Activity profiles, habitat selection and seasonality of body weight in a population of Arabian Spiny-tailed Lizards (Uromastyx aegyptia microlepis Blanford, 1875; Sauria: Agamidae) in Saudi Arabia

2009·1.8 MB·

by  Wilms, Thomas M

© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
BonnerzoologischeBeiträge Band 56 Ileft4 Seiten 259-272 Bonn, November 2009
Activity profiles, habitat selection and seasonality of body weight
in a population ofArabian Spiny-tailed Lizards
{Uromastyx aegyptia microlepis Blanford, 1875; Sauria: Agamidae)
in Saudi Arabia
Thomas M. Wilms'- Philipp Wagner-, Mohammed Shobrak-'' & Wolfgang Bohme-
' Zoologischer Garten Frankfurt, Bemhard-Grzimek-Allee 1, D-60316 Frankfurt am Main, Germany;
E-Mail: [email protected].
-Zoologisches ForschungsmuseumA. Koenig, Adenauerallee 160, D-53113 Bonn, Gennany.
^Biology Department, Science College, TaifUniversity, P.O. Box 888, Taif, Saudi Arabia.
CoH'esponding author.
Abstract. A field study was carried out on the Arabian Spiny-tailed Lizard (Umiiiasn:\ aegyptia wicmlepis) in Saudi
Arabia (Mahazat as-Sayd ProtectedArea) focusing on seasonal differences in activity, body condition and on parame-
ters influencing selection ofburrow sites in this large desert-dwelling lizard. Uromastyxa. microlepis is highly season-
alinrespecttoactivity/visibilityoutsideofitsbuiTowswhichprovideshelteragainstunfavorableclimaticconditions.At
Mahazatas-Saydtheselizardse.xhibitabnnodalactivityinspringandsummerwhileactivityinautumn isunimodalwith
apeak at early afternoon, butthe overall activity is generally low at this time ofthe year More than 73 % oftotal ob-
servedyearly activity takes place in spring and early summer Seasonal changes in availability offood result in differ-
ences ofthe animals'body condition, with a significant decrease ofbody-mass between spring/ autumn and summer/
autumnrespectively. Selectionofburrow sites is largely intluencedby soil typeandvegetation coverage.
KeyWords. Reptilia; Sauria;Agamidae; Uromastyx; Uromastyxaegyptia microlepis; seasonality; habitat selection; ac-
tivity; bodycondition; Mahazatas-SaydProtectedArea; SaudiArabia.
Introduction
The spiny-tailed lizards ofthe genus Uromastyx are in- Blanford, 1875 and Uromastyx a. leptieni Wilms &
habitants ofthe deserts andsemi-deserts ofNorthAfrica, Böhme, 2001 are distinguised (Wilms et al. 2009). The
ArabiaandtheMiddleEast. Currently 15 speciesarecon- present study focuses on theArabian Spiny-tailed Lizard
sideredtobe valid, ofwhich 6 are known to occuron the (U. a. microlepis) which lives in deserts and semideserts
Arabian Peninsula (Wilms et al. 2009). There is a con- ofArabia(SaudiArabia,Yemen,Oman, UnitedArabEmi-
siderable scientific interest regarding the taxonomy and rates, Qatar, Kuwait), in Jordan, Syria, Iraq and coastal
phylogenyofthese animalsbut ecological knowledge re- Iran. Habitats show a marked seasonality regarding cli-
mains scarce and fragmentarywithmost studies focusing maticparameters(temperature,airhumidity, precipitation)
on Uromastyx nigriventris and Uromastyx aegyptia aswellas regardingavailabilityoffood. In contrasttothe
(GRENOT& LoiRAT 1973; Grenot 1976; Bousicila 1986; widerangeofdailyandannual fluctuationsofairandsoil
Vernet et al. 1988; Cunningham 2000, 2001; Al-Haz- temperatures,thetemperatureswithin Uromastyxburrows
Mi 2002; Al-Johany 2003; Al-Hazmi et al. 2005). This areremarkablyconstant(Wilms& Böhme2007& unpub-
isespeciallyinteresting,consideringtheecological impor- lished data). Therefore these bun'ows are considered to
tance, prevalence and local abundance ofthese animals. playanimportantroleasrefuges inrespecttothenno-and
hydroregulationandas shelteragainstpredators. Burrows
Uromastyxaegyptia (Forsskál, 1775) is by farthe largest of Uromastyx aegyptia can be up to 1025 cm long and
member ofthe genus, reaching a maximum body length reach about 180 cm deep in the ground (Bouskila 1983,
mm
ofeven more than 700 and a weight up to 2.500 g 1984), but are, at least at Mahazat as-Sayd, normally not
(Fig. 1). Taxonomically, three subspecies, Uromastyx a. longerthanabout 300-530 cmand 80-120cmdeep(own
aeg}-ptia (Forsskál, 1775), Uromastyx a. microlepis data).
