Embed Size (px)
Citation preview
Grazing by hill sheep andred deer prevents theregeneration of woodland
in many parts of the ScottishHighlands and has also led toextensive loss of heathercover1–3. Conservation bodiesclaim that there has been arapid rise in Highland deernumbers caused by inadequatemanagement and that theseneed to be drastically reduced4.Here we show that the recentincrease in red deer stocks hasprobably been overestimatedand suggest that the gradualrise in numbers since 1970 maybe a consequence of a reduc-tion in sheep stocks and ofchanges in winter weather,rather than of a reduction inculling rate. Although therewould be environmental ben-efits in reducing deer numbers,there is an equal need to reducethe numbers of hill sheep inmany parts of the Highlands.
In the British Isles, few envir-onmental issues have generatedsuch a long-running debate as the manage-ment of the Scottish Highlands, which havebeen grazed by a combination of red deer andsheep for much of the past two hundredyears1,5,6. During this period, heather coverhas diminished and the prevention of treeregeneration by browsing has led to the gradual erosion of many of the remainingnative woodlands2,3.
A recent report commissioned by theWWF (formerly the World Wildlife Fund)and the Royal Society for the Protection ofBirds (RSPB)4 claims that red deer (Cervuselaphus) in the Scottish Highlands increasedfrom 300,000 in 1989 to 450,000 by 2002(Fig. 1a). The report predicts dire environ-mental consequences unless deer numbersare drastically reduced. However, we believethat the concerns of conservation bodiesmay be exaggerated.
First, it is unlikely that Scottish red deernumbers are increasing as rapidly as thereport claims4. A large part of the apparentincrease is due to the arbitrary tripling of theestimated number of deer living in wood-land, where they cannot be counted reliably(from 30,000 in 1990 to 100,000 in 2002;Fig. 1a). When figures are restricted to deerliving on the open hill (where numbers canbe estimated more reliably) and are correctedfor the area and year in which counts weremade,our analysis indicates that the increase
in total deer numbers and average densitysince 1970 has probably been gradual(Fig.1b); also, it is unclear whether the rate ofgrowth has accelerated in recent years.More-over, the confidence interval for any realisticestimate of total hill deer numbers is verylarge: the report4 provides a single estimatefor 2002 without error range (354,000) thatdoes not fall far outside the upper confidencelimit of the total estimated in the 1980s(337,000; ref.6).
Second, it is unlikely that recent increasesin deer numbers have been caused by a relax-ation of management,because (despite somefluctuations between years) annual cullshave been substantially higher since 1985than they were before this (Fig. 1c). Instead,the gradual increase in hill deer numberssince 1970 may reflect continuing reductionsin the number of hill sheep and trends inweather conditions that have led to lesssevere winters and warmer summers, whichhave previously been shown to be associatedwith changes in hill deer numbers6,7. In fact,deer populations in the Highlands may havebeen resource-limited for some time:between 1960 and 1990, annual culls offemales in many parts of the Highlands werebelow the 15–20% level that is necessary toprevent an increase in population size8,9 andthere has been no lasting reduction in totaldeer numbers since then (Fig.1b).Moreover,
brief communications
NATURE | VOL 429 | 20 MAY 2004 | www.nature.com/nature 261
annual changes in population-growth rate (r) values since 1960are probably density-dependent,with r�0 at a total populationsize of 310,000–315,000 deerunder the current culling regime(Fig.1d).
Third, although even moder-ate deer densities can preventwoodland regeneration and damage forestry plantations2,3,9,10,there is limited evidence that deer populations cause extensivechanges in the distribution ofother upland ecosystems8, unlessnumbers are sustained at a highlevel by the provision of supple-mentary food in winter. In manyareas grazed by deer alone, com-munity boundaries appear to bestable9,11. Extensive heather loss(one of the principal concerns ofthe WWF and RSPB) might bemore frequently caused by sheep(which are often maintained atsummer densities that are severaltimes higher than deer) or by acombination of sheep and deer6,9,than by deer alone.
