Forchhammer et al. (1998)
showed that the winter NAO index influenced annual breeding numbers of golden plovers (Pluvialis apricaria) in the
U.K., but that this relationship was lagged by two years and
was explained as an effect of density dependence on numbers. A lag in the relationship between local temperature
(and/or other local weather conditions), and the winter NAO
index seems a more likely explanation than density dependence for the one year time lag found here. Tit populations in
Monks Wood have been stable (great tit) or slightly increasing (blue tit) throughout the study period. Alternatively,
biotic responses to conditions in one year may have consequences which persist into the following year (Masaka and
Maguchi, 2001; Roy et al., 2001). It is also possible that these
relationships with the winter NAO index are an artifact of a
relatively short run of data.
The poor performance of the warmth sum in 2003 may
have been related to the timing of breeding. As in other
studies (e.g., Perrins, 1970; Hinsley et al., 1999), the timing
of breeding of the Monks Wood great tits differed between
years (Table 1), but our calculation of warmth sum was for a
fixed period based on average timing. Thus, in 2003, this
average time period appeared to provide a poor estimate of
the local conditions affecting breeding performance. In a
study of Soay sheep population dynamics, Hallet et al.
(2004) demonstrated how the NAO could out perform local
weather variables as a predictor of sheep mortality because
of complex variation in the timing and interaction of local
weather conditions in some years. Using the relationship
between the winter NAO index and warmth sum (Figure 3) to
calculate the warmth sum for 2003 gave a value of 660,
substantially lower than the value of 771 produced by the
standard method of calculation. After substituting 660 for
771 in the relationships between the parameters of the
mass/height relationships and warmth sum, the results for
2003 no longer appeared as outliers (Figure 2a and 2b). The
influence of local and larger-scale climate conditions on the
timing of breeding by great tits will be reported in more
detail elsewhere (Hinsley et al., unpublished data, 2004).
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
FL-08-05
11/13/06
6:57 PM
Page 1405
Conclusions
The concept of habitat quality is fundamental in the study
of ecology, but is difficult to define and quantify. For many
habitats, and especially woodland, structure is a key component of quality which can be measured at high resolution on
a landscape-scale by airborne lidar (Lefsky et al., 2002;
Bradbury et al., 2005). This ability of lidar to characterize
woodland canopy structure is a prerequisite for large area
habitat mapping (Hyde et al., 2005), e.g., mapping potential
habitat for Delmarva fox squirrels in Delaware State (U.S.A.)
using a habitat suitability model (Nelson et al., 2005). Our
previously reported work on assessing (Hinsley et al., 2002)
and mapping (Hill et al., 2004) habitat quality for tits using
airborne lidar data went a step further by developing actual,
rather than theoretical, models of habitat quality. We used
breeding success (i.e., a measure of fitness), rather than bird
presence or abundance, to identify high quality habitat, and
used lidar data to supply a habitat measure suitable as a
correlate of fitness.
The results presented here suggest that habitat quality
can vary according to local weather conditions, and that
local conditions may in turn be linked to larger scale, global
climate events. Under the current conditions of global
warming and the trend towards warmer springs in northern
Europe, on average, habitat selection in great tits should
favor tall, closed canopy woodland. However, the contrasting effects of weather conditions on habitat quality between
years will tend to preserve a broad spectrum of habitat
preferences in great tits, rather than promoting directional
selection for particular structures.
References
Balmer, D., and L. Milne, 2002. CES comes of age, BTO News, 239:
14–15.
Beebee, T.J.C., 1995. Amphibian breeding and climate, Nature, 374:
219–220.
Both, C., M.E. Visser, and N. Verboven, 1999. Density-dependent
recruitment rates in great tits: The importance of being heavier,
Proceedings of the Royal Society, Series B, 266:465–469.
Bourski, O.V., and W. Forstmeier, 2000. Does interspecific competition affect territorial distribution of birds? A long-term study on
Siberian Phylloscopus warblers, Oikos, 88:341–350.
Bradbury, R.B., R.A. Hill, D.C. Mason, S.A. Hinsley, J.D. Wilson,
H. Balzter, Q.A. Anderson, M.J. Whittingham, I.J. Davenport,
and P.E. Bellamy, 2005. Modelling relationships between birds
and vegetation structure using airborne lidar data: A review
with case studies from agricultural and woodland environments, Ibis, 147,443–452.
Broughton, R.K., S.A. Hinsley, P.E. Bellamy, Hill, R.A., and Rothery,
P., 2006. Marsh Tit territory structure in a British broadleaved
woodland, Ibis, in press.
Brown, J.L., S.H. Li, and N. Bhagabati, 1999. Long-term trend
toward earlier breeding in an American bird: a response to
global warming?, Proceedings of the National Academy of
Science, USA, 96:5565–5569.
Crick, H.P., C. Dudley, D.E. Glue, and D.L. Thomson, 1997. UK
birds are laying eggs earlier, Nature, 388:526.