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
260 Thomas M. Wilms et al.: Arabian Spiny-tailedLizards in SaudiArabia
Fig. 1. Adull UiDDuistyxiicg\¡>lia micmlepis at Mahazat as-Sayd, Saudi Arabia. Photo: T. Wil.\is.
TheprimaryproductionofUromashahabitats isquitelow is SaudiArabia'sonlycompletelyfencedwildlifereserve
andexhibitsstrongseasonal andannualdifferences.There- andisareintroduction siteforHoubaraBustards{Chlamy-
fore it is somewhat surprising that Uromastyx aegyptia, dotis imdulata macqueenii), Arabian Oryx {Oryx leuco-
as all otherspecies ofthe genus, isprimarily herbivorous }yx) and Sand Gazelles (Gazellasubgiitturosa). Mahazat
(Mandaville 1965, Kevork & Al-Uthman 1972, Al- as-Sayd covers an area of2244 km- and lies within one
Ogily& HussAiN 1983, Bousicila 1984 & 1987, Robin- ofthe hottest regions ofthe world (Meigs 1953). It is a
son 1995). Nevertheless remains ofbeetles (Tenebrion- hot and semi-arid to arid desert steppe habitat, typical of
idae, Carabidae), ants, grasshoppers and even scorpions the central plateau ofthe Arabian Peninsula. The eleva-
are also found in the stomach contents as well as in fecal tion is between 900-1050 m above sea level (Lenain et
pellets ofall age classes (Kevork & Al-Uthman 1972, al. 2004). The terrain ofthis area consisted offlat gravel
Wilms 2007, pers. obs.). However, the proportion of plains, known as regs, occasionally interdigitated by dry
%
animal matter in the food is very low and was estimated sandy wadis (Fig. 2). Over 95 ofthe area is covered
%
to be only about 1-2 oftotal food intake. bysandandgravel.AirtemperatureinMahazat,asrecord-
ed in a standard weather shelter, often exceed43 °C dur-
Thepresentpaperanalyzestheseasonality ofactivityand ingthe summer, andoccasionallyreaches 50 °C (Seddon
body condition in this species which is induced by the 1996, Shobrak 1996,Williamsetal. 1999). Rainfall av-
aforementionedenvironmentalvariability.Asecondfocus erages ca. 100 mm per year (Williams et al. 1999) and
wasputontheselectionofhabitatstnicturesby Uromastyx typically occurs between March and May each year, but
a. miciolepis. This study was cairied out on free ranging withoccasional importantrain eventsatothertimes.There
populations ofUromastyxaegyptiamicrolepis at Mahaz- isnopermanentsourceofwaterabovegroundlevel inMa-
at as-Sayd Protected Area, Saudi Arabia. hazat as-Sayd, but ephemeral pools exist for short peri-
ods afterheavy rain (Lenain et al. 2004).
Materialand Methods The main study site was situated just south-east of the
main gate to Mahazatas-Sayd(Al-Muwayh gate), cover-
StudySite ing an area ofapprox. 4 km- and was primarily used to
assess habitat selection, while a second area nearby was
All fieldworkwascarriedoutwithintheboundariesofMa- chosen to establish a 60 km long transect forthe evalua-
hazat as-Sayd Protected Area, Saudi Arabia, which has tion ofactivity profiles.
been protected since 1989. Mahazat as-Sayd (22° 15' N,
41° 50'E) is located in central-western SaudiArabia, ap- The flat gravel plains in this part ofthe Arabian Desert
proximately 170kilometersnorth-eastofTaif Thereserve are intersected by wadis and dominated by sparse vege-
2
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
Bonnerzoologische Beiträge 56 261
tation ofperennial grasses, including Stipagrostis spec, Presence ofspecimens (time), size class [juveniles (total
Panicum turgidumandLasiuntsscindicus,andsmalltrees, length < 200 mm), subadults (total length 200^00 mm),
mavnXyAcacia spec. (Mandeville 1990). adults(total length>400mm)], distanceofspecimenfrom
respective burrow, animal behaviour (e.g. basking, shad-
Thisstudytookplaceinspring, summerandautumn2006 ing, feeding), coloration (dark grey, bright yellow) and
as well as spring and summer 2007. habitat (e.g. soil type, vegetation). To estimate the ther-
malenviromnentoftheobservedspecimensphysicalmod-
els were used to measure operative temperatures (T^) in
Methods appliedto assess activityprofiles different microhabitats in summer 2006 and spring 2007
(Hertzetal. 1993, Shine&Kearney2001). Hollowcop-
mm mm mm
Toassessthedifferencesinseasonalactivityofthe Spiny- percylinders(230 long, 78 diameter,0.6 wall
tailedLizardsatransectwasdefinedwithinan Uromastyx thickness) were used as models. Miniature data loggers
colony. The length ofthis transectwas 60 km and it was (i-buttons, Dallas Semiconductors, Model 1921) were
WD
drivenwitha4 vehicleeighttimesperseason(equals placedinthemodel lumentorecordT^. Dataloggerswere
480 km each) with an average speed of40 km/h. These calibrated inawaterbathagainstamercury-in-glass ther-
transects were driven only on sunny and calm days. As a mometer that had a calibradon traceable to the National
resultofthisprotocolthecoveredtimeperseasonwas 1 Institutes ofStandards andTechnology. To mimic the re-
hours (7 am-7 pm). The transect was driven in the fol- flectance ofa cold (grey colored) Uromastyx the model
lowingseasons: Summer2006(28^^'July-04'hAugust),au- wascoveredwith greyducttape. Temperaturesmeasured
tumn 2006 (26th October-17th November), spring 2007 with the tape covered copper model followed the body
(30'h March-06th April) and early summer 2007 (27'h temperaturesofadead Uromastyxbetterthantemperatures
June-Olst July). measured with the uncovered model. Duct tape covered
models were calibrated against dead as well as live Uro-
The followingdata were registeredbytwopersons (each mastyx under the original climate in the field (data will
concentrating on the area on either side ofthe transect): be published elsewhere).
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
262 Thomas M. Wilms et al.:Arabian Spiny-tailed Lizards in SaudiArabia
Fig. 3. UromastyxbuiTow marked with a tlag at the study site in Mahazat as-Sayd. Photo: T. Wilms.
Methods applied to assess habitatselection was caiTiedout in spring 2006 (08"^May-2"'^June), sum-
mer2006 15'hJuly-01stAugust), autumn2006 (20*Oc-
(
The main study site was in spring 2006 systematically tober-11'h November.) and spring2007 (26* March-15*
scanned for Uromastyx burrows by driving parallel tran- April). Thistrappingmethodhastheadvantagethatacap-
sects(distancesbetweentransectsca. 50 meter). Eachac- tured animal is still able to retreat into the burrow and
tive buiTow was marked with a flag (Fig. 3) and the fol- therefore the risk ofoverheating and loosing the animal
lowingparameterswereregistered: Orientationofburrow is minimized. Captured Uromastyx were marked with a
entrance (N, NE, E, SE, S, SW, W, NW), soil type (fine passive integratedtransponder(PIT),weighted,measured
sand, coarse sand, clay sand, sand/gravel mixture, fine (total length, tail length, head length and width) and sev-
gravel, medium gravel, coarse gravel, rock/sandmixture, eral scale counts were taken. Some specimens were im-
rock), slope, vegetation coverage (0 %, < 5 %, < 15 %, planted with temperature data loggers (i-buttons) and
<25 %),distancetonextAcacia.Atotalof206active Uro- equipped with radio transmitters for a separate study on
mastyxburrowswereregistered. Burrowswereconsidered the thennobiology ofUromastyxa. microlepis (Wilms et
asactive (inhabitedorvisited) eitherifan Uromastyxwas al. in prep.). The specimens were subsequently released
observed directly or iffresh tracks were found near the at their original buiTow.