So what should be done to protect theenvironment of the Scottish Highlands? TheRSPB have shown that it is possible toencourage woodland regeneration andarrest heather loss by reducing grazing pres-sure to very low levels10, but reducing sheepand deer numbers to a comparable extentacross the Highlands would affect manyhuman communities. A more feasible solu-tion could be to reduce the combined stock-ing rate of sheep and deer to a level at whichheather loss is arrested and to protect sensi-tive areas of woodland with fences (despitetheir disadvantages) to allow regeneration tocontinue. Even this aim will be difficult toimplement as sustained control of extensivepopulations is seldom easy or cheap12,13.T. H. Clutton-Brock*, T. Coulson†,J. M. Milner‡*Department of Zoology, University of Cambridge,Cambridge CB2 3EJ, UKe-mail: [email protected]†Department of Biological Sciences, ImperialCollege at Silwood Park, Berkshire SL5 7PY, UK‡Hedmark University College, Evenstad,2480 Koppang, Norway
1. Mitchell, B., Staines, B. W. & Welch, D. Ecology of Red Deer: A
Research Review Relevant to their Management in Scotland
(Institute of Terrestrial Ecology, Cambridge, 1977).
2. Beaumont, D., Dugan, D., Evans, G. & Taylor, S. in Our
Pinewood Heritage (ed. Aldhous, J. R.) 73–95 (Forestry
Commission, Edinburgh, 1994).
3. Milne, J. A., Birch, C. P. D., Hester, A. J., Armstrong, H. &
Red deer stocks in the Highlands of ScotlandA drastic culling of deer may not be the best strategy to arrest erosion of heather cover.
1960 1970 1980 1990 2000
Dee
r num
bers
inth
e H
ighl
ands
1960 1970 1980 1990 2000
Hill
dee
r num
bers
in th
e H
ighl
ands
12 12.2 12.4 12.6 12.8 13
r
1960 1980 2000
Num
ber o
f dee
r cul
led
in th
e H
ighl
ands
02468101214
Fem
ales
cul
led
(%)
a b
c d
500,000
400,000
300,000
200,000
100,000
80,000
60,000
40,000
20,000
0
500,000
400,000
300,000
200,000
100,000
Year
Year Year
ln (Population size)
0.50.40.30.20.1
0–0.1–0.2–0.3–0.4
Figure 1 Estimates of changes in red deer populations in the Scottish Highlands since 1960.
a, WWF/RSPB estimates of total red deer in Scotland4, including (hollow circles) or excluding (filled
circles) an arbitrary increase in the estimated number of woodland deer by 70,000 in 2002 (from
ref. 4); b, total hill deer numbers estimated from a multiple regression model6 and corrected for year of
count, showing 95% confidence limits; c, estimates of annual numbers of adult females (hollow circles)
and all deer (filled circles) culled and of the annual culling rate (per cent) of adult females (triangles);
d, estimates of annual population-growth rate (r ) of hill deer, plotted against population size in the fol-
lowing year, using data shown in b. The plot indicates that population growth is density-dependent and
that r�0 at a population size of 310,000–315,000, given the existing culling regime.
20.5 brief comm MH 13/5/04 5:12 pm Page 261
© 2004 Nature Publishing Group
© 2004 Nature Publishing Group
Robinson, A. The Impact of Vertebrate Herbivores on the Natural
Vegetation of the Scottish Upland: A Review (Scottish Natural
Heritage, Edinburgh, 1998).
4. Hunt, J. F. Impacts of Wild Deer in Scotland: How Fares the Public
Interest? (Report for WWF Scotland and RSPB Scotland, 2003).
5. Hart-Davis, D. Monarch of the Glen (Cape, London, 1978).
6. Clutton-Brock, T. H. & Albon, S. D. Red Deer in the Highlands
(Blackwell Scientific, Oxford, 1989).