Dias, P.C., 1996. Sources and sinks in population biology, Trends in
Ecology and Evolution, 11:326–340.
Ferns, P.N., and S.A. Hinsley, 2004. Immaculate tits: Head plumage
pattern as an indicator of quality in birds, Animal Behaviour,
67:261–272.
Forchhammer, M.C., E. Post, and N.C. Stenseth, 1998. Breeding
phenology and climate, Nature, 391:29–30.
Fretwell, S.D., and H.L. Lucas, 1970. On territorial behaviour and
other factors influencing habitat distribution in birds, Acta
Biotheoretica, 19:16–36.
Gaveau, D.L.A., and R.A. Hill, 2003. Quantifying canopy height
underestimation by laser pulse penetration in small-footprint
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
airborne laser scanning data, Canadian Journal of Remote
Sensing, 29:650–657.
Hallett, T.B., T. Coulson, J.G. Pilkington, T.H. Clutton-Brock, J.M.
Pemberton, and B.T. Grenfell, 2004. Why large-scale climate
indices seem to predict ecological processes better than local
weather, Nature, 430:71–75.
Hill, R.A., S.A. Hinsley, P.E. Bellamy, and H. Balzter, 2003. Ecological applications of airborne laser scanner data: Modelling
woodland bird habitats, Proceedings of Scandlaser: Scientific
Workshop on Airborne Laser Scanning of Forests, 03–04
September, Umeå, Sweden, unpaginated CD-ROM.
Hill, R.A., S.A. Hinsley, D.L.A. Gaveau, and P.E. Bellamy, 2004.
Predicting habitat quality for Great Tits (Parus major) with
airborne laser scanning data, International Journal of Remote
Sensing, Cover Article, 20:4851–4855.
Hill, R.A., and A.G. Thomson, 2005 Mapping woodland species
composition and structure using airborne spectral and lidar
data, International Journal of Remote Sensing, 17:3763–3779.
Hinsley, S.A., P. Rothery, and P.E. Bellamy, 1999. Influence of
woodland area on breeding success in Great Tits Parus major
and Blue Tits Parus caeruleus, Journal of Avian Biology, 30:
271–281.
Hinsley, S.A., R.A. Hill, D.L.A. Gaveau, and P.E. Bellamy, 2002,
Quantifying woodland structure and habitat quality for birds
using airborne laser scanning, Functional Ecology, 16:851–857.
Huang, S., H.N. Pollack, and P.Y. Shen, 2000. Temperature trends
over the past five centuries reconstructed from borehole temperatures, Nature, 403:756–758.
Hurrell, J.W., and H. van Loon, 1997. Decadal variations in climate
associated with the North Atlantic Oscillation, Climate Change,
36:301–326.
Hurrell, J.W., Y. Kushnir, and M. Visbeck, 2001. The North Atlantic
Oscillation, Science, 291:603–605.
Hurrell, J.W., 2005. Climate indices. Winter (Dec-Mar) station based
NAO index, NCAR, Climate Analysis Section, Climate and
Global Dynamics Division, http://www.cgd.ucar.edu/cas/
jhurrell/indices.data.html#naostatdjfm (last date accessed:
19 September 2006).
Hyde, P., R. Dubayah, B. Peterson, J.B. Blair, M. Hofton, C. Hunsaker,
R. Knox, and W. Walker, 2005. Mapping forest structure for
wildlife habitat analysis using waveform lidar: Validation
of montane ecosystems, Remote Sensing of Environment, 96:
427–437.
Lack, D., 1971. Ecological Isolation in Birds, Blackwell Scientific
Publications, Oxford and Edinburgh, 404 p.
Lefsky, M.A., W.B. Cohen, G.G. Parker, and D.J. Harding, 2002.
Lidar remote sensing for ecosystem studies, Bioscience, 52:
19–30.
Lõhmus. A., 2003. Are certain habitats better every year? A review
and case study on birds of prey, Ecography, 26:545–552.
Martin, T.E., 2001. Abiotic vs. biotic influences on habitat selection
of coexisting species: Climate change impacts?, Ecology, 82:
175–188.
Masaka, K., and S. Magushi, 2001. Modelling the masting behaviour
of Betula platyphylla var. japonica using the resource budget
model, Annals of Botany, 88:1049–1055.
Matthysen, E., 1990. Behavioural and ecological correlates of territory
quality in the Eurasian Nuthatch (Sitta europaea), Auk, 107:
86–95.
Matthysen, E., 1998. The Nuthatches, T & AD Poyser Ltd., London,
315 p.
McCleery, R.H., and C.M. Perrins, 1998. Scientific Correspondence,
. . . temperature and egg-laying trends, Nature, 391:30–31.
Monrós, J.S., E.J. Belda, and E. Barba, 2002. Post-fledging survival of
individual great tits: The effect of hatching date and fledging
mass, Oikos, 99:481–488.