opened and clean bunow entrances. Otherwise, the bur-
row was classified as non-active (abandoned). To assess seasonalityofbody condition, abody-mass-in-
dex (bmi) was calculated for each animal using the fol-
Toassesshabitat selection 80bunowsoutofthe206reg- lowingquotient: bmi= weight(g)/totallength (cm). On-
isteredoneswere selectedrandomly andtested against80 ly data ofadult specimens (total length > 400 mm) were
additional random localities within the boundaries ofthe included in the analysis (n=62).
study site, for which the same parameters as for the ac-
tive burrows were collected. y2 analyses wereperfonned
to assess differences between buiTOw locations and ran- Statisticalanalyses ofecologicaldata
domly placed locations.
TheExcel2000and SPSS (15.0) statisticalpackageswere
used to run the analyses, x- analyses have been selected
Methods appliedto assessseasonalit}' ofbody weight to evaluate activity data as well as data on habitat selec-
tion. Touncoverdifferences inbody-mass-indicesMann-
At the main study site Uromastyx were trapped using Whitney CZ-testand Kruskal-Wallisstatisticaltestwereap-
snares put into the entrance ofactive bun^ows. Trapping plied.
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
Bonnerzoologische Beiträge 56 263
100
90
80
70
60
50
40
TTTi i 1 1 1 1 - , I I^T 3200
s s s s s 10
O
rrnii--,ii,,
_
iiilliiiiiiiiiiililiiil
c. 100
201 90
80
70
10- 60
50
lili.. 40
30 i
S S S 8 20 -en
§ i i i 10
O
d. 100
90
80
15 70
60
10
50
5 40
1 m 1 30
0 s g : s í S S 8 3 S S 8 8 20
i i i i ! 1 i ^. í 8 8 8 ¿ 1O0
Fig. 4. Visibility profiles of Uromastyx aeg\>ptia microlepis Fig. 5. Distributionofdifferentbehavioralactivitiesasabso-
atMahazatas-Saydatdifferentseasons;a.Spring;b.EarlySum- lute numbers (black bars) and as percentage oftotal observati-
mer; c. Summer; d. Autumn. ons (grey bars) at different seasons; a. Spring; b. Early Sum-
mer; c. Summer; d. Autumn.
Results Behavior: 1.Baskingnearbun^owentrance;2.Animal faraway
fromburrowentrance;3.Shadingunder/icac/fltree;4.Feeding;
5. Basking on a bush; 6. Basking on a stone.
Activitx'structures
Atotal of245 Uromastyxwere sighted while driving the Profiles ofdaily visibility were produced for every sea-
transects. Theabsolutenumbers fortherespective seasons son by adding up the specimens observed within respec-
are: 97 (spring), 82 (earlysummer), 52 (summer), 14(au- tive intervals of30 minutes (Fig. 4). The diagrams show,
% %
tumn). This is equivalent to 39.59 (spring), 33.46 that U. a. microlepis exhibits a bimodal rhythm in spring
(early summer), 21.22 % (summer) and 5.71 % (autumn) andearlysummerwithtworelativelyhighvisibilitypeaks
oftotal observed presence above ground. There is a very in spring (at morning from 10:30-1 1:00 his respectively
restricted activity during winter (deduced from own data afternoonfrom 17:00-17:30hrs)butonlyonepronounced
on body temperatures, data not shown; Bouskila 1986, peak in early summer (at morning from 8:30-9:00 hrs),
Al-Hazmi etal. 2005).Theseobservationsshowthat Uro- followedbyagreatlyreducedbutneverthelessclearlyvis-
mastyxa. microlepis ishighlyseasonalatMahazatas-Sayd ible peak at early afternoon (13:30-14:00 hrs). The pro-
(X2 = 65.74, P < 0.01, df= 3), with a maximum visibili- files for spring and early summer exhibit a period with a
ty from spring to early summer. significantly reduced visibility during the hottest time of
.
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
264 Thomas M. Wilms et al.: Arabian Spiny-tailedLizards in SaudiArabia
a. a.
100
SubadJts&Miits Adits
90
ml 60
«
200 — - ^1 . J1 1 . . _ 1
SHgis!s S S 8 g S 3 S s
líiííílíií
i ^ 1 Í 1 ^ i é i
b. b.
AlUts SubaUbSAUts
I I s_g^s1gs- gB^sJIg.ls^sJIsnggn.
? •i 8 3 g 3 g
é i i é i i i i i i é ú i i í I i i I í í
S 3 3 S
3 i I i S 5 i 5 ^ ^ ^ ^ á ¿ ä
Fig. 6. Profiles showing the number of visible specimens Fig. 7. Profiles showing the number of visible specimens
(black bars) as well as average distances (in meters) ofall ob- (black bars) as well as the average distances (in meters) from
servedspecimens within eachrespective interval of30minutes the burrows (grey bars). Vertical bars indicate size class ofac-
(greybars);a.Spring; b. EarlySummer;c.Summer;d.Autumn. tivespecimens, a. Spring; b.Early Summer;c. Summer; d.Au-
tumn.
theday(12:30-14:00 hrsin spring; 12:00-13:30hrsinear- visibility profile for autuinn is unimodal and shows an
ly summer). The profile for the summer season shows overall veryrestricted level ofvisibilitywithamaximum
principallyalsoabimodal visibilityrhythmwith thehigh- occuiring between 12:00 hrs and 15:00 hrs, then the
est visibility between 8:30 hrs and 9:00 hrs followed by warmest time ofthe day.
a rapid decrease ofvisibility. After 12:00 hrs specimens
were seen only occasionally (two specimens between These visibility profiles were produced regardless ofthe
13:00-13:30 hrs, three specimens between 15:00-15:30 type ofobserved activity, thus counting a visible speci-
hrs, onespecimenat 17:26hrsandonespecimenat 17:34 men as being active. Figure 5 shows the distribution of
hrs); but nevertheless some specimens were still visible. the different activities like basking, shading, feeding and
The comparison ofthe profiles froin spring, early suin- specimensbeingfaroutsideoftheirrespectiveburrowfor
mer and summer show an obvious trend ofreduction of all four observation periods as absolute numbers (black
afternoon visibility and a shift ofthe first visibility peak bars) as well as percentages thereof(greybars). Calcula-
from lateinomingtoearlyinoming,which ispossiblyre- tionofX- statisticsshowedthattheobservedseasonaldif-
lated to the increase ofmaximuin day temperature. The ferences are significant for the categories "basking near
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
Bonnerzoologische Beiträge 56 265
burrowentrance"{yi=57.451, P<O.Ol, df= 5) and"an- tivityofjuveniles and subadults andonly fiveadultspec-
imal faroutside ofits burrow" (x2 = 24.000, P< 0.01, df imens have been seen active during the period between
= 5), while differences in the other four categories were 10:12-11:40 hrs and at 13:08 hrs and 13:25 hrs (Fig. 6c
not significant. As a consequence, diagrams showing the & 7c). Inautumnvisibilityandactivityisgreatly reduced
numberofvisible animals aswell as the average distance duetothe lowtemperatures.Activitywasobserved forall
ofspecimens from their burrows during each 30 minute size classes only during early afternoon (Fig. 6d & 7d).
interval were created (Fig. 6).
Temperatures in spring are relatively mild, resulting in a Habitatselection
prolongedtimethe lizardshavetospendbaskingnearthe
burrow entrance to reach their preferred body tempera- Ofthe fiveregisteredparametersatburrow locationsthree
tures. After having reached the preferred body tempera- proved to show significant deviations between observed
ture some animals start leavingthe directvicinityoftheir and expected frequencies. These are: orientation ofbur-
ownburrow. From 12:30-14:00hrspresenceofspecimens row entrances (P < 0.01), soil type (P < 0.01), and vege-
is greatlyreduced, butthose animals still present exhibit tation coverage (P < 0.01). The slope ofthe area around
a high degree ofactivity indicated by the long distance burrow entrances as well as the distances ofbun^ow en-
jfromtheirrespectiveburrow(Fig. 6a). Evaluatingtheac- trances to next Acacia were not significantly different
tivity data in respect to animal size revealed, that from comparedto circumstances atrandom locations (datanot
12:07 hrs until 14:41 hrs only large adults were seen far shown).
outsideoftheirburrows(Fig. 7a). Obviouslytemperatures
duringmiddayare even in springhigh enoughto forcethe Orientation ofburrow entrances: X- analysis revealed a
juveniles and subadults under ground, seeking shelter in significant deviation ofthe observed distribution ofbur-
their burrows as thermal refuges and therefore accounts rowentrance directions fromthe expecteddistributionof
fortherelatively lownumberofactive specimens. In ear- burrow entrance directions (X- = 30.27; P < 0.01; df=7).
ly summer average as well as absolute temperatures are The highest deviation from the expected value exists for
higherthanduringspring,thusthebaskingperiodnearthe buiTow entrances directed to south-west and south-east
burrow entrance on early morning is shorter; thus result- (Fig. 8).
inginanearlierstartof"farfromtheburrowactivity"(Fig.
6b). Obviously temperatures at morning are already too Selectedsoiltype atburrow location: Nine different soil
hightoallowactivityofjuvenilesandsubadultsawayfrom types were found at the 80 randomly selected locations
their burrows because of the risk of overheating. From within the study area (fine sand, coarse sand, clay sand,
8:28hrsuntil 14:00hrsonlyadultshavebeenseenactive. sand/gravel mixture, fine gravel, medium gravel, coarse
After 14:12 hrs also subadults started "far from the bur- gravel, rock/sand mixture, rock). X- analysis ofobserved
rowactivity"(Fig. 7b). InsummervisibilityofUromastyx frequenciesofselected soiltypes atburrow locations and
wasgenerallygreatlyreduced. Therewasnoobservedac- ofsoiltypesatrandomlyplaced locationswithinthe study
NE SE SM W NW
IObservedfrequency Expectedfrequency
Fig. 8. Observed andexpected frequencies ofUromastyxburrow entrance directions at Mahazat as-Sayd.
;
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
266 Thomas M. Wilms et al.: Arabian Spiny-tailed Lizards in SaudiArabia
Fig. 9. Selected soil types at Uwmastyxbunow locations
a. coarse sand; b. fine gravel; c. medium gravel.
arearevealed significantdifferences(yi=40.30; P<0.0
1
df= 8). All but one ofthe randomly selected Uromastyw
burrows were located at places with medium gravel, fine
gravel andcoarse sand(Fig. 9). Theonlyexceptional bur-
row was found between black rocks. Occasionally the
original soiltypeinthevicinityofUwiuasñ'xbun^owswas
coveredby athin layerofdrifting sand, which didnot in-
terferewiththephysical propeitiesoftheunderlayingsoil.
In such cases only the main substrate was considered for
the statistical analysis. Sandy soil was therefore defmed
as loose deep sand. Fig. 10. Vegetationcoverageestimatedwitiiin aradiusoffne
meters aroundentrances; a.0%; b.<5%; c.< 15"'o; d.<25%.
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
Bonnerzoologische Beiträge 56 267
Selected vegetation coverage at burrow location: Vege- sex oftherespective specimens (P<0,189. Fig. 11 b). To
tationcoveragewas estimatedwithinaradius offiveme- check ifbetween seasons bmi differences arc statistical-
ters aroundthe burrow entrances respectively around the lysignificant,dataforall fourseasons werepairwisetest-
randomly selected locations (0%, < 5%, < 15%, < 25%; edusingMann-Whitney iy-test. Thesetestsshow, thatdif-
Fig. 10). yi analysis revealed a significant difference be- ferencesbetweenspring2006andautumn2006werehigh-
tweenvegetationcoverage frequencyatburrow locations ly significant (P< 0.001) as well as forsummer2006 and
and at randomly placed locations {yi = 49.40; P < 0.01; autumn 2006 (P< 0.004). All other pairs showed no sig-
df= 3). The data show that Uromastyx avoid area with nificances.
0%coverage but show apreference forareas with < 5%
vegetation coverage. For places with < 15 % and < 25 % To furtherevaluatetherelationshipofbmibetween spring
vegetationcoveragenodifferencesbetweenobservedand 2006andautumn 2006respectivelysummer2006andau-
expected frequencies were evident. tumn2006 Mann-Whitney í7-testswereperformedonda-
taformalesand females separately.As result,differences
for males were significant between spring 2006 and au-
Seasonality ofbody weight tumn2006 (P<0.001)butnot forfemaleswhilebetween
summer 2006 and autumn 2006 differences were signif-
Morphometric data of62 adult specimens (total length > icant for females (P < 0.016) but not for males. After
400 mm) caught in spring 2006, summer 2006, autumn Fowler et al. (1998) Mann-Whitney í7-test may be used
2006 and spring 2007 were used to calculate body-mass- withvery low sampling units, but in thatcase there must
indices (bmi). Bmi in spring 2006 was 1.1961-3.2414 be no overlap ofobservations between the two samples
(2.4348+/0.6033), insummer2006itwas 1.1057-3.1269 to reject H,,. This precondition ofthe test was met by the
(2.2172+/-0.6146). inautumn2006itwas0.8454-2.2981 dataset for females with values for summer being
(1.4972+/-0.4257) and in spring 2007 it was 1.81-2.74andforautumn 1.08-1.43,while itwasnotmet
1.1872-2.8921 (1.9828 +/- 0.4901). Kruskal-Wallis sta- by the dataset for males with values for spring being
tistics were performedto test data for significances (Fig. 1.99-3.24and forautumn0.85-2.30. Thereforewereject
11),whichrevealedthatbmidifferssignificantlyin respect Hq only for the females' dataset.
to season (P < 0.001; Fig. 11 a) but not regarding to the
a.)Season: 1 =spring2006: 2=summer2006: 3=autumn2006:4= 2007 b.) Sex: 1 = male; 2 =female
season N Mean Rank sex N Mean Rank
bmi 1 12 44,08
bmi 1 36 34,06
2 10 37,40
2 26 27,96
3 12 15,67
Total 62
4 28 30,79
Total 62
Test Statistics(a,b)
TestStatistics(a,b)
bmi
bmi
Chi-Square 1,722
Chi-Square 16,193
df 3 df 1
Asymp. Sig. ,001 Asymp. Sig. ,189
a KruskalWallisTest
a Kruskal Wallis Test
b GroupingVariable: season
b Grouping Variable: sex
Fig. 11. Kruskal-Wallisteststatisticstestingbody-mass-indexdataforsignificanceregardingseason(Fig. 1la)andsex(Fig. 1lb).
© Biodiversity Heritage Library, http://www.biodiversitylibrary.org/; www.zoologicalbulletin.de; www.biologiezentrum.at
268 Thomas M. Wilms et al:Arabian Spiny-tailed Lizards in SaudiArabia
Discussion tionswerealsoconductedin earlyspring(February) show-
ing that the animals have only one daily activity period
Activit}'structures in this season (11.00-15.00 hrs). After Bouskila (1986)
theperiodwiththehighestactivityisfromMarchto Sep-
As has been shown in this study Uromastyxaegyptia mi- temberinapopulation ofUromastyxa. aegyptiainIsrael,
%
crolepis is highly seasonal, with this seasonality not be- whilefromOctoberto Februaryonly70 oftheobserved
ing restricted to behavioral means ofavoidance oftem- Uromastyxburrows showed signs ofactivity. The lowest
porarilyunsuitablemicroenvironmentsbutalsotochanges levels ofactivity were recorded from Decemberto Mid-
in the physical constitution of the animals (changes of February,wereall observationsofactivespecimenswere
bmi). The evaluation ofthemioregulation in this species made on clear and sunny days only.
is not the primaiy objective ofthe present paperand will
bediscussedelsewhere (Wilms etal. in prep.). Neverthe- In another study conducted by Al-Johany (2003) near
less, theobserved seasonality accounts to some degreeto Riyadh (Thumama; King Khalid Wildlife Research Cen-
interactions between the thermal environment and the ter)anunimodal activity structurewasobservedin spring
lizard and is therefore here briefly discussed. (May) starting between 9.00 and 10.00 hrs and ending
aroundonehourbefore sunset. Summeractivity(August)
Differences in temporal patterns of activity, the use of was foundto be bimodal withone activityperiodstarting
space, and body temperature relationships are not inde- 8.00-9.00 hrs until midday and another period from
pendent. Many lizardsbehaviorally regulatebodytemper- 15.00-16.00hrsuntil sunset(18.30hrs). Wesuggest,that
aturesby shuttlingbetween sun and shade orhotandcold thedifferentspringactivityprofilesestablishedinthepres-
microenvironments to alter heat flux, by modifying pos- ent study compared with the result ofAl-Johany (2003)
turetoaltersurfaceareasexposedtoheatsourcesorsinks, might be explained by the different methodologyusedto
and by regulating activity times (Huey 1974. Pianka assess activity. While in the present study a transect
]986). These arewell acceptedparadigms ofreptile ecol- method was applied Al-Johany (2003) observed the an-
ogyandespecially true forextreme habitats likethose in- imals from a hiding place within the study area, which
habited by the studied Uronuist}'x at Mahazat as-Sayd. probably lead to biased results by overestimating activi-
ty during midday by observing single actively roaming
The visibility profiles of Uromastyx at Mahazat as-Sayd specimens. As has been shown in the present study, few
show a bimodal structure in spring, early summer and to specimens are visible above ground during midday but
a lesser degree in summer combined with a general de- some of these show an extraordinary high activity by
creaseofvisibility from springto summer. Visibilitystnic- walking far from their respective burrow. All observed
turechangesfromabimodaltoanunimodal rhythm from specimensbeingactiveduringthehottestperiodoftheday
summer to autumn reflecting changes in the availability were adult. Adult Uromastyx a. microlepis have a body
ofsuitablethermal microhabitats. Thevisibilityprofile for mass ofup to 2.500 g and have therefore a higher ther-
autumn shows an overall very restricted level ofvisibili- mal inertia than smallerconspeciflcs. This property isre-
tywithamaximumoccurringbetween 12.00hrsand 15.00 ducingtheriskofoverheatingforlarge specimensandal-
hrs, then the hottest time ofthe day. lows them a prolonged activity phase under unsuitable
thermal conditions compared to smaller specimens.
Theseresultsareingoodaccordancewiththeobservations
ofAl-Hazmi et al. (2005) made in central Saudi Arabia Operative temperatures (T^,) for observation periods in
(Al-Gassim region), with two activity peaks in spring spring2007andsummer2006areavailablefromthreedif-
(April) and summer (July) (9.00-11.00 hrs and ferentmicrohabitats (sandyplain, gravel plain and40cm
15.00-17.00hrs inspring, 7.00-10.00hrsand 16.00-18.00 deep in the entrance ofan Uromastyx burrow). There is
hrs in summer) and only a single activity peak in autumn only a slight difference regarding T^^(average ofall days
(mid October) (11.00-13.00 hrs). In this study observa- with observations) between spring and summer with al-
Table 1. Operative temperatures ofthree different microhabitats in spring and summer in Mahazat as-Sayd (only temperatures
between 07.00 hrs and 19.00 hrs were used forcalculation).
(min,spring) (min,summer) (max,spring) (max,summer) averageTg(spring) averageTg(summer)
Sandy plain 27.8 °C 27.6 °C 54.4 °C 55.5 °C 45.3 °C 47.2 °C
Gravel plain 28.8 °C 28.4 °C 53.3 °C 54.8 °C 44.6 °C 46.4 °C
Burrowentrance 29.8 °C 35.1 °C 39.4 °C 38.1 °C 36.7 °C 37.2 °C