7. Clutton-Brock, T. H. & Albon, S. D. Nature 358, 11–12 (1992).
8. Clutton-Brock, T. H., Crawley, M. J. & Milner, J. M. in A Highland
Deer Herd and its Habitat 237–255 (Red Lion, London, 2002).
9. Milner, J. M., Alexander, J. & Griffin, C. in A Highland Deer Herd
and its Habitat 109–146 (Red Lion House, London, 2002).
10. Hester, A. J., Iason, G. R., Stolte, A. & Sim, D. Vegetation Survey:
Abernethy Forest RSPB Reserve (RSPB Reserve, Edinburgh, 1998).
11. Darling, F. F. A Herd of Red Deer (Oxford Univ. Press, 1937).
12. Caughley, G. in Rangelands: A Resource under Siege (eds Joss, P. J.,
Lynch, P. W. & Williams, O. B.) 135–140 (Australia Academy of
Science, 1986).
13. Caughley, G. in Problems in Management of Locally Abundant
Wild Animals (eds Jewell, P. A., Holt, S. & Hart, D.) 7–19
(Academic Press, New York, 1981)
Competing financial interests: declared (see online version of this
communication for details).
Vision
Steady-state misbindingof colour and motion
When you see a red ball rolling acrossthe floor, the ball’s redness, round-ness and motion appear to be
unified and inseparably bound together asfeatures of the ball. But neurophysiologicalevidence indicates that visual features suchas colour, shape and motion are processed inseparate regions of the brain1. Here wedescribe an illusion that exploits this separa-tion, causing colour and motion to berecombined incorrectly while a stable stimu-lus is being viewed continuously.
The illusion is seen in stimuli containingtwo sheets of random dots,where one sheet ismoving up and one is moving down. Thesheets contain dots of two colours such thatthe central and peripheral portions of thestimuli combine colour and motion in opposite fashions (Fig. 1a). On the upward-moving sheet, dots in the centre are red anddots in the periphery are green.On the down-ward-moving sheet, dots in the centre aregreen and dots in the periphery are red. (Formovie,see supplementary information.)
Observers gazing at the centre of the dis-play perceive peripheral dots erroneously:they ‘bind’ colour and motion in the wrongcombination. The entire display thereforeappears to be covered by a sheet of red dotsthat are moving upwards and a sheet of greendots that are moving downwards.
To quantify this effect, observers wereasked to report whether the majority ofperipheral red dots were moving up or downover a series of trials in which we varied thepercentage of peripheral dots moving ineither direction. In control trials, where thecentral portion of the stimulus was omitted,responses of observers (n�5; 3 naive)roughly followed the physical stimulus(Fig. 1b, black line). But when the central
portion was present, responses shiftedmarkedly, and in the same direction as thecentral dots (Fig. 1b, red and green lines;MANOVA,F(10,16)�35.2,P<0.001).
To ensure that observers’ responses werereflecting their perception and not a simplebias to respond to the central dots, we gener-ated stimuli in which one peripheral regionmatched the centre and the other containedthe opposite motion. Observers (n�4; 3naive; all from the first experiment) perceivedboth sides to move with the centre,performingpoorly in reporting which side moved theother way (mean�s.e.m.: 47�8% correct;sensitivity measure d'��0.12�0.38). Onthe other hand, they performed well in reporting which side was which when thestimulus centre was blank (86�2.5% correct;d'�1.5�0.2,paired t-test P<0.5).
We propose that the illusion described hereis the result of an ambiguity-resolving mecha-nism. Because vision is clearest at the point ofgaze,it would normally be advantageous to useinformation from the centre to resolve periph-eral ambiguity if the physical source is likely tobe a common uniform surface. Uniform sur-faces are strongly implied by the contiguousand precisely equal values of colour andmotion present across this stimulus.Peripher-al features are bound in such a way as to accordwith the usually more reliable centre,althoughin this case the binding is mistaken.
Previous investigations of basic visualfeature binding required the use of stimuliwith brief presentation times or rapidlychanging features to induce errors or ineffi-ciencies2–4. The illusion presented hereoccurs despite fully attentive and continuousviewing,avoiding the common confounds ofmemory, expectation and task strategy5. Itsolidifies the evidence for the existence of abinding problem and provides a substratefor neurophysiological investigation.Daw-An Wu*, Ryota Kanai†,Shinsuke Shimojo*‡§*Division of Biology and ‡Computation and NeuralSystems, California Institute of Technology, 139-74,Pasadena, California 91125, USAe-mail: [email protected]†Psychonomics Division, Helmholtz ResearchInstitute, Universiteit Utrecht, 3584 CS Utrecht,The Netherlands§Human and Information Science Laboratory,NTT Communication Science Laboratories,Atsugi, Kanagawa 243-0198, Japan1. Livingstone, M. & Hubel, D. Science 240, 740–749 (1988).
2. Treisman, A. M. & Gelade, G. Cogn. Psychol. 12, 97–136 (1980).
3. Moutoussis, K. & Zeki, S. Proc. R. Lond. B 264, 393–399 (1997).
4. Nishida, S. & Johnston, A. Curr. Biol. 12, 359–368 (2002).
5. Wolfe, J. M. & Cave, K. R. Neuron 24, 11–17 (1999).
Supplementary information accompanies this communication on
Nature’s website.
Competing financial interests: declared none.
Corrigendum
Insecticide resistance in mosquito vectorsM. Weill, G. Lutfalla, K. Mogensen, F. Chandre,A. Berthomieu, C. Berticat, N. Pasteur, A. Philips,P. Fort, M. RaymondNature 423, 136–137 (2003).It has been drawn to Nature’s attention that M. W., N. P.,P. F. and M. R. are named as inventors on two patentapplications relevant to this work (FR2002000762220020620, filed in June 2002; and FR20020021379920021105, filed in November 2002), which shouldtherefore have been declared as competing financialinterests.
brief communications arising online➧ www.nature.com/bca
Evolutionary biology: Lamprey Hox genes and theevolution of jawsY. Takio, M. Pasqualetti, S. Kuraku, S. Hirano, F. M. Rijli,S. Kuratani (doi:10.1038/nature02616)
brief communications
262 NATURE | VOL 429 | 20 MAY 2004 | www.nature.com/nature
Figure 1 Steady-state misbinding stimulus and observers' reports.
a, Sample transparent motion stimulus. Two sets of random dots
move rigidly in opposite directions. 'Real' circles show details of
the physical stimulus. In the upward-moving surface, dots near
fixation are red, and dots beyond 6º of horizontal eccentricity are
green; in the downward-moving surface, the reverse coloration
applies. White bars marked the transition line. 'Perceived' circles
show details of the illusory percept. Observers incorrectly pair
colour and motion in the periphery. Across the entire stimulus,
upward-moving dots appear red and downward-moving dots
appear green, forming two homogeneous surfaces of dots moving
in opposite directions. b, Proportion of trials in which observers
judge that most red dots beyond the white bars are moving
upwards, plotted as a function of the actual percentage of red dots
moving upwards; time to view stimuli was unlimited. In control
trials (black line), the central region is blank, and responses follow
the stimulus. When the central region contains upward-moving red
dots and downward-moving green dots (red line), responses shift
predominantly to 'up'. When the central region contains the oppo-
site motion (green line), responses shift predominantly to 'down'.
Real
Perceived
0 25 50 75 1000
25
50
75
100
Peripheral red dots moving upwards (%)
‘Up
’ res
pon
ses
(%)
a
b
20.5 brief comm MH 13/5/04 5:12 pm Page 262
© 2004 Nature Publishing Group
© 2004 Nature Publishing Group