Muller, K.L., J.A. Stamps, V.V. Krishan, and N.H. Willits, 1997. The
effects of conspecific attraction and habitat quality on habitat
selection in territorial birds (Troglodytes aedon), American
Naturalist, 150:650–66.
Murcia, C., 1995. Edge effects in fragmented forests: Implications for
conservation, Trends in Ecology and Evolution, 10:58–62.
D e c e m b e r 2 0 0 6 1405
FL-08-05
11/13/06
6:57 PM
Page 1406
Nelson, R., C. Keller, and M. Ratnaswamy, 2005. Locating and
estimating the extent of Delmarva fox squirrel habitat using an
airborne LiDAR profiler, Remote Sensing of Environment, 96:
292–301.
Newton, I., 1991. Habitat variation and population regulation in
Sparrowhawks, Ibis, Supplement 1, 133:76–88.
Nott, M.P., D.F. Desante, R.B. Siegel, and P. Pyle, 2002. Influences
of the El Niño/Southern Oscillation and the North Atlantic
Oscillation on avian productivity in forests of the Pacific
Northwest of North America, Global Ecology and Biogeography,
11:333–342.
Pasinelli, G., 2001. Breeding performance of the Middle Spotted
Woodpecker Dendrocopos medius in relation to weather and
territory quality, Ardea, 89:353–361.
Patenaude, G., R.A. Hill, R. Milne, D.L.A Gaveau, B.B.J, Briggs, and
T.P. Dawson, 2004. Quantifying forest above ground carbon
content using LiDAR remote sensing, Remote Sensing of Environment, 93,368–380.
Perrins, C.M., 1970. The timing of birds’ breeding season, Ibis,
112:242–255.
Perrins, C.M., 1979. British Tits, Collins, London, 304 p.
Perrins, C.M., and R.H. McCleery, 1989. Laying dates and clutch
size in the Great Tit, Wilson Bulletin, 101:236–253.
Perrins, C.M., and R.H. McCleery, 2001. The effect of fledgling
mass on the lives of great tits Parus major, Ardea, 89:
135–142.
Pulliam, H.R., and B.J. Danielson, 1991. Sources, sinks and habitat
selection: A landscape perspective on population dynamics,
American Naturalist, Supplement 137:S50-S66.
Rathcke, B.J., 2000. Hurricane causes resource and pollination
limitation of fruit set in a bird-pollinated shrub, Ecology, 81:
1951–1958.
Riddington, R., and A.G. Gosler, 1995. Differences in reproductive
success and parental qualities between habitats in the Great Tit
Parus major, Ibis, 137:371–378.
Roy, D.B., P. Rothery, D. Moss, E. Pollard, and J.A. Thomas, 2001.
Butterfly numbers and weather: Predicting trends in abundance
1406 D e c e m b e r 2 0 0 6
and the future effects of climate change, Journal of Animal
Ecology, 70:201–217.
Rytkönen, S., and I. Krams, 2003. Does foraging behaviour explain
the poor breeding success of great tits Parus major in northern
Europe?, Journal of Avian Biology, 34:288–297.
Sanz, J.J., 1996. Environmental restrictions on reproduction in the
Pied Flycatcher Ficedula hypoleuca, Ardea, 83:421–430.
Sanz, J.J., 2002. Climate change and breeding parameters of great
and blue tits throughout the western Palearctic, Global Change
Biology, 8:409–422.
Sanz, J.J., 2003. Large-scale effect of climate change on breeding
parameters of pied flycatchers in Western Europe, Ecography,
26:45–50.
Schreiber, E.A., and R.W. Schreiber, 1989. Insights into seabird
ecology from a global ‘natural experiment,’ National Geographic
Research, 5:64–81.
Siikamäki, P., 1995. Determinants of clutch size and reproductive
success in the Pied Flycatcher, Biological Research Report 41,
University of Jyväskylä, Finland, 35 p.
Slagsvold, T,. 1976. Annual and geographical variation in the time of
breeding of the Great Tit Parus major and the Pied Flycatcher
Ficedula hypoleuca in relation to environmental phenology and
spring temperature, Ornis Scandinavica, 7: 127–145.
Steele, R.C., and R.C. Welch, 1973. Monks Wood. A Nature Reserve
Record, The Nature Conservancy, Huntingdon, 337 p.
Thomas, D.W., J. Blondel, P. Perret, M.M. Lambrechts, and J.R.
Speakman, 2001. Energetic and fitness costs of mismatching
resource supply and demand in seasonally breeding birds,
Science, 291:2598–2600.
Townsend, D.J., 1982. The Lazarus syndrome in Grey Plovers,
Wader Study Bulletin, 34:11–12.
van Balen, J.H., 1973. A comparative study of the breeding ecology of
the Great Tit (Parus major) in different habitats, Ardea, 61:1–93.
Visser, M.E., A.J. van Noordwijk, J.M. Tinbergen, and C.M. Lessells,
1998. Warmer springs lead to mistimed reproduction in great
tits (Parus major), Proceedings of the Royal Society, Series B,
265:1867–1870.
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING