leeg
The effects of fluctuating microclimate
on the questing behaviour and
survival of Ixodes ricinus
Maris van der Baan
A bachelor thesis study for Van Hall Larenstein University of Applied Sciences, The Netherlands. BSc Forest and nature management, Tropical Forestry. Carried out at the research institute Alterra, Wageningen University and Research Centre
Name of graduate student: Maris van der Baan
This is a bachelor thesis study for: Van Hall Larenstein University of Applied Sciences, The Netherlands
BSc Forest and nature management, Tropical Forestry
Research institute: Alterra, Wageningen University and Research Centre
Supervisors: Dr. Ir. N.W. (Nico) van de Brink, Alterra - Wageningen University & Research Centre
Dr. Ir. PJ (Peter) van der Meer, Van Hall Larenstein, University of Applied Sciences
Date: 6 January 2014
Search terms: Tick, temperature, humidity
The effects of fluctuating microclimate
on the questing behaviour and
survival of Ixodes ricinus
ABSTRACT
Lyme borreliosis, commonly known as Lyme disease, is a vector-borne disease that is caused by the bacterium Borrelia burgdorferi sensu lato. Infected ticks can transmit this disease and infect humans. Decreasing human Lyme infections could be achieved by vegetation control , since mortality and questing behavior of ticks are related to relative humidity and temperature which may be regulated through vegetation control. The main objective of this study was to determine whether and how daily fluctuating temperatures, relative humidity and saturation deficit affect questing behaviour and mortality of the sheep tick Ixodes ricinus. In total, 96 ticks were collected and distributed over 12 arenas which contained 8 vantage points for questing activity. Relative humidity, temperature and saturation deficit were registered by loggers in each arena, and means, minima, maxima and the number of hours above the previously suggested thresholds of 7°C and 80%RH were calculated over the 24 hours prior to each observation. Four arenas were placed in a greenhouse with temperatures fluctuating between 3°C and 15°C and RH between 64% and 100%, another four were placed in a greenhouse with temperatures between 12°C and 21°C and RH between 57% and 96%, and another four were placed in a greenhouse with temperature between 16°C and 26°C and RH between 64% and 93%. Tick questing quantity and mortality was observed for four weeks. A logistic regression analysis showed that questing activity was positively related to temperature and the number of hours above 7°C and that mortality was negatively related to relative humidity, maximum temperature, and the number of hours above 80% RH, and positively related to mean and minimum saturation deficit. After closer examination of graphs it was suggested that questing is limited by cold temperatures, and that questing under warmer temperatures is less affected by temperature, which suggests that dynamics in questing behavior at higher temperatures may be explained by something else. However, this has not been tested. Some vegetation control measures have been proposed for forest management to reduce the amount of tick encounters by humans.
T
ABLE
OF
CONTENTS
Abstract...3
1. Introduction...7
2. Research question and Hypotheses...9
3. Literature review...10
3.1 Tick ecology...10
3.2 Relative humidity and saturation deficit...11
3.3 Temperature...12
3.4 Other factors influencing questing and survival...13
3.5 Microclimate change and vegetation control...13
4. Materials and methods...15
4.1 Tick collection...15
4.2 Arenas...15
4.3 Data collection...17
4.4 Data analysis...17
5. Results...19
5.1 Overview of the microclimates in the compartments...19
5.2 Questing...20
5.3 Mortality...24
6. Discussion...28
7. Recommendations on forest management...30
8. Conclusions...31
9. Acknowledgements...32
10. References...32
1. INTRODUCTION
Lyme borreliosis, commonly known as Lyme disease, is a vector-borne disease that is caused by the bacterium Borrelia burgdorferi sensu lato. The bacteria are transmitted by ticks that have been infected by previous hosts. Once a tick is infected, it can transmit the bacterium to next hosts, including birds, rodents, reptiles, small- and big mammals such as deer, but also human (Jaenson et
al., 1994). The bacterium has several subspecies, of which Borrelia afzelii and Borrelia garinii
account for most human infections in Europe. Other subspecies in Europe are Borrelia burgdorferi,
Borrelia spielammi, and Borrelia bavariensis, and in North America Borrelia burgdorferi is the only
active agent (Stanek et al., 2012). The symptoms of Lyme borreliosis vary among the different bacterium subspecies. Early symptoms include erythema migrans (a red circle around the tick bite), neurological symptoms and rarely cardiac symptoms. Crodermatitis chronica atrophicans (a skin disorder) and arthritis mostly manifest at later stages, although the symptoms do not always occur chronologically, and most patients do not suffer from all symptoms (Stanek et al., 2012). In Europe, the main vector that transmits Lyme borreliosis is Ixodes ricinus, in Asia I. persulcatus, in Northeastern and upper Midwestern America I. scapularis, and in Western America I. pacificus (Stanek et a.l, 2012).
A meta-analysis by Rauter and Hartung (2005), covering studies from all over Europe, showed that tick infections by Borrelia burgdorferi were found in almost all parts of Europe. In The Netherlands there has been an increase of tick bites and erythema migrans cases between 1993 and 2005, according to a study by Hofhuis et al. (2006). In 1993, 39 cases of erythema migrans were reported per 100.000 inhabitants, and 103 cases per 100.000 inhabitants in 2005. Although there is not yet a clear scientific explanation for this dramatic increase, possible explanations include increase in populations of infected ticks, changes in recreational behaviour, creation of ecological connections between nature areas, changes in host abundance and recent milder winters (Gassner, 2010). Considering the dramatic increase of Lyme infections in ticks, it is obviously important to continue with and even increase the amount of studies on causes for this increase and on the possibilities to decreasing human infection rates.
Decreasing human infections with Lyme could be achieved by prevention with repellents, but this is only short term and temporary. Vegetation control to control tick populations and behaviour seems to offer other options, since I. ricinus is mostly restricted to forests (or other vegetation cover) where the litter layer is sufficiently moist even during the driest periods of the year (Gray, 1998). Studies have been carried out on the effects of microclimatic conditions on population density, survival rate and the questing activity of ticks (e.g. Dantas-Torres and Otranto, 2013; Mejlon and Jaenson, 1997; Perret et al., 2003; Randolph et al., 2002; Ruiz-Fons et al., 2012; Vail and Smith, 2002) which is the posture ticks hold when they are reaching out for hosts. These microclimatic conditions include temperature, relative humidity (RH) and saturation deficit, which is the difference between the actual vapour pressure in the air and the total potential saturation vapour pressure at a given temperature. Some studies in vegetation control (e.g. Williams and Ward, 2010; Tack et al., 2013) showed that by changes in microclimatic conditions the population densities and questing behaviour may be influenced, which may eventually lead to a decrease in human Lyme infections. These studies therefore offer important insights for possibilities in forest management. This study was a follow-up study of a thesis study also conducted at Alterra Wageningen University and Research Centre. This previous study (Paree, 2013) was conducted under laboratorial conditions, where RH and temperature were regulated (at 10, 15 and 25°C and 40, 60 and 80% RH),
and confirmed that I. ricinus nymphs survived better at high RH and higher temperatures. Although no significant relation was found between questing and temperature or RH, there were larger amounts of ticks questing at higher RH.
Conducting studies under constant laboratorial conditions facilitates statistical analysis, but it may not represent conditions in the field. Climatic conditions fluctuate in the field, and a study conducted under fluctuating conditions might therefore show other interesting results than those conducted under constant conditions. Moreover, many studies consider the effect of yearly fluctuations in temperature or RH, by taking into account average daily temperatures or RH, but not the daily fluctuations. To decrease the number of human Lyme infections through forest management, considering e.g. vegetation control or recreation and the time of the day people enter the forest, an understanding of the effect of climatic fluctuations during the day on tick behaviour and survival is also needed.
2. RESEARCH
QUESTION
AND H
YPOTHESES
The main objective of this study was to determine whether and how daily fluctuating temperatures, relative humidity and saturation deficit affect questing behaviour and mortality of the sheep tick Ixodes ricinus.
Research question
What is the effect of daily fluctuations in temperature, relative humidity and saturation deficit under semi controlled conditions on questing quantity and mortality of Ixodes ricinus?
Hypotheses
Questing behaviour:
1. The number of questing ticks is positively related to relative humidity 2. The number of questing ticks is positively related to temperature 3. The number of questing ticks is negatively related to saturation deficit
4. The number of questing ticks is positively related to the number of hours above 7°C Mortality rate:
1. Mortality rate is negatively related to relative humidity 2. Mortality rate is negatively related to air temperature 3. Mortality rate is positively related to saturation deficit
3. LITERATURE
REVIEW
3.1 T
ICK ECOLOGYThe spatial distribution of I. ricinus covers temperate climates throughout Europe and North Africa, approximately from 10°W (Ireland) to 45°E (Ural mountains, Russia) and from 60°N (Sweden) to 30°N (Egypt) (Randolph et al., 2002). During their lifetime, I. ricinus ticks develop into three subsequent stages after hatching from an egg: larvae, nymph and adult. To develop into the next stage ticks need one blood meal. In order to find a blood meal, they ascend vantage points like grasses,
herbs or bushes, on which they wait with their front legs stretched out until a host passes, also called “questing” (see figure 1). Once a host passes, the ticks easily attach to it and walk to feeding places such as ears, neck and between the legs (Gassner, 2010). Here they can remain for 3 to 7 days until they are fully engorged (Gassner, 2010).
There are many ways how ticks can trace hosts. As with many other parasites, odour plays a very important role for locating hosts. Just as mosquitos
react to ammonia (Syed and Leal, 2009; Logan et al. 2008), ticks react to chemicals found in the human body by the use of receptor neurons in their Haller’s organ, including CO2, (Steullet and Guerin, 1992), organic acids and ammonia (Steullet and Guerin, 1994). The distribution of ticks is also determined by previous vantage point occupation, as suggested by Healy and Bourke (2008). Their study, conducted in a woodland, showed that some of the stems were more frequently occupied than others, and they suggest that this can be explained by odour since the sites were often crossed by sika and red deer, although they did not include this in their study.
Another study showed that nymphs reacted more often to movement of humans than adult ticks did (Vassallo and Pérez-eid, 2002). Since the nymphal stage is most responsible for Borrelia transmission (with the exception of adult females of I. persulcatus) (Gray, 1998), this may be another important factor for human Lyme infections.
In a study conducted near Stockholm, larvae were mostly found on small mammals like shrews (Sorex spp.) and rodents (Apodemus spp., Clethrionomys glareolus and Microtus agrestis), whereas nymphs and adults mostly chose larger hosts like roe deer (Capreolus capreolus) and hares (Lepus spp.) (Tälleklint and Jaenson, 1994). To maintain their population density it is therefore crucial for ticks that large animals are in the area, which play an important role in reproduction (Gray, 1998). Here, a distinction can be made between “reproduction” hosts and “reservoir” hosts, with the latter referring to the hosts that transmit the Lyme disease, such as rodents and birds which only larvae and nymphs feed on (although few exceptions have been observed where adults also feed on small animals) (Gray, 1998). In some cases, a reproduction host is also a reservoir host. In some areas which are not accessible to any reproduction hosts, tick populations can even be maintained by population “input” by birds and rodents that carry larvae and nymphs from surrounding areas and deposit the fed ticks in the area (Kirstein et al., 1997).
The average height at which ticks quest differs per species, stage and gender, and is obviously dependent on the height of the vegetation. However, all stages have been found to quest at heights of even 140 cm (Mejlon and Jaenson, 1997). Mejlon and Jaenson (1997) found only larvae and nymphs in low vegetation (≤80 cm), where they were most abundant between 0-9 cm and 30-39
Figure 1 . Questing female of Ixodes ricinus. Source: www.wageningenur.nl
cm respectively. In high vegetation (up to 140 cm), larvae were most abundant between 10-19cm, nymphs between 50-59 and adults between 60-79, which equals the height of the roe deer (Capreolus capreolus) (Mejlon and Jaenson, 1997). This vertical distribution could be explained by the tolerance to desiccation which is generally greater at later stages, but also by the choice of host, which is different among the stages (Mejlon and Jaenson, 1997).
In conditions that are unsuitable for questing or when a tick is developing into the next life stage, they can enter a state called diapause (Gray, 1998). Unlike quiescence, where they rest and absorb water to return questing later, ticks remain at the bottom for a longer period during diapause. Diapause can occur at all life stages, and strongly influences the season of questing (Gray, 1998). Ticks feed mostly during spring and early summer, and some also in autumn (Gray, 1991). This is partly due to the availability of hosts, but it is mostly determined by diapause, where day length seems to be a very important factor triggering diapause (Gray, 1998).
3.2 R
ELATIVE HUMIDITY AND SATURATION DEFICITAlthough it can take up to several months to develop into the next stage, depending on seasonal and climatic conditions (Gassner, 2010), the tick’s survival during each life stage is limited by the fixed energy resources from its current life stage (Perret et al., 2003) and the ability to maintain its water resources in very dry environments (Gassner, 2010), since ticks lose water during questing (Perret, 2003). Ticks can only absorb water vapour from the local microclimate (Kahl and Knülle, 1988), and are very vulnerable to RH levels below 80% (Kahl and Knülle 1988). Risks of desiccation are highest when ticks are in the questing phase, which can take several weeks, or in the developmental phase, when they transform into the next stage or lay eggs on or near the soil surface (Gray, 1998). In general, the risk of desiccation decreases at later stages (Mejlon and Jaenson, 1997). To replenish their water resources ticks go into a state called quiescence, when they descend to the litter layer to take up water vapour from the atmosphere (Kahl and Alidousti, 1997), generally when it becomes dark (Perret et al., 2003). Ticks can detect light intensity with their bilaterally placed strings of photoreceptors (Perret et al., 2003).
In a study by Lane et al. (1995) RH positively affected the number of questing ticks of Dermacentor
occidentalis. This correlation was also found by Vail and Smith (2002), when RH had a positive
effect on the questing height of I. scapularis. In a study by Dantas-Torres and Otranto (2013) the daily number of ticks collected in a forested area in southern Italy was negatively related to saturation deficit (SD), which is an indicator of the dryness of the air, and can be calculated by the formula:
SD=
(
1−
RH
100
)
4.9463 e
0.0621 T
(where: saturation deficits is expressed in millimetres of mercury, RH is in percentages, T is the temperature in degrees Celcius. Source: Perret et al., 2000).
SD was also found to positively affect the distance ticks walk after quiescence (Perret et al., 2003). SD does not seem to have an effect on the distance ticks walk after questing (Perret et al., 2003). Moreover, increases of SD decreases questing duration, whereas quiescence duration is not related to SD (Perret et al., 2003). Desiccating climates thus decrease the periods of questing and may therefore contribute to a decline in tick populations (Randolph et al., 2002). This is confirmed by Perret et al. (2000) in Switzerland and by Randolph et al. (2002) in the UK, where field experiments showed that sudden increases in SD occurred simultaneously with sudden decreases in tick populations. This demonstrates that SD plays an important role in tick population maintenance.
Randolph and Storey (1999) found that, under drier conditions, less nymphs were located in the upper layers of the herbage, and that the number of nymphs feeding on small rodents increased, as well as the rate of fat use. However, less larvae quested or fed on small rodents under dry conditions than under more humid conditions, and they suggested that larvae therefore become quiescent to prevent desiccation. They suggest that these changes in ratio of larvae to nymphs feeding on rodents due to decreasing RH may eventually contribute to variation in transmission of Lyme Borrelia.
3.3 T
EMPERATURESeveral studies have shown that I. ricinus survival and questing activity are strongly associated with temperature (Ruiz-Fons et al. 2012; Ogden et al. 2004; Dantas-Torres and Otranto 2013). In temperate climates, the yearly population dynamics are influenced by low temperatures, since winter temperatures are too low for interstadial development and questing (Tack et al., 2013). The optimal temperatures for development into next life stages lies between 8 - 11°C (Gassner, 2010). Temperature may decrease the duration of development periods to next life stages, as suggested by Ogden et al. (2004). During a study conducted in Switzerland, no ticks could be found when the maximum daily temperature did not exceed 1.9°C, or when the average daily temperature did not exceed 1.2°C (Perret et al., 2000). In the same study, they proposed a threshold of 7°C at which ticks become active again in spring. They suggest that the variation of tick density among the three study years could be explained by temperature, ranging between -4°C and 28°C over three years, where high density of questing ticks was associated with higher temperatures during spring. This positive relation was also found by Ruiz-Fons et al. (2012), where mean annual temperature affected questing tick abundance in Spain. They found that the abundance of questing larvae was greater at sites with a higher mean annual temperature (which ranged between 8.0–12.7°C among the sites), and at sites with fewer days with minimum temperatures below freezing point. They suggest that high questing larvae abundance is caused by higher reproduction of adult ticks, and that survival is mostly determined by the severity of the winter, provided that there is no moisture stress. In contrast, the daily number of ticks found in Italy in a study by Dantas-Torres and Otranto (2013) was negatively related to daily mean temperature, where most of the ticks were collected between 5°C and 19,9°C, but was not related to monthly mean temperature. This suggests that annual population dynamics are positively affected by annual temperatures, whereas daily tick abundance is negatively affected by temperature.
Interestingly, a study conducted in Africa, also considering Ixodid ticks, showed there were higher species diversity levels at sites with little variation in annual temperature and warmer temperatures (Estrada-Peña et al., 2012). They suggest each tick species and tick stage has a “climate niche”, including suitable meteorological and environmental conditions but also suitable host presence. Even though Africa does not have temperate climates, this study shows that populations of tick species may be influenced by variation in climatic factors, and not only just average, minimum or maximum conditions.
The height at which ticks quest on a stem does not seem to be related to temperature (Vail and Smith, 2002; Mejlon and Jaenson, 1997), although ticks walk greater distances at higher temperatures and are longer positioned in questing posture at 25°C than at lower or higher temperatures (Vail and Smith, 2002).
3.4 O
THER FACTORS INFLUENCING QUESTING AND SURVIVALSeveral other factors are important for consideration in decreasing human Lyme infections. These cannot be influenced directly through forest management, but they play an important role in vulnerability to desiccation and influence their questing behaviour.
As stated before, the survival of ticks is limited by their fixed energy resources, which can only be regained by a blood meal before developing into the next stage. Every time a tick ascends and descends a stem, but also when a tick takes up water vapour from the air, it requires energy (Perret et al., 2003). Higher fat contents makes a tick less susceptible for desiccation (Gassner, 2010). Randolph and Storey (1999) found that ticks were more active under drier conditions when they had just had a blood meal. This could be explained by the ticks’ ability to accumulate water from the air, which takes a lot of energy from the recent blood meal (Gassner, 2010). The benefits of high fat content also play a role in the infection of Borrelia in ticks. For Borrelia species it is beneficial for their reproduction if they would be able to influence tick behaviour to increase the chance a host is encountered. The spread of Borrelia infection can usually occur fast since a rodent remains infective for the rest of its life once it has been infected by a tick (Gern et al. 1994) and can therefore infect hundreds of larvae. Consequently, once these larvae become infective nymphs after development, Borrelia infections can spread fast. It has been found that infected nymphs were more active and had an increased energy supply due to a higher lipid fraction, compared to uninfected nymphs (Gassner, 2010). This increased energy supply in combination with increased activity may eventually lead to an increased survival rate of ticks and an increased spread of Borrelia infections. Herrman and Gern (2010) confirmed this when they found that the survival rate of infected I. ricinus ticks was higher than uninfected ticks, even under temperature and RH stress. This demonstrates that there is a mutualism between ticks and Borrelia species.
3.5 M
ICROCLIMATE CHANGE AND VEGETATION CONTROLI. ricinus is mostly restricted to forests (or other vegetation cover) where the litter layer is
sufficiently moist (≥80% RH) even during the driest periods of the year (Gray, 1998). Some areas where temperatures reach high levels during summer, in combination with low precipitation, tick populations decline and activity is low (Mannelli et al., 2011). However, a study in Ireland showed that ticks could even quest under hot and dry conditions with temperatures up to 35°C as long as there would be vegetation cover to provide moist (Gray, 1984). Vegetation, however, can also be affected in areas where temperatures are hot and precipitation is low. In some cases this is detrimental for ticks (when a moist litter layer is absent), but in some cases it can be beneficial for ticks when the species composition of trees or shrubs changes. For instance, fallen leaves of Beech (Fagus sylvatica), that can replace Norway spruce (Picea abies), provide a better microclimate for ticks (Manelli et al., 2011).
The restriction to a moist litter layer may offer possibilities in vegetation control, where unsuitable microclimate conditions can be created. Recently, an experiment was conducted by Tack et al. (2013) to examine the effects of changes in the environment by vegetation removal. This study showed that the abundance of larvae, nymphs and adults decreased where shrubs were cleared, even up to 2 years after clearing. However, they suggest that this is probably not due to changes in microclimate, but rather changes in host abundance. The largest temperature differences occurred during the winter, and they suggested that this did not explain the differences in tick abundance, since ticks are cold-hardy and can survive 24-hour exposures to temperatures below -14°C (Dautel and Knülle, 1997). Therefore, they suggested it could be explained by the absence of hosts that prefer shrub cover, for instance bank voles (Hille and Mortelliti, 2010). They suggested that shrub
clearing could be an effective and long-term (2 years) prevention of ticks bites, but that it should obviously be in line with sustainable forest management without implementing total vegetation clearing.
A study conducted in different forest stands (Lauterbach et al., 2013) showed there were more questing nymphs in thicket (as a young forest succession stage) than in pole wood or timber stages, suggesting an important impact of forest management on questing tick abundance. However, they mention that the effect of variable climatic conditions have probably more effect than forest management practices, in particular where mild winter temperatures result in higher tick abundance in the next year (Lauterbach et al. 2013). They stress that the effect of microclimate changes through vegetation control in combination with annual climate changes (e.g. increasing winter temperatures) can therefore only be studied with long-term data.
4. MATERIALS
AND
METHODS
4.1 T
ICK COLLECTIONTicks were collected from one population to ensure there were no genetic variations. They were collected in September and November at the Grebbeberg in Rhenen in The Netherlands, using the flagging method that has been used in many other studies (e.g. Perret et al. 2000; Williams and Ward, 2010; Vail and Smith, 2002) (see figure 2). A cotton cloth of one by one meter, attached to a stick, was dragged over low vegetation and checked every ten meters for nymphs. Nymphs that are
questing for hosts easily attach to the cloth. After transportation, the ticks were stored for two months in tubes with a wet cotton cloth to prevent desiccation, at a temperature of 5°C. In total, only 96 ticks could be used for the experiment, after many ticks had died during RH control testing and during the first experiment outside (see box, initial methodology: RH control).
4.2 A
RENASThe ticks were placed in glass arenas of 18 by 25 cm and 22 cm high, filled with a 5 cm layer of
gypsum on the bottom, in which five wooden skewers of 15 cm, and three skewers of 10 cm were stabbed vertically, that represented grass stems for questing (see figure 3). Some of the skewers that started moulding after a few weeks were replaced by new skewers when no ticks were questing on the skewer. Each arena was covered with fine nylon mesh glued to the glass with nail polish to prevent ticks from escaping. On this mesh, an opening was created, that could be closed with a lid, to facilitate fast observations and refilling the flasks and removing the loggers or skewers. Two small flasks with water were placed in each arena, with a strip of cotton (20 by 0,5 cm) under the lid leading to the bottom. This was done to offer a water source to prevent that ticks would die from desiccation. This did not increase the RH in the arena. Temperature and RH were registered by loggers (type: HOBO U23 Pro v2 Temperature/Relative Humidity Data Logger - U23-001), which were hung up in the arenas at half height. In total, 12 arenas were placed at three blocks of foudifferent locations in a greenhouse with each different, controlled temperatures (RH was not controlled). One block was placed in a greenhouse with controlled temperatures between around 16 and 27°C, another block under controlled
Figure 2. Collection of Ixodes ricinus at the Grebbeberg, Rhenen the Netherlands with the flagging method
temperatures between around 12 and 21°C, and the last was placed in a greenhouse where the temperature was equal to the temperature outside (which did not exceed 16°C during the observations), but would not drop below 3°C. Two of the four arenas in each block were covered with wood, to create shade, and the other two were covered with glass, to allow light to come in, and were fastened diagonally on a wooden plan with a rubber band. This was initially designed for outside, to prevent rain from coming in. However, it was still used to create some variation between the four arenas.
BOX. INITIAL METHODOLOGY: RH CONTROL
The eventual design of and the temperature and RH control in the arenas were not similar to how it was designed initially. The aim had been to place the arenas outside to allow for daily fluctuations in temperature and RH, but also to create different RH levels among the arenas by the use of silica gel that absorbs water vapour. Many methods to regulate RH with silica gel were tested before the start of the experiment. This took two months, which resulted in delay of the experiment. The aim was to create four different arena types, including one arena with increased RH (95-100%), one with slightly decreased RH (40-50%) one with extremely decreased RH (<20%), where each arena would be covered with a plastic foil to maintain the controlled RH. The fourth arena would be kept open to allow for natural RH fluctuations. It turned out to be extremely difficult to create an arena with <20% RH. It was important to place a water source in this arena, since previous studies showed that ticks die at extremely low RH. The aim was therefore to decrease the RH of the entire arena, while there was still a water source available that would not affect these low RH levels. Several attempts were done, including a barrier by hanging a small bucket up-side-down in which a water tray with gauze was placed on the bottom, so that the ticks could walk under the bucket onto the water tray. Bags with silica gel that absorbs water were hung on the outside of the bucket so that any water that would evaporate out of the bucket would be absorbed immediately, and not evaporate into the arena. This seemed to be effective regarding the RH (it often remained below 20% RH, sometimes even below 10%, and there was still water in the tray), but the ticks were unable to detect the water tray and still died within two days. Other trials included “open” water sources in the arena, to allow for a water vapour gradient. For this, a lot of silica gel (400 gr) was hung in bags on the ceiling of the arena, small flasks were filled with water, from which small strips of cotton cloth were leading to the bottom of the arena, so that water would be transported by the cotton. In the case where the gypsum bottom of the arena was covered with plastic foil to prevent that water would be absorbed by the gypsum, the RH dropped below 20%, but the ticks were found dead on the edges of the arena that were not covered by plastic. When the bottom was covered with plastic foil, the RH did not drop below 40%. It was not possible to find a method to create an RH below 20% while the water source was detectable for the ticks, since there was either too much water that could not be absorbed by the silica gel, or the ticks could not find the water. During the testing of this arena type, often all ticks died after even one day, if they could not find the water source and RH reached below 20%. Since no method was found to control extremely low RH after two months, and the weather was becoming colder, the decision was made to only focus on daily natural fluctuations of RH and temperature without controlling them, by covering the arenas only with a gauze. 700 ticks were collected in September in three days. However, during the testing of these methods, many ticks died from desiccation and could therefore not be used for the eventual experiment. Moreover, some of the ticks died in the refrigerator where they were stored temporarily, due to water drops from the wet cotton that ensured a high RH level, but caused some of the ticks to drown. In November, only another 23 ticks could be collected from the forest at the Grebbeberg due to the cold weather. Eventually, only 96 ticks could be used for the experiment.
Tick activity was observed at 1000 hours and at 1430 hours, using an observation form (see Appendix: Observation form), for four weeks from mid November to mid December. The number of questing and walking ticks were noted in the morning and the afternoon. Some of the ticks that were located on the bottom suddenly quested when they saw the observer, and were therefore not considered as “questing” but as walking or lying on the bottom. Only ticks that had deliberately ascended a skewer, the wall, the roof or even one of the water flasks were considered as “questing”. This included ticks that were already positioned in questing posture, but also ticks that were located on a skewer, the wall, the roof or a flask and started questing when they saw the observer. Ascendance of a skewer, the wall, roof or a flask was therefore a criteria for “questing”, rather than the questing posture, since it is the ascendance of a vantage point which makes ticks vulnerable for desiccation in the field. The number of dead ticks were only noted during the morning since this was quite time consuming and would possibly cause too much disturbance. Ticks that seemed to be dead were taken out of the arena and exposed to breath to trigger movement, or taken to a warmer greenhouse compartment (especially the ticks in the cold greenhouse compartment that sometimes showed little movement). When no movement was seen after five minutes they were considered dead. The data were imported in excel and analysed on a daily basis. Temperature (in °C) and RH (in %) were measured every hour by data loggers, to obtain a reasonable dataset of the fluctuations. Saturation deficit was calculated using the equation that has been given in the chapter Literature Review.
4.4 D
ATA ANALYSISTo answer the research question, which includes daily fluctuations in temperature (T), relative humidity (RH) and saturation deficit (SD), a distinction is made between the following variables: mean, minimum and maximum T, RH and SD. Moreover, the proposed thresholds of 7°C, below which ticks become less active (Perret et al., 2000), and 80%RH below which ticks are vulnerable to desiccation (Kahl and Knülle 1988) were also included in the analysis, expressed in #hours ≥7°C and #hours ≥80%RH. These variables, questing rate and mortality rate were calculated in Microsoft Excel, and important in GenStat to carry out a logistic regression analysis. All the different variables were calculated over the 24 hours prior to each observation, and observations started after 24 hours. The logistic regression analysis enabled analysing the binomial data of the number of questing and dead ticks, by transforming the questing and mortality rates (=[#questing or dead ticks]/[#total ticks in the arena) into logits, using the formula:
logit( p)=ln
(
p
1− p
)
Where p stands for probability, which is [#questing or dead ticks] / [#total ticks in the arena]. In total, 22 logistic regression analyses were executed, where confidence limits for estimates were set to 95%. Firstly, it was important to exclude whether day part of the observations interacted with the microclimatic variables, i.e. whether the effects of the microclimatic variables on questing behaviour and mortality were different in the morning and the afternoon. This was done by multiplying the variables with the factor “day part”. If the outcome if this interaction was not significant, day part was added as an extra, non-interactive factor in all regression analyses.
For each regression analysis, firstly the significance of the overall analysis was tested without making a distinction between the different variables. The outcome, the F-probability (=p-value),
was the probability that the coefficient of all variables together is zero (where a single variable can still be significant if the F-probability is not).
After that, every variable was regressed against the logits of questing and mortality rate. Since the arenas may have had an effect on the questing behaviour and mortality of the ticks as well, arena was included as a “random factor”, to separate the effect of the individual arenas from the effects of T, RH and SD. Depending on whether there was interaction between the variables and the day part of observation, day part was added as an interactive or non-interactive factor.
Since mortality was only registered during the morning, only the data of T, RH and SD in the morning were included for mortality analysis. Graphs were drawn with Genstat, by plotting the variables against the transformed logits of the questing and mortality rates.
5. RESULTS
5.1 O
VERVIEW OF THE MICROCLIMATES IN THE COMPARTMENTS An overview of T, RH and SD in the different compartments is given in table 1. Compartment 16 was the coldest, which was a greenhouse that was open to allow for fluctuations similar to those outside, but would be controlled so that T would not drop below 3°C, and it received no more light than outside. Compartment 8 was the warmest, and received 16 hours of light. Compartment 20 also received 16 hours of light. Although the differences in light duration between the compartments may have had an influence on the data, there was no equipment available to log light intensity.Table 1.Temperature, RH and SD conditions per compartment
Compartment 8 Compartment 16 Compartment 20 T, average T, minimum 20,53 16,84 7,23 3,33 16,00 12,20 T, maximum 26,26 15,15 21,50 RH, average 82,36 92,62 85,49 Rh, minimum 64,87 64,49 57,42 RH, maximum 93,87 99,76 96,47 SD, average 3,22 0,61 2,01 Sd, minimum 0,96 0,02 0,41 SD, maximum 8,37 3,75 6,82
There was no interaction between any variable and the part of the day of observations (morning or afternoon), meaning that the effects of T, RH and SD on questing behaviour and mortality were not different in the morning and the afternoon (see table 2). Therefore, day part was added as an extra, non-interactive factor in all regression analysis.
Table 2. T-probabilities of the interaction tests between day part of observation and variables
Variable Interaction between day part of observation and variable (t-probability) Tmean 0.151 Tmin 0.202 Tmax 0.477 #Hours ≥7°C 0.314 RHmean 0.140 RHmin 0.207 RHmax 0.113 #Hours ≥80% RH 0.084 SDmean 0.084 SDmin 0.072 SDmax 0.176 Where:
- T-probability = p-value of the interaction between day part of observation and the variables
5.2 Q
UESTINGBelow, the questing rates per compartment can be found in figure 4. It is interesting to see that the dynamics of questing in compartment 16 (cold) seemed to be more restricted to temperature (which affected questing significantly, see table 3) than the questing in compartment 8 (warm), since the fluctuations in questing in compartment 16 were more or less parallel to the fluctuations in mean temperature. Questing in compartment 8 did not show a distinctive parallel with temperature. 0 2 4 6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12
Questing and Tmean per observation
Questing ratio (comp 8, warm T)
Mean questing ratio Comp 8
Questing ratio (comp 20, moderate T)
Mean questing ratio Comp 20
Questing ratio (comp 16, cold T)
Mean questing ratio Comp 16 Tmean, compartment 8 Tmean, compartment 20 Tmean, compartment 16 Thresshold of 7°C Observations Q u e sti n g ra te M e an t e m p e ra tu re
Figure 4 . Questing rates (left axis) and mean temperatures (right axis) per compartment. The threshold of 7°C is also given.
In figures 5, 6 and 7 a better view is given of the dynamics of temperature, including minimum and maximum temperatures. Here we can see clearly that even when temperatures dropped below the proposed threshold of 7°C in compartment 16, there was still questing activity. In compartment 8 (warm temperatures), there was an increase in temperature and drop in RH during the 30th
observation (day 15) (see figure 5, observation 30, and figure 11, day 15), where T reached 26°C and RH dropped to 73%. At this moment there was a smaller questing rate than at lower
0 2 4 6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12
Questing and temperature, comp. 16
Que stin g ra-tio per ob- ser- va-tion Me an que stin g ra-tio Tma x Tme an Tmi n Observations Q u e sti n g ra ti o M e an t e m p e ra tu re
temperatures. The outcomes of the logistic regression analyses on questing behaviour can be found in table 3, where every row gives the outcome of a different analysis. All regression analyses showed that some of the variation was explained by the part of the day the observations were done (all t-probability values were highly significant). This can also be seen in figures 8 and 9, where morning observations (the lowest trend) showed less questing activity than in the afternoon. Temperature was in all cases (mean, min, max and #Hours ≥7°C) significant, showing higher questing activity at higher temperatures.
This positive trend can also be seen in figures 8 and 9. Interestingly, this study confirms that there is threshold of 7°C above which the questing activity of ticks is significantly greater. However, it has not been tested whether the number of hours ≥6°C or ≥8°C was also significant, or any other temperature close to 7°C. Some studies show that ticks quest below 7°C as well (Jaenson and lindgren, 2011; Dantas-Torres, 2013), which is confirmed by this study. The threshold may therefore not be that strict. RH and SD were in all cases (mean, min, max and #Hours ≥80% RH) not significant.
Figure 7. Questing (left axis) and temperature (right axis), compartment 20
0 2 4 6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12
Questing and temperature, comp. 20
Qu es tin g ra tio pe r ob ser va tio n M ea n qu es tin g ra tio Tm ax Days Q u e sti n g ra ti o M e an t e m p e ra tu re
Figure 6. Questing (left axis) and temperature (right axis), compartment 16
0 2 4 6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12
Questing and temperature, comp. 20
Ques ting ratio per ob- ser- va-tion Mea n ques ting ratio Tmax Tmea n Tmin Thre ssh old of 7°C Days Q u e sti n g ra ti o M e an t e m p e ra tu re
Figure 8. Logit transformed questing rates against Tmean, morning (lower green line) and afternoon (upper red line)
Figure 9. Logit transformed questing rates against #Hours ≥7°C, morning (lower green line) and afternoon (upper red line)
Table 3. Results questing (regression analysis) F-probability (p-value) T-probability per variable T probability of the factor
“day part” (morning or afternoon) Positive or negative (only given if significant) <.001 TMean <0.001 <0.001 positive <.001 TMin <0.001 <0.001 positive <.001 TMax <0.001 <0.001 positive <.001 #Hours ≥7°C <0.001 <0.001 positive <.001 RHMean N.S. <0.001 <.001 RHMin N.S. <0.001 <.001 RHMax N.S. <0.001 <.001 #Hours ≥80% RH N.S. <0.001 <.001 SDMean N.S. <0.001 <.001 SDMin N.S. <0.001 <.001 SDMax N.S. <0.001 Where:
- F-probability = the probability that the overall analysis was significant
- T-probability per variable = p-value of the relation between the variable and questing logits
- T probability of the factor “day part” (morning or afternoon) = p-value of the difference in questing
between morning and afternoon
- Positive or negative: indicates positive or negative relation (only relevant if significant)
5.3 M
ORTALITYThe accumulative mortality per compartment can be seen in figure 10, which also shows the differences in RHmean among the compartments. This graph shows that many ticks died at the start, and that the mortality was lowest in compartment 8. Only a few more ticks died in compartment 20 compared to compartment 16. During the last four days in compartment 8 there was a sudden increase in mortality (3 ticks died in 4 days whereas during the previous observations 4 ticks had died in 11 days), as well as in compartment 16. Figures 11, 12 and 13 give a better view of the dynamics of RH, including minimum, maximum and mean RH, and the threshold of 80%RH. Here we can see clearly that in all 3 compartments RH regularly drops below 80%, and that the number of dead ticks often coincides with low RH levels. The outcomes of the regression analysis shows that in all cases RH (mean, min, max and #hours ≥80%RH) affected mortality rate negatively, indicating that dry conditions resulted in higher mortality (see table 4). These trends can also be seen in figures 14 and 15, where the logit transformed mortality rates are plotted against RHmean and #Hours ≥80%RH. SDmean (see figure 16) and SDmin showed a positive relation with mortality, also indicating that drier conditions resulted in higher mortality. SDmax was only just not significant (T-probability = 0.066) Mean, minimum, maximum temperature and the number of hours above 7°C did not affect mortality, which is in contrast with questing. 1 2 3 4 5 6 7 8 9 1011121314151617 55 60 65 70 75 80 85 90 95 100 0 1 2 3 4 5 6 7
RH and mortality, comp 8
R h m ax R h m ea n R h mi n Days R H ( % ) N u m b e r o f d e ad ti ck s
Figure 10. Accumulative mortality in number of ticks
1 2 3 4 5 6 7 8 9 1011121314151617 55 60 65 70 75 80 85 90 95 100 0 1 2 3 4 5 6 7
RH and mortality, comp 8
Rh m ax Rh m ea n Rh mi n Days R H ( % ) N u m b e r o f d e ad ti ck s
Figure 11. RH (left axis) and number of dead ticks (right axis), compartment 8
1 2 3 4 5 6 7 8 9 1011121314151617 55 60 65 70 75 80 85 90 95 100 0 1 2 3 4 5 6 7
RH and mortality, comp 16
Rhm ax Rhm ean Rhmi n Thre sh-old 80% RH #dea d ticks Days R H ( % ) N u m b e r o f d e ad ti ck s
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 55 60 65 70 75 80 85 90 95 100 0 1 2 3 4 5 6 7
RH and mortality, comp 20
Rhmax Rhmean Rhmin Thresh old 80%RH #dead ticks Days R H ( % ) N u m b e r o f d e ad ti ck s
Figure 13. RH (left axis) and number of dead ticks (right axis), compartment 20
Figure 16. Logit transformed mortality rates against SDmea Figure 15. Logit transformed mortality rates against #Hours ≥80%RH
Table 4. Results mortality (regression analysis)
F-probability T-probability per variable Positive or negative
(only given if significant)
0.078 TMean: N.S. 0.179 TMin N.S. 0.002 Tmax 0.010 negative 0.114 #Hours ≥7°C N.S. <0.001 RHMean <0.001 negative 0.007 RHMin <0.001 negative <0.001 RHMax <0.001 negative <0.001 #Hours ≥80%RH <0.001 negative <0.001 SDMean <0.001 positive <0.001 SDMin <0.001 positive 0.087 SDMax N.S. Where:
- F-probability = the probability that the overall analysis was significant
- T-probability per variable = p-value of the relation between the variable and questing logits - T probability of the factor “day part” (morning or afternoon) = p-value of the difference in questing between morning and afternoon
6. DISCUSSION
It was not possible to exclude the influence of the observer on the questing behaviour of the ticks, although the observations were done as quickly as possible. Sometimes it was necessary to place a hand in the arena (when the loggers needed to be replaced or checked for batteries, when moulded skewers had to be replaced or when seemingly dead ticks needed to be checked). A few ticks that were questing on the roof or on a skewer dropped themselves when they saw a hand. This may have influenced the data of the subsequent observations.
Annual dynamics in tick populations in temperate climates are highly seasonal, since the duration of development through the life stages is negatively affected by temperature (Ogden, 2004), while the ability to develop is dependent on the severity of the winter (Randolph, 2004). Perret et al. (2000) suggested a threshold of 7°C below which ticks are inactive, which largely affects the dynamics in tick populations. This study could confirm that questing activity below 7°C was indeed significantly different than activity above this threshold, although there was still questing activity at lower temperatures. This was found earlier by Dantas-Torres and (2013) who found actively questing I. ricinus ticks at cold days during the winter, when daily temperatures ranged between 0.5°C to 4.9°C. In a study in Sweden it was suggested that I. ricinus is active at temperatures ≥5°C (Jaenson and lindgren, 2011). In a study by Herrman and Gern (2013) where they exposed ticks to changing temperatures between -10 to 13°C in a laboratory, it was found that the frequency of variation in low temperature is determinant for the survival of ticks. This might suggest that there are different effects of winter temperatures, where low temperatures negatively affect development, whereas variation in low temperature may affect survival of individual ticks.
While low winter temperatures seem to affect annual dynamics in tick populations, higher temperatures affect daily questing activity, but may also affect mortality. In temperate zones, questing tick abundance seems highest during the summer months (Hudson et al., 2001, see Dantas-Torres). However, in hot and dry climates, tick abundance seems limited by high temperatures causing higher mortality during the summer (Dantas-Torres, 2013). Tick abundance may therefore be larger during winter, spring and autumn, as suggested by Dantas-Torres (2013) in a study in Southern Italy. In their study, daily questing tick abundance was negatively related with daily mean temperature, where ticks were less abundant during summer. Interestingly, they also found ticks questing during the winter at temperatures between 0.5°C to 4.9, and they suggested other factors may play a role in tick behavior in those areas, possibly associated with vegetation (Dantas-Torres, 2013). This may suggest that temperature is related to questing but that it also limits survival if it reaches high levels combined with low RH. Although this study could confirm that there is less questing activity below 7°C than at higher temperatures, it was not possible to determine whether high temperatures indeed affected tick survival negatively. Moreover, due to the limited sample size, it was not possible to determine whether, for instance, maximum or minimum temperature had more effect on questing behaviour, nor whether there would be a threshold above which the effect of temperature on questing behaviour is different than below the threshold. For instance, questing behaviour at low temperatures seemed to be more limited by temperature, since the graph of compartment 16 showed a clear parallel between questing and temperature. In contrast, the dynamics in questing behaviour at high temperatures (in compartment 8) were not as parallel to temperature as in compartment 16, suggesting that the dynamics in questing behaviour under warmer conditions may (also) be affected by something else. Vail and Smith (2002) found that the duration of questing posture was greater at 25°C than at lower or higher temperatures, with a range of 10 - 30°C, indicating that questing and temperature are
positively related below 25°C, whereas they are negatively related above 25°C. However, questing can still occurs even until temperatures of 35°C, provided that there is a moist litter layer (Gray, 1984). If 25°C is indeed the ideal questing temperature (provided that there is no moisture stress), any variation in questing behaviour at this temperature (which was more or less the temperature in compartment 8) is probably explained by something else. This might also explain why some studies in questing tick abundance show contrasting results, where some of the outcomes suggest a negative relation between temperature and questing (e.g. Dantas-Torres in Southern Italy), while others suggest a positive relation (e.g. Zakovska in Czech Republic), although many other underlying causes may have had an influence on these contrasting results.
In the regression analysis of this current study all compartments were combined, showing an overall effect of all T variables, but no extra test was executed to see whether the effects of warmer temperatures were indeed different than the effects of lower temperatures. However, during the 30th observation (day 15) there was a sudden increase in temperature and a decrease in questing in
compartment 8, followed by an increase in mortality during the following days, which is in line with the results of Vail and Smith (2002) mentioned above. Due to a low sun that had entered the arenas in the morning, temperatures rose high combined with a drop in RH (73%). This possibly indicates a negative effect of very high temperatures and the drop in RH on both questing and survival. The logistic regression analysis shows that there was indeed an effect of Tmax on mortality, but it had a negative relation, rather than a positive (indicating that higher Tmax resulted in lower mortality). However, considering that this was only a single event that temperatures reached such high levels, it would have been unlikely that this single event had an effect on the outcomes. It is possible that until moderately high temperatures (around 25°C) warmer conditions result in lower mortality and higher questing rate, and that hot conditions (>25°C) result in higher mortality and lower questing rate, but that this did not influence the outcome of the regression. Since the compartments in this study did not show enough temperatures above 25°C, this threshold could not be proven. Except for the negative relation between Tmax and mortality no relations with the other temperature variables were found.
The results do show a negative effect of relative humidity on mortality, which is in line with previous studies (e.g.). However, no relation could be found between RH and questing, although previous studies found a an increase in questing activity under higher RH (e.g. Randolph and Storey, 1999; Lane et al 1995; Vail and Smith, 2002; Dantas-Torres and Otranto, 2013). This is probably due to the high levels of RH in this study. Although in all compartments RH dropped below 80% regularly, the maximum and even average RH reached above 80% every day (except for compartment 8 during the first two days). The number of hours below 80% did have an effect on questing, but the overall range of RH was probably still too high to show significant differences in questing behaviour at different RH levels. SD was also not related to questing, but since SD is related to RH, this is probably also caused by the high levels of RH.
7. RECOMMENDATIONS
ON
FOREST
MANAGEMENT
As stated in this report, dynamics in tick populations are highly seasonal and are limited by the severity of winters, and the temperatures during hot summers. Reducing tick population over a large area without very drastic forest measures (such as large scale clearance) is obviously not recommended, since the importance of forests and biodiversity are globally recognized. However, local changes in the microclimate can have an impact on the local distribution of ticks by creating unsuitable habitats where temperature are too high in the summer, too low in the winter or where RH levels become too low.
Studies in shrub cover demonstrated that high shrub cover had higher tick abundance than low cover (Tack et al., 2012) or clearance (Tack et al., 2013). Clearance seemed to be most efficient for reducing tick abundance, since abundance was suppressed for 2 years after clearing (Tack et al., 2013). These cleared plots showed lower winter temperatures, but they suggested that the low tick abundance was not due to changes in winter temperatures, since ticks survived 24-hour exposures to temperatures below -14°C in a study by Dautel and Knülle (1997). Therefore, they suggested it could be explained by the absence of hosts that prefer shrub cover, for instance bank voles (Hille and Mortelliti, 2010). However, Herrman and Gern (2013) showed that not the coldness of winter temperatures limits survival but the variation in temperature. The low abundance of ticks in cleared plots could therefore still (partly) be explained by great temperature variations in the winter. For forest management measures it is, however, not very important to know what the underlying causes are. Whether the habitat is made unsuitable for ticks directly by changing the microclimate or indirectly by creating unsuitable host habitats does not make a difference.
If clearance would be applied in only a few places in a forest, the overall annual population could be decreased, but this may not have much impact if the other areas are still very suitable habitats. However, since recreationists in The Netherlands and perhaps other European countries are restricted to forest tracks, these areas need only be taken near forest tracks. Cleared areas can be left for regrowth of the vegetation after the winter. If the low tick abundance in cleared plots is indeed (partly) explained by the fluctuations of winter temperatures, rather than mean temperatures, the time period that these areas should remain cleared may depend on how long ticks survive in diapause during fluctuating temperatures. However, since regrowth does not occur at low temperatures, this is may not relevant. Clearance along forest tracks during summer can also cause tick mortality where RH levels of the litter layer are decreased and temperatures rise. Some of the current forest measures actually increase tick populations. For instance, the improvement of the vertical forest structure includes shrub layers that are suitable habitats for ticks, and may eventually even lead to increases in tick populations. These measures could be taken in special designated areas that are away from forest tracks, to prevent tick encounters.
Finally, the forest and nature areas that are entered by people belong to tick habitat. Unless total clearance or any other measure that destroy entire tick populations is implemented, ticks will always be present in recreation areas. Since I do not plea for such destruction of populations, and I assume many foresters, ecologists and conservationists agree, this is not an option. Therefore, people that enter tick habitats should always be cautious and check on any tick bites after they leave the area, and, if any tick is spotted on the skin, it should be removed before 24 hours, since infection does not occur within 24 hours after the bite (Mladenovic et al., 2010). Moreover, the best way to prevent tick bites or infection is by providing people of information about Lyme disease.
8. C
ONCLUSIONS
With such a small sample size and short period of only four weeks it was difficult to determine whether there are other underlying causes that affect questing behaviour and mortality. There was still a large amount of the variation in questing and mortality unexplained, and it is difficult to determine whether the questing of ticks was mostly determined by, for instance, the minimum or the maximum temperatures. Since some of the changes in RH, T and SD may have long term effects on mortality and questing behaviour. Means, minima, maxima and the number of hours above the thresholds of 7°C and 80%RH were calculated over the 24 hours prior to each observation, but it is possible that the effects are on a longer or shorter term. Some of the hypotheses should be rejected since no relation could be found with the variable. This may be caused by the small sample size. Previous studies showed more relations to microclimate than this current study. These findings are limited and no further conclusions can be drawn on those variables where no relation was found.
However, this study still provides outcomes that are in line with previous studies, with the following conclusions:
Questing
Temperature showed a positive relation with the number of questing ticks The number of questing ticks was positively related to the number of hours above 7°C
Mortality
Relative humidity was negatively related to mortality
Among the temperature variables, only maximum temperature showed a negative relation with mortality
Mean and minimum saturation deficit was positively related to mortality Mortality was negatively related to the number of hours above 80% RH If there are any more thresholds in the effects of variables (for instance that temperature may affect questing differently under 25°C than above 25°C), it would be interesting and worthy to carry out more in-depth research on this. Lyme disease and other tick-borne diseases are spread out over different climates in several continents. It is important to know whether the effects of temperature, relative humidity or saturation deficit are indeed different at lower or higher levels. If so, it may differ per climate what measures could be taken in tick habitat to reduce human Lyme infections. This was a follow-up study of the study by Paree (2013) where the effects of constant microclimatic conditions in a laboratorial setting were tested. The results were similar, since the same effects were found for temperature, relative humidity and saturation deficit. However, Paree could not find any relation between mortality and saturation deficit, nor between temperature and questing.
9. ACKNOWLEDGEMENTS
I would like to thank my collegues at Alterra Wageningen University and Research Centre for assisting and helping me during my thesis study, in particular my supervisor Nico van den Brink, and Wim Dimmers. My thanks also go to Peter van der Meer, who has been my second supervisor at Van Hall Larenstein University of Applied sciences. And lastly, I would also like to thank the forest managers at Natuur Monumenten who gave me permission to collect ticks at the Grebbeberg.
10. R
EFERENCES
Dantas-Torres, F., Otranto, D. (2013), Species diversity and abundance of ticks in three habitats in southern Italy. Ticks and Tick-borne Diseases 4:251–255.
Dautel, H. and W. Knulle (1997), Cold hardiness, supercooling ability and causes of low-temperature mortality in the soft
tick, Argas reflexus, and the hard tick, Ixodes ricinus (Acari: Ixodoidea) from Central Europe. Journal of Insect
Physiology 43(9): 843-854.
Estrada-Peña, A., Estrada-Sánchez, A., Estrada-Sánchez, D., (2012). Occurrence patterns of Afrotropical ticks (Acari: Ixodidae)
in the climate space are not correlated with their taxonomic relationships. PLoS ONE 7: e36976.
Gassner, F. (2010), Tick tactics. Interactions between habitat characteristics, hosts and mictroorganisms in relation to the
biology of the sheep tick I. ricinus. Wageningen, Wageningen UR. PhD: 162.
Gern, L., Siegenthaler, M., Hu, C. M., Leuba-Garcia, S., Humair, P. F., and Moret, J. (1994), Borrelia burgdorferi in rodents
(Apodemus flavicollis and A. sylvaticus): duration and enhancement of infectivity for Ixodes ricinus ticks. European
Journal of Epidemiology 10: 75-80.
Gray, J. S. (1984), Studies on the dynamics of active populations of the sheep tick, Ixodes ricinus L., in Co. Wicklow, Ireland. Acarologia 25: 167–178.
Gray, J. S. (1991), The development and seasonal activity of the tick Ixodes ricinus: a vector of Lyme borreliosis. Review of Medical and Veterinary Entomology 79:323-333.
Gray, J. S. (1998), The ecology of ticks transmitting Lyme borreliosis. Experimental and Applied Acarology 22:249-258. Healy, J. A. E., and Bourke, P. (2008), Aggregation in the Tick Ixodes ricinus (Acari: Ixodidae): Use and Reuse of Questing
Vantage Points. Journal of Medical Entomology 45(2): 222-228.
Herrmann, C., and Gern, L., (2010), Survival of Ixodes ricinus (Acari: Ixodidae) under challenging conditions of temperature
and humidity is influenced by Borrelia burgdorferi sensu lato infection. Journal of Medical Entomology 47 (6), 1196–
1204.
Herrmann, C., Gern, L., (2013), Survival of Ixodes ricinus (Acari: Ixodidae) nymphs under cold conditions is negatively
influenced by frequent temperature variations. Ticks Tick Borne Dis. 4:445–451.
Hille, S.M., Mortelliti, A. (2010), Microhabitat partitioning of Apodemus flavicollis and Myodes glareolus in the sub-montane
Alps: a preliminary assessment. Hystrix, the Italian Journal of Mammalogy 21:157–163
Hofhuis, A., Van der Giessen, J. W. , Borgsteede, F. H., Wielinga, P. R., Notermans, D. W. and Van Pelt, W. (2006), Lyme
borreliosis in the Netherlands: strong increase in GP consultations and hospital admissions in past 10 years. Euro
Surveillance, 11(25):pii=2978.
Hudson, P.J., Rizzoli, A., Rosà, R., Chemini, C., Jones, L.D., Gould, E.A., (2001), Tickborne encephalitis virus in northern Italy:
molecular analysis, relationships with density and seasonal dynamics of Ixodes ricinus. Medical and Veterinary
Entomology. 15, 304–313.
Jaenson, T.G.T., Tälleklint, L., Lundqvist, L., Olsén, B., Chirico, J., Mejlon, H., 1994. Geographical distribution, host
associations, and vector roles of ticks (Acari: Ixodidae, Argasidae) in Sweden. Journal of Medical Entomology Med.
Entomol. 31: 240–256.
Jaenson, T.G., Lindgren, E. (2011), The range of Ixodes ricinus and the risk of contracting Lyme borreliosis will increase
northwards when the vegetation period becomes longer. Ticks Tick Borne Dis 2: 44–49.
Kahl, O. and Alidousti, I. (1997). Bodies of liquid water as a source of water gain for I. ricinus (Acari: Ixodidae). Experimental and Applied Acarology. 21: 731.
Kahl, O., and Knülle, W. (1988), Water vapour uptake from subsaturated atmospheres by engorged immature Ixodid ticks. Experimental and Applied Acarology 4: 73-83.
Kirstein, F., Rijpkema, S., Molkenboer, M. and Gray, J.S (1997), The distribution and prevalence of B. burgdorferi
genomospecies in Ixodes ricinus ticks in Ireland. European Journal of Epidemiology 13: 67–72.
Lane, R. S., Kleinjan, J. E. and Schoeler, G. B. (1995), Diel Activity of Nymphal Dermacentor occidentalis and I. pacificus
(Acari: Ixodidae) in Relation to Meteorological Factors and Host Activity Periods. University of California: 32(3):
290-299, Berkeley.
Lauterbach, R., Wells, K., O’Hara, R. B., Kalko, E. K. V. and Renner, S. C. (2013), Variable Strength of Forest Stand Attributes
and Weather Conditions on the Questing Activity of Ixodes ricinus Ticks over Years in Managed Forests. PLoS ONE
8(1): e55365.
Logan, J., Birkett, M., Clark, S., Powers, S., Seal, N., Wadhams, L., Mordue, L. and Pickett, J. (2008), Identification of human-derived volatile chemicals that interfere with attraction of Aedes aegypti mosquitoes. Journal of Chemical Ecolology 34 (3): 308–322.
Mannelli, A., Bertolotti, L., Gern, L. and Gray, J. (2011), Ecology of Borrelia burgdorferi sensu lato in Europe: transmission
dynamics in multi-host systems, influence of molecular processes and effects of climate change. FEMS Microbiology
