The German bee monitoring project: a long term …The German bee monitoring project: a long term...

21
Apidologie Available online at: c INRA/DIB-AGIB/EDP Sciences, 2010 www.apidologie.org DOI: 10.1051/apido/2010014 Original article The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch 1, **, Werner von der Ohe 2, **, Hannes Kaatz 3 , **, Annette Schroeder 4 , Christoph Otten 5 , Ralph B ¨ uchler 6 , Stefan Berg 7 , Wolfgang Ritter 8 , Werner M ¨ uhlen 9 , Sebastian Gisder 1 , Marina Meixner 6 , Gerhard Liebig 4 , Peter Rosenkranz 4, ** 1 Institute for Bee Research, Friedrich-Engels-Str. 32, 16540 Hohen Neuendorf, Germany 2 LAVES Institut für Bienenkunde Celle, Herzogin-Eleonore-Allee 5, 29221 Celle, Germany 3 Universität Halle-Wittenberg, Institut für Biologie, Molekulare Ökologie, Hoher Weg 4, 06099 Halle, Germany 4 Apicultural State Institute, University of Hohenheim, August-von-Hartmannstrasse 13, 70599 Stuttgart, Germany 5 DLR Fachzentrum Bienen und Imkerei, Im Bannen 38-54, 56727 Mayen, Germany 6 LLH Bieneninstitut Kirchhain, Erlenstr. 9, 35274 Kirchhain, Germany 7 LWG Fachzentrum Bienen, An der Steige 15, 97206 Veitshöchheim, Germany 8 CVUA Freiburg, National and OIE Reference Laboratory, Am Moosweiher 2, 79108 Freiburg, Germany 9 Landwirtschaftskammer Nordrhein-Westfalen, Aufgabengebiet Bienenkunde, Nevingho40, 48147 Münster, Germany Received 18 November 2009 – Revised 17 January 2010 – Accepted 30 January 2010 Abstract – The Western honey bee, Apis mellifera, is the most important animal pollinator in agriculture worldwide providing more than 90% of the commercial pollination services. Due to the development in agriculture the demands for honey bee pollination are steadily increasing stressing the pollination capacity of the global managed honey bee population. Hence, the long-term decline of managed honey bee hives in Europe and North-America is of great concern and stimulated intensive research into the possible factors presumably causing honey bee colony collapse. We here present a four-year study involving more than 1200 bee colonies from about 120 apiaries which were monitored for the entire study period. Bee samples were collected twice a year to analyze various pathogenic factors including the ectoparasitic mite Varroa de- structor, fungi (Nosema spec., Ascosphaera apis), the bacterium Paenibacillus larvae, and several viruses. Data on environmental factors, beekeeping management practice, and pesticides were also collected. All data were statistically analyzed in respect to the overwintering mortality of the colonies. We can demon- strate for several factors that they are significantly related to the observed winter losses of the monitored honey bee colonies: (i) high varroa infestation level, (ii) infection with deformed wing virus (DWV) and acute bee paralysis virus (ABPV) in autumn, (iii) queen age, and (iv) weakness of the colonies in autumn. No eects could be observed for Nosema spec. or pesticides. The implications of these findings will be discussed. colony losses / Varroa / DWV / ABPV / Nosema / pesticides Corresponding author: E. Genersch, [email protected] * Manuscript editor: Marla Spivak ** These authors contributed equally. 1. INTRODUCTION The western honey bee Apis mellifera L. is among the most important productive live- stock due to the role of managed honey bee Article published by EDP Sciences

Transcript of The German bee monitoring project: a long term …The German bee monitoring project: a long term...

Page 1: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

Apidologie Available online at:c© INRA/DIB-AGIB/EDP Sciences, 2010 www.apidologie.orgDOI: 10.1051/apido/2010014

Original article

The German bee monitoring project: a long term studyto understand periodically high winter losses of honey bee

colonies*

Elke Genersch1,**, Werner von der Ohe2,**, Hannes Kaatz3 ,**,Annette Schroeder4, Christoph Otten5, Ralph Buchler6, Stefan Berg7,

Wolfgang Ritter8, Werner Muhlen9, Sebastian Gisder1, Marina Meixner6,Gerhard Liebig4, Peter Rosenkranz4,**

1 Institute for Bee Research, Friedrich-Engels-Str. 32, 16540 Hohen Neuendorf, Germany2 LAVES Institut für Bienenkunde Celle, Herzogin-Eleonore-Allee 5, 29221 Celle, Germany

3 Universität Halle-Wittenberg, Institut für Biologie, Molekulare Ökologie, Hoher Weg 4, 06099 Halle, Germany4 Apicultural State Institute, University of Hohenheim, August-von-Hartmannstrasse 13, 70599 Stuttgart,

Germany5 DLR Fachzentrum Bienen und Imkerei, Im Bannen 38-54, 56727 Mayen, Germany

6 LLH Bieneninstitut Kirchhain, Erlenstr. 9, 35274 Kirchhain, Germany7 LWG Fachzentrum Bienen, An der Steige 15, 97206 Veitshöchheim, Germany

8 CVUA Freiburg, National and OIE Reference Laboratory, Am Moosweiher 2, 79108 Freiburg, Germany9 Landwirtschaftskammer Nordrhein-Westfalen, Aufgabengebiet Bienenkunde, Nevinghoff 40, 48147 Münster,

Germany

Received 18 November 2009 – Revised 17 January 2010 – Accepted 30 January 2010

Abstract – The Western honey bee, Apis mellifera, is the most important animal pollinator in agricultureworldwide providing more than 90% of the commercial pollination services. Due to the development inagriculture the demands for honey bee pollination are steadily increasing stressing the pollination capacityof the global managed honey bee population. Hence, the long-term decline of managed honey bee hives inEurope and North-America is of great concern and stimulated intensive research into the possible factorspresumably causing honey bee colony collapse. We here present a four-year study involving more than1200 bee colonies from about 120 apiaries which were monitored for the entire study period. Bee sampleswere collected twice a year to analyze various pathogenic factors including the ectoparasitic mite Varroa de-structor, fungi (Nosema spec., Ascosphaera apis), the bacterium Paenibacillus larvae, and several viruses.Data on environmental factors, beekeeping management practice, and pesticides were also collected. Alldata were statistically analyzed in respect to the overwintering mortality of the colonies. We can demon-strate for several factors that they are significantly related to the observed winter losses of the monitoredhoney bee colonies: (i) high varroa infestation level, (ii) infection with deformed wing virus (DWV) andacute bee paralysis virus (ABPV) in autumn, (iii) queen age, and (iv) weakness of the colonies in autumn.No effects could be observed for Nosema spec. or pesticides. The implications of these findings will bediscussed.

colony losses / Varroa / DWV / ABPV / Nosema / pesticides

Corresponding author: E. Genersch,[email protected]* Manuscript editor: Marla Spivak** These authors contributed equally.

1. INTRODUCTION

The western honey bee Apis mellifera L.is among the most important productive live-stock due to the role of managed honey bee

Article published by EDP Sciences

Page 2: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

2 E. Genersch et al.

colonies in pollination of many crops, partic-ularly of specialty crops such as nuts, berries,fruits and vegetables. Therefore, the economicvalue of honey production plays only a mi-nor role compared to the economic value ofhoney bees as pollinators in agriculture (Morseand Calderone, 2000). For European crops itwas estimated that 84% of crop species dependat least to some extent upon animal pollina-tion, with honey bees being the most impor-tant animal pollinator (Williams, 1994). How-ever, this figure is misleading since it does nottake into account the importance of the crop toconsumers. As the majority of the world’s sta-ple foods are wind- or passively self-pollinated(wheat, corn, rice), or are vegetatively propa-gated (potatoes), their production does not de-pend on and increase with animal pollinators(insects, birds, and bats). These crops accountfor 65% of global food production, leaving asmuch as 35% depending on pollinating ani-mals (Klein et al., 2007). 90% of commercialpollination services are provided by managedhoney bees, making honey bees the most im-portant commercial pollinator in Europe andworldwide. The demands for agricultural pol-lination are increasing (Aizen et al., 2008)stressing the pollination capacity of the globalmanaged honey bee population. Hence, it isnot surprising that although the global pop-ulation of managed honey bee hives has in-creased ∼45% during the last half century(Aizen and Harder, 2009) the long-term de-clines of managed honey bee hives in the USAand some European countries became an is-sue of widespread interest and concern (Pettisand Delaplane, 2010; Moritz et al., 2010). Asa consequence of these concerns research intothe many factors presumably afflicting honeybees has been intensified in the recent past.The main focus lies on elucidating the role ofpathogens and environmental factors, mainlypesticides, in decreased honey bee vitality andincreased colony losses.

Concerning the role of pathogens, thereis no question that the global health ofhoney bees is at risk, threatened by parasiticmites (Varroa destructor, Acarapis woodi,Tropilaelaps spec.), fungi (Nosema spec., As-cosphaera apis), bacteria (Paenibacillus lar-vae, Melissococcus plutonius), viruses, and

vermin (small hive beetle). The most recentexamples of catastrophic colony losses linkedto – but not fully explained by – pathogenshave been (i) the as yet mysterious ColonyCollapse Disorder (CCD), which resulted inhuge honey bee losses in the USA and else-where (Cox-Foster et al., 2007; Oldroyd, 2007;vanEngelsdorp et al., 2007), as well as (ii)massive colony losses in Spain since 2006attributed to Nosema ceranae (Higes et al.,2008; Higes et al., 2006; Higes, 2010). In ad-dition, honey bees are negatively affected bymany pesticides and fungicides used in agri-culture and the chronic exposure to acaricidesneeded to combat Varroa destructor in api-culture (Barnett et al., 2007; Desneux et al.,2007; Karise, 2007; Moncharmont et al., 2003;vanEngelsdorp et al., 2009a; Johnson et al.,2010).

In the winter 2002/2003, the beekeepersin Germany experienced unusual high winterlosses with about 30% of the German honeybee population reported dead in spring 2003.Losses were not equally distributed among thebeekeepers; instead, the mean of 30% was theresult of many beekeepers that lost 80–100%of their hives on one hand and many that ob-served normal winter losses on the other hand.Similar to CCD, no easy explanation couldbe found for this phenomenon but has beenreported from other European countries sincethen (Potts et al., 2010).

In response to these 2002/2003 winterlosses the German Bee Monitoring Projectwas initiated in autumn 2004. The aim of thisproject was to unravel factors which are re-sponsible for increased colony winter losses.The overall idea was to collect in advancecolony data and samples of bees and hiveproducts from a great number of colonies inorder to use them later for a retrospectiveexplanation of colony mortality. In order tobest achieve this aim, more than 1200 beecolonies from about 120 apiaries (10 coloniesper apiary) were monitored from autumn 2004until now. Lost colonies were replaced withcolonies originating from the same apiary,preferably with nuclei made from the lostcolony in the previous year. Data on the pres-ence of viral, bacterial and fungal pathogens,on varroa infestation level, on the health status

Page 3: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 3

and strength of the colonies at different timesof the year, on mite control regimes, on expo-sure to certain crops, on pesticide residues inrape pollen, and on beekeeping practice werecollected by professional bee inspectors andthe beekeepers themselves. Rape pollen waschosen for pesticide analyses because in Ger-many oil seed rape is the most important nectarand pollen source for honey bees in late spring(Horn, 2009) and a risk for the contaminationof bee products due to the common and in-dispensable application of pesticides (Meixneret al., 2009). Colony mortality was recordedand correlated in a statistical analysis usingmore than 4000 data sets from 2004–2008.The results of this monitoring project will bediscussed in the context of the ongoing colonylosses in Europe and North-America.

2. MATERIAL AND METHODS

2.1. Structure and organizationof the project

In reaction to the unusually high colony lossesin Germany in the winter of 2002/2003 the GermanBee Monitoring Project was established in autumn2004 after several round table discussions involv-ing bee scientists, experts of the German Ministry ofAgriculture, beekeepers, farmers organizations, andrepresentatives of agrochemical companies. Theproject was headed by a project board consisting ofthese partners. Nine scientific bee institutes of dif-ferent Federal States in Germany were responsiblefor the coordination of the field work, the data col-lection, and the supervision of the beekeepers in-volved in the project. Only those data and sampleswhich originated from selected beekeepers and theirmonitored colonies (see below) were included in thestudy and subsequent statistical analyses. Each beeinstitute supervised 6 to 24 beekeepers and farm-ers organisations and the bee inspectors and/or sci-entists were obliged to visit them at least at twoof the three sampling dates each year (autumn sur-vey before wintering in October, spring survey afterwintering in March/ April or summer survey). Dur-ing these visits the inspectors/scientists (i) collecteddetailed information about the previous period, (ii)took samples of bees and hive products from eachof the 10 monitoring colonies, and (iii) estimatedthe population size of the colonies. A single mon-

itoring period lasted from September to August ofthe following year.

2.2. Description of the apiariesand monitoring of the honey beecolonies

At the beginning of the monitoring project,selected beekeepers in Germany were asked to par-ticipate with 10 colonies designated for participa-tion in the German bee monitoring project. Selec-tion of the beekeepers – and, hence, the colonies –aimed at establishing a cohort representing ‘all bee-keeping’ in Germany especially in respect to (i) ge-ographical distribution (Fig. 1), (ii) the number ofmanaged colonies (between 10 and several hundredcolonies, Fig. 2), (iii) scale of beekeeping (hobby-ist, semi-professional and professional beekeepers,Fig. 2), and (iv) the main nectar flow plants in theproximity of the apiary (i.e., the main source of nec-tar and pollen) with special focus on intensive cropslike oil seed rape, sunflowers, and corn which havebeen suspected of causing a negative impact on thehealth of honey bee colonies (Tab. I). The projectstarted in autumn 2004 with 112 beekeepers. In the2005/2006 season already 123 beekeepers provideddata, with 120 beekeepers continuing participationin 2006/2007 and 117 in 2007/2008.

Each participating beekeeper randomly selectedten colonies from his apiary to serve as ‘monitoringcolonies’. If such a colony collapsed in the course ofthe study it was replaced with another colony of thesame apiary, preferably with a nucleus made fromthe collapsed colony in the previous year. The tenselected colonies were managed by the beekeeperlike the other colonies in the apiary and accordingto his/her usual beekeeping practice including mi-gration to specific honey crops, production of nu-clei, requeening, and varroa treatment. This was toensure that the colonies involved in the project re-flected the entire range of variation in types of hivesand management techniques common in and typicalfor Germany.

2.3. Factors analyzed in individualhoney bee colonies

2.3.1. Questionnaires

Prior to the start of the project, participating bee-keepers answered a basic questionnaire concerning

Page 4: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

4 E. Genersch et al.

Table I. Main nectar flow plants in the vicinity of the apiaries as provided by the monitoring beekeepers atthe start of the project.

[%]Main nectar flow plants No Low Middle High Not definedOil seed rape (Brassica napus) 36 11 16 37Sunflower (Helianthus annuus) 90 3 3 4Corn (Zea mais) 44 18 16 22Honey dew 32 25 16 20 7

Figure 1. Geographical distribution of the apiariesof the German monitoring project during the year2005.

the total number of colonies, exact location of theapiary and the colonies (if several sites were used),details of the management system (type of hive, mi-gratory beekeeping, and mode of colony multipli-cation). With additional yearly questionnaires thebeekeepers provided information about their honeyyields, migrations, production of nuclei, requeeningof colonies, varroa treatment(s), abnormal popula-tion dynamics, and visible symptoms of diseases.This information was evaluated and verified as faras possible during the regular visits of the bee in-spectors/scientists.

Figure 2. Size of the beekeeping business (num-ber of colonies, upper diagram) and percentages ofhobby and professional beekeepers among the par-ticipants in the monitoring project (lower diagram).

2.3.2. Record of colony winter losses

During the spring survey the beekeepers pro-vided the number of colonies which collapsed overwinter. A colony was considered dead if (i) no beeswere present any more or (ii) the colony was tooweak to have a chance to recover during spring (ap-proximately less than three bee frames occupied bybees after winter). Colonies which collapsed be-tween April and September were to be recorded but

Page 5: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 5

no such losses were reported during the course ofthe study.

2.3.3. Estimation of the populationdynamics of the monitoringcolonies

The population size of each monitoring colonywas estimated before (October) and after (March /April) wintering, preferably in colonies with littleor no brood, i.e. after (autumn) and before (spring)massive brood rearing. For this purpose, the num-ber of frames covered by bees was counted. Thedefinition of “one frame covered by bees” was de-termined in training courses of the supervisors inorder to standardize this measure as far as possi-ble. In general the following procedure was used:all hives were opened and from two-story hives theupper magazines were tilt forwards. By doing so,all spaces between the combs could be inspectedand the numbers of combs covered by bees wererecorded. Depending on the climatic region, thedate for the spring estimate varied between the api-aries. To avoid overestimating the population size ofthe overwintered colony the population estimationhad to be performed prior to the emergence of thefirst spring brood. Therefore, the last accepted pe-riod for measuring the starting population was the15th week of the year.

The quotient of the population size before andafter the wintering of the colonies were calculatedas “overwintering quotient” and represented a mea-sure of the weakening of the colonies over winterand was used to analyze the effect of oilseed rapeand the amount of pesticides in bee bread on thewintering of honey bee colonies.

2.3.4. Sampling and analysis of adulthoney bees

Samples of about 150 adult bees were taken inOctober, spring and summer. For sampling, a combwith bees was taken from the periphery of the broodnest and the winter cluster, respectively, and thebees were shaken on a piece of plastic wrap andthen put into labelled plastic vials. Samples wereimmediately stored at –20 ◦C until analysis.

The bees were analyzed for the followingpathogens:

Varroa destructor: from the autumn (October)samples, all bees were individually analyzed for

varroa mites and the infestation level was calculatedas ‘number of mites per 100 bees’ and given as ‘per-centage of infestation’. From the analyzed samples,a total of 3589 colony-years with full data sets couldbe used for the statistical analysis.

Nosema spec.: from all spring samples about20 bees were homogenized and after the additionof 2 mL water analyzed microscopically (400X).According to the number of Nosema spec. sporeswithin the visual field positive samples were clas-sified into weakly (<20 spores), medium (20–100 spores) and strongly (>100 spores) infected.In 1868 colonies from 2005 to 2007 autumn sam-ples were also analyzed for Nosema infection andthen used for statistical analysis in relation to win-ter losses.

Honey bee viruses: to limit the costs for theanalysis, only one third of the autumn sampleswere analyzed for five honey bee viruses whichwere considered relevant in respect to colony losses:Kashmir bee virus (KBV), acute bee paralysisvirus (ABPV), sacbrood virus (SBV), deformedwing virus (DWV), and Israeli acute paralysis virus(IAPV). From each bee sample to be analyzed tenbees were taken and decapitated and subsequentlytotal RNA was extracted from the heads for detec-tion of SBV, ABPV, DWV, and IAPV (Siede et al.,2008; Yue and Genersch, 2005). For detection ofKBV and also for IAPV, total RNA was extractedfrom the abdomens. KBV detection was only per-formed in the first three years of the project and thenstopped since very few KBV positive bees weredetected only in 2006. Instead, bee samples from2007 were analyzed for the newly detected IAPVimplicated in colony losses in the US (Cox-Fosteret al., 2007; Maori et al., 2007). RNA extractionwas performed using standard methods (RNeasyKit, Qiagen) as described previously (Genersch,2005; Yue et al., 2006). The rather moderate sam-pling size and the pooling of the bees will not al-low the detection of the odd infected bee in thecolony but will detect an infection level above 20%(Fries et al., 1984) which can be considered biolog-ically relevant. In addition, a recent study demon-strated that analyzing individual bees has no ad-vantage over analyzing pooled bees and that a poolof 20 bees is sufficient to reliably quantify viruslevels in colonies (Highfield et al., 2009). To de-tect viral RNA, one-step RT-PCR was performedaccording to standard protocols (One-step RT.PCRkit; Qiagen) and as previously described (Genersch,2005; Yue et al., 2006). The following tempera-ture scheme was used: 30 min at 50 ◦C, 15 min at

Page 6: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

6 E. Genersch et al.

Table II. Primer sequences used for virus detection.

Length of AnnealingVirus Primer sequence amplicon temperature ReferenceKBV 5’GATGAACGTCGACCTATTGA 3’ 414 bp 50.5 (Stoltz et al., 1995)

5’TGTGGGTTGGCTATGAGTCA 3’ABPV 5’CATATTGGCGAGCCACTATG 3’ 398 bp 49.5 (Bakonyi et al., 2002)

5’CCACTTCCACACAACTATCG 3’DWV 5’CCTGCTAATCAACAAGGACCTGG 3’ 355 bp 52.0 (Genersch, 2005)

5’CAGAACCAATGTCTAACGCTAACCC 3’SBV 5’GTGGCAGTGTCAGATAATCC 3’ 816 bp 52.0 (Yue et al., 2006)

5’GTCAGAGAATGCGTAGTTCC 3’IAPV 5’GAGCGTCGATCCCCCGTATGG 3’ 524 bp 55.0 (Maori et al., 2007)

5’TCCATTACCACTGCTCCGACAC 3’

95 ◦C followed by 35 cycles with 30 s at 94 ◦C,30 s at the appropriate annealing temperature (seeTab. II), 30 s at 72 ◦C followed by a final elonga-tion step for 10 min at 72 ◦C. PCR products (5 μLper reaction) were analyzed on a 1.0% agarose gel.The agarose gel was stained with ethidium bro-mide and visualized by UV light. Correlation of theelectrophoretic mobility of the amplicons with theexpected size (Tab. II) was interpreted as specificdetection. Specificity of the amplicons was furtherverified by sequencing (Medigenomix) random am-plicons.

During the first two project years, also Americanfoulbrood (Paenibacillus larvae) and tracheal mites(Acarapis woodi) were analyzed following the OIEguidelines and protocols given by the GermanNational Reference Laboratory for Bee Diseases(Freiburg). However, tracheal mites were never ob-served and AFB only rarely detected. Therefore,these analyses were not continued in order to saveresources.

2.4. Residue analysis

Samples of bee bread (appr. 10 × 10 cm) col-lected after the blooming period of oilseed rape(Brassica napus) were used for residue analysis.Oilseed rape provides a huge amount of nectar andpollen for honey bees in Germany but also rep-resents a source for contamination with pesticidesthrough seed dressing and spray application duringthe blooming period. Because most pesticides arelipophilic, pollen is considered the best matrix formeasuring the exposure of a honey bee colony topesticides. In the years 2005 and 2006, fifty apiarieswere selected for residue analysis based on the mi-croscopic pollen analysis of the honey. Only those

which revealed a high input of rape nectar were con-sidered suitable due to a potential exposure to pesti-cides. In 2007, bee bread samples of nearly all api-aries (n = 110) were analyzed.

All samples were split in two parts, one forthe pollen analysis and one for the residue analy-sis. For the chemical analyses a multi-method (LC-MS/MS, GC-MS) was adapted which allowed thedetection and quantification of 258 active ingredi-ents. 5 g beebread were extracted with acetonitrile.After removal of fat and remaining proteins by cool-ing to –20 ◦C overnight, solvent was cleaned-upusing gel-permeation-chromatography (GPC). Theextract was further cleaned by SPE cartridges con-taining C18, aminopropyl and graphitized carbonblack. The final extract was analyzed by GC-MSand LC-MS/MS for 258 pesticides and pesticidemetabolites. The limits of quantification were be-tween 3 and 10 μg/kg, in a few cases 15 μg/kg. Forall neonicotinoids the limits of detection were at thelevel of 1 μg/kg.

2.5. Effect of locations with accessto oilseed rape

In 2006, we analyzed a possible effect of oilseedrape on the wintering of honey bee colonies. Theamount of rape pollen was determined in bulk sam-ples of honeys from the first harvest from apiarieswith different access to oilseed rape (n = 142).The overwintering quotient from October 2006to March/April 2007 was calculated for the sam-pled colonies and correlated with the amount ofrape pollen as determined by microscopic honeyanalysis.

Page 7: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 7

Table III. Level of Varroa destructor infestation – as determined from adult bee samples collected in au-tumn – in all colonies and given separately for those colonies which survived or collapsed in the subsequentwinter.

Infestation level in % ± sd

2004 2005 2006 2007 Σ n P-value1

All colonies 3.1 ± 5.8 4.7 ± 8.9 4.4 ± 8.6 5.1 ± 8.5(N) (315) (1065) (1092) (1117) (3589)

Surviving colonies 3.1 ± 5.9 3.2 ± 5.9 3.5 ± 6.7 3.6 ± 6.4(N) (311) (927) (1013) (966) (3217)

Collapsed colonies 1.7 ± 2.0 14.6 ± 17.0 16.5 ± 18.0 14.8 ± 13.5 < 0.000001(N) (4) (138) (79) (151) (372)

1 Varroa infestation in October was compared using Mann-Whitney U-test.

2.6. Data evaluation and statisticalanalysis

All data were fed into a central database pro-grammed especially for this project. Each data setrepresents the complete parameters for one colonyin one year (i.e. colony parameters of the autumn,spring and summer survey including colony size, allbeekeeping practices such as type, date and num-ber of Varroa treatments, migratory beekeeping,honey yield, colony management, the data from thequestionnaire (see Sect. 2.3.1), and all laboratorydata including honey, pollen and pathogen analy-sis). For statistical analysis, only complete data setswere considered. In addition, these complete datasets had to pass a plausibility check. Finally, of 5198colony-level data sets, only 4313 could be used forthe statistical analysis.

Some parameters like Nosema spec. in autumnand honey bee viruses were not analyzed in allcolonies in each year and, therefore, for these pa-rameters only a reduced number of data sets couldbe analyzed. The exact number of data sets used foreach parameter is given in the results (Tabs. III–VI).

We compared surviving and collapsed coloniesemploying nonparametric tests because the basalassumptions of parametric tests (i.e. normality andconstant variance) were not satisfied. All statisti-cal analyses were carried out using Statistica 6.0.Kruskal-Wallis- and median-test proved that therewere no significant differences in the distribution ofcolony losses between the years (P > 0.05). There-fore, the data sets of the four years were analyzedtogether. For each parameter given in Tables III–VInon parametric Mann-Whitney U-tests (varroa in-festation rates) and chi2 tests (honey bee viruses,Nosema infection, beekeeping management) were

performed by comparing the survival rates in in-fested colonies with those in non-infested colonies.A P-value < 0.05 was considered significant.

3. RESULTS

3.1. Colony winter losses

It has to be mentioned that the monitoringbeekeepers in general reported a satisfactorydevelopment of their colonies during the beeseasons of the study period. This is supportedby average honey yields per colony of 39.5 kgin 2005, 49.0 kg in 2006 and 46.3 kg in 2007.Of the 4393 colonies included in the analysisover the four year period, 504 colonies diedover winter but most did not show the de-scribed symptoms of colony collapse disorder(CCD) (vanEngelsdorp et al., 2007). In 80 ofthe 504 dead colonies the reason for the col-lapse could clearly be explained by queen loss(50), food shortage (17), crime (12) and AFB(1). These colonies were not considered in thefollowing statistical analysis. Therefore, to un-ravel the reasons for inexplicable or not easilyexplainable winter losses 3889 surviving and424 dead colonies were used for further analy-sis.

The average percentage of winter lossesranged from 3.8% (2004/05) to 15.2% in2005/06 (Fig. 3). However, the losses werenot distributed equally among the participat-ing beekeepers. An analysis of all data setsof the four years showed that the majorityof the beekeepers had no or only moderate

Page 8: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

8 E. Genersch et al.

Table IV. Incidence of Nosema spec. and honey bee viruses in the adult bee samples from autumn.

2004 2005 2006 2007 Σ n

Noseman 164 688 1072 1924positive % 31.1 21.4 13.8strong infection % 1.2 0.7 2.4

ABPVn 182 276 296 350 1104positive % 8.8 5.8 6.4 11.7

SBVn 182 276 296 350 1104positive % 15.4 9.8 5.4 7.4

DWVn 182 276 296 350 1104positive % 4.4 11.2 20.6 33.4

KBVn 182 218 196 - 596positive % 0.0 0.0 1.0 -

IAPVn - - - 341 341positive % - - - 0.0

Figure 3. Proportion of lost colonies among themonitored colonies of the beekeepers participatingin the project during the four years of the monitor-ing phase. For each year, the total number of lostcolonies was calculated as proportion of all coloniesparticipating in the project.

colony losses during the project period andonly in 14.2% of the analyzed cases the losseswere higher than 20% (Fig. 4). This distri-bution – many beekeepers with no or fewcolony losses and few beekeepers with highlosses – were similar for all four winters asconfirmed by Kruskal-Wallis- and median-test(P > 0.05). In addition to annual variationsin winter losses regional variations were alsoobserved, however, these regional differenceswere not consistent over the four years period.In addition, higher colony losses were not con-sistently related to certain beekeepers and api-aries.

Figure 4. Average distribution of the colony lossesduring the winters 2004/ 05 – 2007/ 08 (n = 436beekeeper). One data set represents one beekeeperwith 10 colonies.

3.2. Effect of pathogens and parasiteson colony winter losses

The varroa infestation rates in October aregiven in Table III. The infestation rate forVarroa destructor ranged from 3.1 in 2004to 5.1 in 2007. A statistically highly signifi-cant (P < 0.000001) difference between thevarroa infestation rate of surviving coloniesand of colonies which collapsed over wintercould be demonstrated. Summarizing all datasets, the average varroa infestation in survivingcolonies (av. ± s.e.: 3.4±0.1) was significantlylower than in lost colonies (av. ± s.e.: 15.1 ±

Page 9: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 9

Figure 5. Relation between colony mortality inwinter and varroa infestation level calculated asnumber of mites per 100 bees.

0.7). Even if one considers the huge stan-dard deviation of the individual annual val-ues the damage rate of some colonies mighthave exceeded the damage threshold level ofmore than 10 mites per 100 bees in autumn(Liebig, 2001) (Fig. 5). Plotting colony mor-tality against ‘varroa mites per 100 bees’ re-vealed that with as little as 10 mites per 100bees around 20% of the colonies were proneto collapse over winter and an average of50% mortality could be expected if more than20 mites per 100 bees were present in a colonyin autumn (Fig. 5). However, some of thecolonies with high infestation rates of morethan 30% in October were able to survive thewinter (Fig. 5). Colony mortality and varroainfestation levels were significantly correlated(Spearman rank order correlation r = 0.996,P < 0.00001).

The prevalence of Nosema infection in au-tumn varied between ∼31% Nosema posi-tive colonies in 2005 (with a relatively lownumber of colonies) and less than 14% in2007 (Tab. IV). Most of the Nosema infectedcolonies did only reveal weak infection levels;as few as 0.7% to 2.4% of the colonies wereconsidered strongly infected.

Bee samples were qualitatively analyzed forthe presence of DWV, ABPV, SBV KBV, andIAPV (Tab. IV). While KBV and IAPV couldrarely (1.0%) or never be detected in any ofthe samples, respectively, DWV, ABPV, andSBV were more prevalent. The incidence ofthese three viruses varied independently from

year to year. The incidence of ABPV variedbetween 5.8% in autumn 2005 and 11.7% inautumn 2007. The proportion of SBV positivesamples differed between 5.4% in 2006 and15.4% in 2004, while autumn samples test-ing positive for DWV varied between 4.4% in2004 and 33.4% in 2007. In general, the season2007/2008 showed the highest level of viral in-fections in the sampled and analyzed colonies.

In addition to the significant relationshipbetween varroa infestation level and colonywinter losses, the occurrence of DWV andABPV was significantly higher in lost than insurviving colonies (Tab. V). No effect could beproven for SBV and Nosema spec. (Tab. V).Analysis of the viral infection status of surviv-ing and collapsed colonies with chi2-tests re-vealed that the presence of DWV in autumnwas related with a surprisingly high signifi-cance (P = 0.00001) with winter losses. Inother words, colonies which contained clini-cally infected bees (DWV viral RNA in totalhead RNA) in autumn had a lower chance tosurvive winter than colonies which tested neg-ative for DWV. A similar relationship could bedemonstrated for ABPV. Again, a significantrelation between ABPV infection detected inautumn and colony collapse over the follow-ing winter could be established (P = 0.0039).

3.3. Effect of beekeeping managementon colony winter losses

No effect could be confirmed for the mate-rial of the hive (wood vs. Styrofoam) and the“starting condition”, respectively (Tab. VI).The latter refers to the common practiceamong German beekeepers to establish newcolonies at the end of the season in orderto have stronger and/or healthier colonies foroverwintering. These new colonies could beestablished as nuclei (n) or could be composedof nuclei and “old” colonies (o) in differentcombinations (Tab. VI). A comparison of allpossible colony types (n, o, n+n, o+o, n+o) didnot reveal any statistical significance (Chi2,df = 4; 1.75; P = 0.78).

A clear significant effect, however, wasproven for the age of the queen: colonieswhich survived the winter had on average

Page 10: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

10 E. Genersch et al.

Table V. Effects of pathogen infection and parasite infestation in October on winter losses of honey bees.

TotalNo. of P-value

Factor colonies analyzed No. of survived colonies No. of collapsed colonies (chi2)total pathogen pathogen total pathogen pathogen

positive negative positive negativeDWV 1104 995 173 822 109 44 65 0.00001ABPV 1104 995 75 920 109 17 92 0.0039KBV 596 543 2 541 53 0 53 0.658SBV 1104 995 99 896 109 6 103 0.202Nosema spec. 1924 1744 317 1427 180 29 151 0.492

Table VI. Effects of beekeeping management on winter losses of honeybee colonies.

Total Total TotalNo. of No. of No. of

Factor analyzed colonies surviving colonies collapsed colonies P-valueType of beehive 4313 3889 424

0.94 (chi2)wood / styrofoam wood / styrofoam(2594) / (1295) (282) / (142)

Starting condition1 4293 3876 4170.78 (chi2)O N C O N C

(2731) (724) (421) (317) (65) (35)Queen age years (n) 4021 3639 382

0 1 2 3 4 0 1 2 3 4 0.0052 (chi2)(2002) (1238) (192) (5) (2) (156) (181) (45) (0) (0)

Colony strength in October 4313 3889 424 < 0.000001 (t-test)(frames with bees ± sd) 12.3 ± 5.1 10.0 ± 5.4

1 Old colony from previous year (O), newly formed colony during summer season (N), or combined colonies(C) either o + o, o + n, n + n.

significantly younger queens compared to thecolonies which collapsed during winter (chi2,P < 0.000001, Tab. VI). In other words: youngqueens lowered the risk for a colony to col-lapse during the winter independently from theabove mentioned starting conditions. A cleareffect on the overwintering success was alsoconfirmed for the colony strength (= num-ber of bees) in October: the 3889 survivingcolonies during the four year period occupiedon average 12.3 ± 5.1 bee spaces compared toonly 10.0 ± 5.4 in the dead colonies. Thesedifferences were highly significant (chi2, P <0.000001, Tab. VI).

3.4. Effect of oilseed rapeon the overwintering quotient

The correlation analysis reveals a positivebut not significant correlation between the

amount of rape pollen in the honeys harvestedin summer and the overwintering coefficient(Fig. 6). Therefore, the hypothesis could notbe verified that intensive contact of honey beecolonies to oilseed rape has a negative influ-ence on overwintering.

3.5. Residue analysis of pesticides in beebread

42 active ingredients have been detected inthe 105 analyzed samples of 2005 and 2006(Tab. VII). In many positive samples more thanone substance could be found. Only 25 sam-ples did not reveal any measurable contam-ination (below limit of detection). The mostabundant active substances were coumaphos(46, varroa treatment), boscalid (35, fungi-cide) and terbuthylazine (32, herbicide). Somesamples showed quite high residue amounts

Page 11: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 11

amount of rape pollen [%]

10 20 30 40 50 60 70 80 90 100

ove

rwin

teri

ng

co

effi

cien

t

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.00

Figure 6. Relation between amounts of rape pollenin honey harvested in summer 2006 and the over-wintering coefficient of the colonies in the subse-quent winter 2006/2007. Statistical analysis of thesedata according to Pearson revealed no correlation(r = 0.173, P = 0.12).

(the herbizide azoxystrobin, and the fungizidetolylfluanid, for instance). However, these highresidue amounts did not correlate with poorcolony development. There was no signifi-cant difference in overwintering quotient be-tween apiaries with no pesticide residues inthe bee bread and those with higher amountsof residues (more than 10 μg/kg of at leastone substance; chi2P = 0.999; F-Test P =0.938; n = 40, bee bread 2006, overwinter-ing 2006–2007). The most prevalent insec-ticide was thiacloprid (9, max. 199 μg/kg).Other detected insecticides were dimethoate(3 samples), azetamiprid (2), pirimicarb (2),tau-fluvalinate (2), and Lambda-cyhalothrin(1). The amounts of the active substances inthese cases were below 10 μg/kg, except fordimethoate (20 μg/kg).

The results of the 110 bee bread samplesfrom 2007 did not reveal striking differencesto the results of the samples 2005 and 2006in terms of the percentage of positive samplesand the amounts of active ingredients. 42 ac-tive substances have been detected between 1and 67 times in the 110 samples (Tab. VII).The main substances were again coumaphos(33 times), boscalid (67 times), thiacloprid (62times), and terbuthylazine (48 times).

Due to their high toxicity to bees the neoni-cotinoids were of particular interest. However,clothianidin was not detected and imidaclopridwas detected only once (3 μg/kg) in the 215samples collected from 2005–2007.

4. DISCUSSION

For several reasons the German bee moni-toring project represents a worldwide uniqueapproach for the analysis of unusually highwinter losses:

1. The project was established as a close co-operation between beekeepers and bee sci-entists enabling the monitoring of coloniesmanaged by ‘normal’ beekeepers who con-tinued to practice their usual beekeepingroutine.

2. To reflect the beekeeping situation in all ofGermany in respect to regional peculiari-ties, the participating apiaries (about 120)with more than 1200 colonies were dis-tributed nationwide.

3. A long-term ongoing study was imple-mented to observe annual variations.

4. Data on bee pathology, on residue analysisof bee bread and information on the envi-ronmental conditions at the site of the api-ary and beekeeping management practiceswere recorded to be analyzed in relation tocolony winter losses.

Such a project requires an enormous effort forgeneral coordination, for the supervision of theparticipating beekeepers, and for the mainte-nance of the central data base. However, thecooperation with the beekeepers worked outquite well and without major conflicts whichis confirmed by the remarkable low fluctua-tion of less than 5% of the participants overthe years. This was and is a prerequisite for theplanned long term continuation of the moni-toring project. The overall aims of this projectwere (i) to analyze the occurrence of periodi-cally high winter losses on the basis of verifi-able data and (ii) to correlate such losses withthe factors measured within the project.

4.1. Colony losses

Periodic colony winter losses of 30% andmore have been reported in Germany alreadyfor more than 50 years (Gnädinger, 1984)and recently from other countries (Ellis et al.,2010; Giray et al., 2010; vanEngelsdorp et al.,2008). However, during the four winters from

Page 12: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

12 E. Genersch et al.

Table VII. : Number of pollen samples of the years 2005 and 2006 (total n = 105) and the year 2007 (totaln = 110) which tested positive for the analyzed pesticides.

Active ingredient 2005 and 2006 2007Insecticides / Acetamipride 2 0Acaricides Bromopropylate 8 4

Clofentezine 1 0Coumaphos 46 33Dimethoate 3 3Fenpyroximate 0 2Flufenoxurone 0 1Imidacloprid 0 1Indoxacarbe 0 1Lambda-cyhalothrine 1 2Methiocarbe 0 22Methoxyfenozide 0 5Pirimicarb-desmethyl 0 7Pirimicarbe 2 3Tau-fluvalinate 2 4Tebufenozide 2 4Tebufenpyrad 0 1Thiacloprid 9 62

Fungicides Azoxystrobin 10 12Bitertanol 2 0Boscalid 35 67Carbendazime 6 7Cymoxanile 3 0Cyproconazole 4 0Cyprodinile 11 0Difenconazol 1 3Dimethomorph 3 0Diphenylamine 0 2Epoxiconazole 0 1Fenpropimorph 1 7Fludioxonil 8 13Flusilazole 2 3Iprodione 2 1Iprovalicarb 1 1Kresoxim-methyl 0 4Metalaxyl 4 0Myclobutanil 5 3Penconazol 0 1Pyraclostrobine 2 10Pyrimethanil 0 6Tebuconazole 12 3Tolyfluanid 4 2Triadimenol 1 0Trifloxystrobine 3 0Vinclozolin 0 1

Page 13: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 13

Table VII. Continued.

Active ingredient 2005 and 2006 2007Herbicides Chloridazone 5 3

Ethofumesate 3 3Isoproturone 6 25Metamitrone 1 4Metobromuron 1 0Metolachlore 7 15Metoxurone 2 0Metribuzine 1 0Pendimethaline 1 0Prosulfocarb 2 18Terbuthylazine 32 44

2004/ 05 until 2007/ 08 the mean winter lossesof all monitoring beekeepers ranged only be-tween 4% and 15%, but regionally higherlosses were reported. An advantage of thecooperation with experienced beekeepers isthat fundamental mistakes in the managementpractice should be rare and, therefore, unex-pected colony losses may not be the result ofbeekeeping mismanagement. This assumptionis supported by the fact that only 12 of 4393died from starvation during winter.

Colony winter losses were not equally dis-tributed among the participating beekeepers.Combining the datasets of all four years, onlyin 14.2% of the cases the loss rates were higherthan 20%. This also means that the majorityof the beekeepers had little or only moderatelosses. However, as higher colony losses werenot consistently related to certain beekeepersand apiaries, respectively, we can state that thecolony losses during our monitoring projectcannot exclusively be explained by the factor‘beekeeper’.

4.2. Reasons for colony losses

The German Bee monitoring project pro-vided statistical evidence that certain factorsare involved in causing winter losses of beecolonies. The identified factors were (i) highmite infestation levels, (ii) clinically relevantDWV infections in autumn, (iii) ABPV in-fections in autumn, (iv) old queens, and (v)relative colony weakness before overwinter-ing. The main cause of overwintering prob-

lems was undoubtedly infestation with the ec-toparasitic mite Varroa destructor followed byviral infections. Bee colonies exhibiting oneor more of the above mentioned factors mighthave a rather small chance to survive winteraccording to the results of the project.

The infestation with Varroa destructor infall clearly revealed the highest relation withwinter losses of honey bee colonies. As a mea-sure for the infestation rate we used “var-roa mites per 100 bees” in October. At thattime honey bee colonies in Germany alreadyhave produced their winter bee population andusually have little or no brood. It is believedthat varroa damages at the colony level al-ready are engendered by the infestation dur-ing late summer, when the host populationdeclines, the relative varroa parasitization in-creases and consequently the production ofhealthy long-living winter bees is negativelyimpacted (Amdam et al., 2004; Fries et al.,1994). However, the infestation rate of the“October bees” obviously represents an ex-cellent measure to predict the risk of colonywinter losses. For infestation rates between0 and 20% a nearly exponential increase ofthe winter losses can be observed. Figure 5indicates a threshold of 6% for the infesta-tion rate to keep the average colony losses be-low 10%, which is considered an acceptablecolony loss rate for winter. This confirms re-cent field studies (Liebig, 2001) showing thatinfestation rates of the winter bees of morethan 7% were critical for the winter survivalof colonies under German conditions. In the

Page 14: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

14 E. Genersch et al.

USA (Delaplane and Hood, 1999) and Canada(Currie and Gatien, 2006) infestation rates of10% and even more were suggested as thresh-olds for economic damages. Our results in-dicate that these thresholds may be too high,at least for German conditions. Our survivingcolonies had average infestation rates of ap-proximately 3% whereas the infestation of col-lapsed colonies was 5 fold higher, on average,with a huge variation (Tab. III). Furthermore,it is likely that more highly infested colonieswhich did not collapse exhibit sublethal dam-ages which may affect the spring developmentafter overwintering.

Surprisingly, winter mortality only in-creased from ∼55% to ∼65% for colonieswith infestation rates ranging from 30% to80% (Fig. 5). This contradicts field experi-ments where non-treated colonies with infesta-tion rates of more than 30% during summer donot have a chance to survive the following win-ter (Fries et al., 2003; Rosenkranz et al., 2006)on the one hand, but, on the other hand, figuressimilar to those obtained in the current studyhave also been observed in managed coloniesin cold climate (Strange and Sheppard, 2001).The surviving monitoring colonies with highvarroa infestation may, therefore, be victims ofreinvasion after the production of winter bees(Goodwin et al., 2006; Greatti et al., 1992;Renz and Rosenkranz, 2001) or may have beenfree of secondary infection.

Additionally, we have to consider thatnearly all of these colonies had been treatedat minimum once against varroa during sum-mer, usually with formic acid. Our results alsoclearly demonstrate that the varroa treatmentsperformed so far are not sufficiently effectiveto ensure a regionwide reduction of the varroainfestation and, therefore, minimize the risk ofreinvasion.

We can, therefore, state that Varroa destruc-tor still represents the major threat for thewinter survival of honey bee colonies in Ger-many. Varroa as a main factor for winter losseshave also been proven in recent surveys fromEurope (Topolska et al., 2008) and the USA(vanEngelsdorp et al., 2008). The infestationof the “October bees” could be used in exten-sion work to predict the chance for survivalof colonies before winter. Varroa treatments

should be performed in a way to achieve a beeinfestation rate in autumn of less than 5%.

In the literature several studies determiningthe viral infection status of honey bee coloniescan be found (Baker and Schroeder, 2008;Berenyi et al., 2006; Tentcheva et al., 2004).The general picture emerging from these stud-ies is that DWV is the most prevalent virus inEurope with more than 90% of the analyzedcolonies being infected. The same seems to betrue for Germany (Yue and Genersch, 2005)although no epidemiological study has beenperformed so far. Unfortunately, DWV inci-dence of 90–100% in all colonies regardlessof whether they are strong, weak or collaps-ing does not allow correlating DWV infectionwith colony losses since the mere presenceof DWV in otherwise healthy bees is obvi-ously of no clinical relevance (de Mirandaand Fries, 2008; Yue et al., 2007). Recently itwas shown that the detection of DWV RNAin total RNA extracted from bee heads cor-related with clinical symptoms like crippledwings (Yue and Genersch, 2005). However,a small proportion of seemingly healthy beesalso show DWV infection in the head (Yueet al., 2007) which is interesting in the con-text of a recent study involving experimentalDWV infection of adult bees. Injecting DWVinto the hemolymph caused learning deficitspointing to neurological symptoms being asso-ciated with DWV infections (Iqbal and Müller,2007). Such learning deficits might affect thefitness of individual bees and, hence, colonyperformance. We therefore chose to diagnoseDWV infections by using extracts from beeheads rather than from entire bees. Not surpris-ingly, we found a rather low rate of DWV pos-itive colonies compared with previous studies(Baker and Schroeder, 2008; Berenyi et al.,2006; Tentcheva et al., 2004) ranging between4.4% in autumn 2004 and 33.4% in autumn2007. Statistically relating these data with theobserved winter losses revealed a highly sig-nificant (P = 0.00001) negative effect of DWVinfection (characterized by viral detection in‘head’) in autumn on winter survival. Mostof the collapsed colonies also had high var-roa infestation levels confirming the strong as-sociation of clinical DWV infections with V.destructor infestation (Ball, 1983, 1989; Ball

Page 15: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 15

and Allen, 1988; Bowen-Walker et al., 1999;Gisder et al., 2009; Martin, 2001; Shen et al.,2005; Yang and Cox-Foster, 2005; Yue andGenersch, 2005). However, a few collapsedcolonies with DWV showed no or low miteinfestation levels (data not shown). This sug-gests that DWV can contribute to colony col-lapse even in the absence of V. destructor asalso implicated by a recent study (Highfieldet al., 2009). Another possibility is that a clin-ical DWV infection of the colony initiated byhigh mite infestation rates during spring andsummer can cause colony collapse in wintereven though V. destructor had been success-fully eliminated during late summer and au-tumn treatments.

ABPV infection in autumn was also signif-icantly related (P = 0.0039) to colony col-lapse in the following winter confirming theresults of a recent study from a small regionin Germany (Siede et al., 2008). In this studyvirus detection in extracts from entire bees wascompared with detection in head extracts. Itwas shown that both methods were equallymeaningful although ABPV detection in totalRNA from head produced slightly more sig-nificant results (Siede et al., 2008). Again, al-though most of the collapsed colonies withABPV showed high mite infestation levels, afew of these colonies had low infestation lev-els. The same explanations as already outlinedfor DWV might also hold true for some ABPVassociated colony collapses: Either ABPV cansometimes cause colony collapse even in theabsence of V. destructor or eliminating V. de-structor once a fatal ABPV infection has al-ready been initiated in the colony does nolonger change the fate of the colony.

No negative effect of SBV or KBV onwinter survival could be demonstrated in ourstudy. The rates of infection were consistentwith other studies (Baker and Schroeder, 2008;Berenyi et al., 2006; Tentcheva et al., 2004).For detection of SBV again only RNA ex-tracted from bees’ heads has been used as op-posed to whole bee extracts commonly usedin the literature. SBV is a brood pathogenbut persistence in the hypopharyngeal glandsof adult bees drives virus transmission in thecolony (Bailey and Ball, 1991). Due to thistissue tropism of SBV, using head extracts of

adult bees for SBV diagnosis was the methodof choice for increasing detection sensitivity.KBV was only detected in a few colonies. Thisis in accordance with the worldwide distribu-tion of KBV which is most prevalent in NorthAmerica and New Zealand but less prevalentin Europe (de Miranda et al., 2010) whereABPV is most prevalent (de Miranda et al.,2010). In addition, KBV was also the leastprevalent virus found in other European stud-ies on the incidence of bee viruses in diseasedand healthy colonies (Baker and Schroeder,2008; Berenyi et al., 2006; Tentcheva et al.,2004). For both SBV and KBV the miss-ing correlation with colony collapse is in ac-cordance with previous studies revealing arather low incidence of these two viruses andeven a higher prevalence in healthy colonieswhen compared to weak or collapsing colonies(Berenyi et al., 2006; Tentcheva et al., 2004).

In summary, a clear relation between win-ter losses and viral infection in autumn couldonly be established for DWV and ABPV.ABPV can be considered a member of thehighly virulent ABPV-KBV-IPAV virus com-plex (de Miranda et al., 2010) with all mem-bers being extremely virulent when injectedinto pupae or adults (Bailey and Ball, 1991;Bailey et al., 1963). It is therefore not sur-prising that ABPV as well as IAPV are impli-cated in colony losses (Cox-Foster et al., 2007;Siede et al., 2008) and that ABPV was in-volved in winter losses of monitoring coloniesof the study at hand. So far, DWV has beenimplicated in colony losses in only one study(Highfield et al., 2009). Other studies mighthave missed this relation because if 90–100%of the colonies are diagnosed as DWV-positivebut only 10–30% of the colonies collapse sta-tistical tests will not reveal a relation betweenDWV infection and colony collapse. There-fore, it is important to differentiate betweenclinically irrelevant and clinically relevant in-fections of colonies and to detect only thosecolonies which carry a clinically relevant in-fection. This can be achieved by quantify-ing DWV loads in asymptomatic bees andcolonies since DWV loads exceeding 1 ×108 copies per asymptomatic worker bee inwinter seem to be fatal for the colony evenin the absence of high mite infestation levels

Page 16: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

16 E. Genersch et al.

(Highfield et al., 2009) or by considering thedifferences in DWV tissue tropism betweenovertly and covertly infected bees (de Mirandaand Genersch, 2010; Gisder et al., 2009; Yueand Genersch, 2005) and restricting DWV di-agnosis to total RNA extracted from head asdone in the study at hand.

Recently, infections with Nosema ceranaeleading to an usual form of nosemosis havebeen implicated in severe colony losses inSpain and it was suggested that this unusualform of nosemosis is the main cause of inex-plicable colony losses and CCD-like phenom-ena in Europe if not worldwide (Higes et al.,2006–2009; Martin-Hernandez et al., 2007)due to the high virulence of Nosema ceranaeand its exceptional biotic potential even athigher temperatures (Martin-Hernandez et al.,2009). These assumptions are in contrast toseveral other studies identifying IAPV as reli-able marker for CCD (Cox-Foster et al., 2007)or showing that CCD symptoms can be re-duced by anti-viral treatment (Maori et al.,2009) or ruling out Nosema spec. as cause forcolony losses (Chauzat et al., 2007; Johnsonet al., 2009; vanEngelsdorp et al., 2009b; Chenand Huang, 2010). Likewise, the results ob-tained with the German bee monitoring projectdid not reveal any relation between infectionwith Nosema spec. and winter losses althoughboth Nosema species are prevalent in Germany(Klee et al., 2007). Since no losses occurredduring summer although colonies were in-fected by Nosema spec. it can also be ruledout that infection with Nosema spec. killedcolonies between spring and autumn as de-scribed in the Spanish studies (Higes et al.,2008; Martin-Hernandez et al., 2007). A weakpoint of the study at hand is that the differen-tiation between the Nosema species has beenperformed only sporadically and, therefore,could not be included in the statistical anal-yses. Nevertheless, colony losses caused byNosema ceranae would not have been maskedby this approach and, therefore, the interpre-tation that Nosema spec. did not cause colonylosses in Germany during the study period isvalid.

Another factor which could be significantlyrelated to winter losses was the age of thequeen heading the monitored colony. For the

first time we could demonstrate that coloniesheaded by young queens have a significantlyhigher chance to survive the winter comparedto colonies with older queens. A possible rea-son for this queen-age-effect could be a sig-nificantly higher brood and bee production incolonies with young queens accompanied bya lower infestation with varroa mites (Akyolet al., 2007). However, the detailed reasonsfor the higher vitality of colonies headed byyounger queens remain elusive.

The analysis of pesticide residues in pollen(bee bread) as performed in the course of theGerman bee monitoring project was the firstsuch screening in Germany. As expected, theresults show that pollen is contaminated witha plethora of chemical substances originatingfrom the agricultural practice of using pesti-cides but also from the apicultural necessityof using acaricides. During bloom of oilseedrape many pesticides are used and, hence,they can be detected in many pollen samples.Likewise, pollen samples from apiaries in re-gions with intensive cultivation of rape showeda higher contamination level. It is generallyassumed that although individual substancesmight not have a negative effect on individ-ual bees and colonies (i.e. non-toxic for bees),the simultaneous contamination of pollen withseveral agrochemical substances will havea negative effect on larvae or nurse beesconsuming such multi-contaminated pollen.These presumed sublethal effects than nega-tively influence colony development eventu-ally leading to colony collapse. The contam-inations identified in bee bread in the courseof the German bee monitoring project wereindeed mainly substances which are consid-ered non-toxic for bees. In addition, the ob-served amounts of the residues were quitelow, i.e. three orders of magnitude lowerthan the respective LD50 (http://sitem.herts.ac.uk/aeru/footprint/en/index.htm). Accordingly,no relation between contamination of pollenand colony development or winter lossescould be demonstrated in the course of theproject although special emphasis was putinto this aspect. Still, further investigationsand controlled experiments with improvedmethodology (Pham-Delègue et al., 2002) areundoubtedly necessary because several studies

Page 17: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 17

did prove negative effects of pesticides onhoney bees (Decourtye et al., 2003, 2004;Moncharmont et al., 2003; Johnson et al.,2010).

4.3. Conclusion

A panel of factors have been analyzedfor their role in winter losses of honey beecolonies in Germany. Among all these factors,infestation with Varroa destructor turned outto play the key role. Based on the results pre-sented it is safe to state that Varroa destructoris the dominant killer of honey bee coloniesduring winter. In addition to high varroa in-festation levels, DWV and ABPV infectionsin autumn significantly lower the winter sur-vival of honey bee colonies as do old queensheading overwintering colonies. That a weakcolony has not the best chance to survive thewinter is rather trivial but the fact that we ob-served such winter losses due to colony weak-ness shows that beekeepers still winter weakcolonies. It is safe to assume that these identi-fied factors are not specific for winter losses inGermany but that these results have wider im-plications. Varroa destructor, viral infections,old queens, colony weakness for sure are alsoresponsible for winter losses in many otherEuropean regions and may be even in partsof North-America. This does not rule out thatfrom year to year other, additional factors alsoplay a role in colony losses and that the rea-sons for the periodically occurring, unusuallyhigh winter losses of more than 30% are differ-ent from what we observed during the last fiveyears with rather normal winter losses. Thecontinuation of the project is important to al-low the generation of a database that can pro-vide an explanation for winter losses using sta-tistical evidence.

A negative effect of pesticide residues inbee bread from spring on the survival of thebee colonies in the subsequent winter couldnot be proven, however, our approach wasnot proposed to record sublethal and chroniceffects of multiply contaminated pollen. Forsuch issues, more extensive sampling proce-dures and enhanced methods are required.

From the results of this study we candeduce a general recommendation for bee-keepers who want to successfully bring theircolonies through the winter season: an effec-tive treatment against Varroa destructor is thebest life insurance for honey bee colonies. Inaddition, wintering strong colonies headed byyoung queens will improve the chances of thecolonies to stay alive over winter. Followingthese recommendations will not generate eter-nal honey bee colonies but will definitely re-duce colony winter mortality.

ACKNOWLEDGEMENTS

The project was financially supported by theGerman IVA (Industrieverband Agrar) and theGerman Beekeepers’ Association (DIB) and non-financially supported by the State Ministry ofNutrition, Agriculture and Consumer Protection(BMELV). E.G. was supported by grants from theMinistries for Agriculture from Brandenburg andSachsen-Anhalt, Germany. In addition, we are in-debted to all the beekeepers who cooperated withinthis monitoring project and to all those who con-tributed to the success of the project by fruitfuldiscussions and by assisting with the field and labwork. We thank Dr. Martens (LUFA, Speyer, Ger-many) for performing the residue analysis in pollen.

Le programme de surveillance de l’abeille enAllemagne : une étude à long terme pourcomprendre les pertes hivernales importantesconstatées périodiquement dans les coloniesd’abeilles.

perte des colonies / Varroa /DWV / virus des ailesdéformées / APV/ virus de la paralysie aiguë /Nosema / pesticides

Zusammenfassung – Das Deutsche Bienenmoni-toring: Eine Langzeitstudie zum Verständnis pe-riodisch auftretender, hoher Winterverluste beiHonigbienenvölkern. Die Honigbiene Apis melli-fera ist weltweit der wichtigste Bestäuber in derLandwirtschaft und nach aktuellen Schätzungenwird der globale Bedarf an kommerzieller Bestäu-bung weiter steigen. Dadurch stellt der seit Jahrenzu beobachtende stetige Rückgang der Bienenvöl-ker in Nord-Amerika und Europa ein ernsthaftesProblem für die Landwirtschaft dar. Für die Ab-nahme der Bienenvölker werden neben wirtschaft-lichen Faktoren vor allem periodisch auftretendeVölkerverluste verantwortlich gemacht, für die aber

Page 18: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

18 E. Genersch et al.

eine eindeutige Ursachenanalyse bisher fehlt.Zur Ursachenaufklärung von Winterverlusten führ-ten wir von 2004 bis 2009 ein Monitoringprojektdurch, in dem mehr als 1200 Bienenvölker auf125 über ganz Deutschland verteilten Bienenstän-den (Abb. 1) kontinuierlich beprobt und kontrolliertwurden. Die beteiligten „Monitoringimker“ stelltenhierfür 10 ihrer Völker zur Verfügung und liefer-ten Daten zu Honigerträgen, Wanderungen und Ab-legerbildung. Mitarbeiter der Bieneninstitute nah-men zweimal im Jahr Bienenproben für Krankheits-untersuchungen (Nosema spec, Varroa destructor,4 verschiedene Bienenviren) sowie Bienenbrotpro-ben für Rückstandsuntersuchungen. Die Stärke derBienenvölker wurde bei der Ein- und Auswinte-rung bestimmt; als „Überwinterungsverlust“ wur-den Völker definiert, die tot waren bzw. nicht genugBienen für eine erfolgreiche Frühjahrsentwicklungaufwiesen.Die Winterverluste schwankten zwischen 3,5 % und15,2 % (Abb. 3) mit ungleicher Verteilung innerhalbder beteiligten Imker (Abb. 4). Für die Ursachen-analyse wurden die überlebenden mit den zusam-mengebrochenen Völkern verglichen. Dabei zeig-ten sich die größten und hochsignifikanten (P <0,000001, U-Test) Unterschiede beim Varroabefallder Bienen im Oktober (Tab. III, Abb. 5). Eben-falls hochsignifikante Unterschiede ergaben sich fürdie Bienenviren DWV (P < 0,00001) und APBV(P < 0,0039), nicht jedoch für KBV, SBV und denNosemabefall (Tab. V). Erstaunlicherweise warenVölker mit jungen Königinnen signifikant seltenervon Winterverlusten betroffen als mit älteren Kö-niginnen (Tab. VI), während z. B. Beutenmaterialoder Rähmchenmaß keine Rolle spielten.Bei den insgesamt in drei Jahren auf Pestizidrück-stände untersuchten 215 Bienenbrotproben wurdeninsgesamt über 50 Wirkstoffe (von 256) nachge-wiesen, die meisten im Spurenbereich. Häufig wur-den mehrere Wirkstoffe gefunden und nur etwasmehr als 20 % der Proben waren frei von messba-ren Rückständen (Tab. VII). Neonikotinoide wur-den nur in einer einzigen Probe nachgewiesen. Eskonnte keine Korrelation von Rückstandswerten mitWinterverlusten festgestellt werden. Es gab auchkeinen Zusammenhang zwischen der Überwinte-rung von Bienenvölkern und dem Umfang des zu-vor eingetragenen Rapshonigs (Abb. 6).Unser Projekt zeigt, dass der Varroabefall im Herbst(zusammen mit den assoziierten Sekundärinfektio-nen) eine Hauptursache für Überwinterungsverlustedarstellt. Eine konsequente Varroabehandlung undstarke Bienenvölker mit jungen Königinnen sinddaher die wichtigste Empfehlung, um Winterverlu-sten vorzubeugen. Ein zusätzlicher Einfluss der üb-rigen Faktoren kann nicht ausgeschlossen werden,hierfür sind aber modifizierte Versuchsansätze not-wendig.

Völkerverluste / Varroa / DWV / ABPV / Nose-ma / Pestizide

REFERENCES

Aizen M.A., Harder L.D. (2009) The global stock ofdomesticated honey bees is growing slower thanagricultural demand for pollination, Curr. Biol. 19,915–918.

Aizen M., Garibaldi L., Cunningham S., Klein A.(2008) Long-term global trends in crop yield andproduction reveal no current pollination shortagebut increasing pollinator dependency, Curr. Biol.18, 1572–1575.

Akyol E., Yeninar H., Karatepe M., Karatepe B.,Özkök D. (2007) Effects of queen ages on Varroa(Varroa destructor) infestation level in honey bee(Apis mellifera caucasica) colonies and colonyperformance, Ital. J. Anim. Sci. 6, 143–149.

Amdam G.V., Hartfelder K., Norberg K., Hagen A.,Omholt S.W. (2004) Altered physiology in workerhoney bees (Hymenoptera: Apidae) infested withthe mite Varroa destructor (Acari: Varroidae):A factor in colony loss during overwintering? J.Econ. Entomol. 97, 741–747.

Bailey L., Ball B.V. (1991) Honey Bee Pathology,Academic Press, New York, London.

Bailey L., Gibbs A.J., Woods R.D. (1963) Two virusesfrom adult honey bees (Apis mellifera Linnaeus),Virology 21, 390–395.

Baker A., Schroeder D. (2008) Occurrence and geneticanalysis of picorna-like viruses infecting workerbees of Apis mellifera L. populations in Devon,South West England, J. Invertebr. Pathol. 98, 239–242.

Bakonyi T., Farkas R., Szendroi A., Dobos-Kovacs M.,Rusvai M. (2002) Detection of acute bee paralysisvirus by RT-PCR in honey bee and Varroa destruc-tor field samples: rapid screening of representativeHungarian apiaries, Apidologie 33, 63–74.

Ball B.V. (1983) The association of Varroa jacobsoniwith virus diseases of honey bees, Exp. Appl.Acarol. 19, 607–613.

Ball B.V. (1989) Varroa jacobsoni as a virusvector, In Present status of varroatosis inEurope and progress in the varroa mite control,Office for Official Publications of the EuropeanCommunities, Luxembourg.

Ball B.V., Allen M.E. (1988) The prevalence ofpathogens in honey bee (Apis mellifera) colonoesinfested with the parasitic mite Varroa jacobsoni,Ann. Appl. Biol. 113, 237–244.

Barnett E.A., Charlton A.J., Fletcher M.R. (2007)Incidents of bee poisoning with pesticides in theUnited Kingdom, 1994-2003, Pest Manag. Sci. 63,1051–1057.

Berenyi O., Bakonyi T., Derakhshifar I., KöglbergerH., Nowotny N. (2006) Occurence of six honeybee viruses in diseased Austrian apiaries, Appl.Environ. Microbiol. 72, 2414–2420.

Page 19: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 19

Bowen-Walker P.L., Martin S.J., Gunn A. (1999) Thetransmission of deformed wing virus betweenhoney bees (Apis mellifera L.) by the ectoparasiticmite Varroa jacobsoni Oud., J. Invertebr. Pathol.73, 101–106.

Chauzat M.P., Higes M., Martin-Hernandez R., MeanaA., Cougoule N., Faucon J.P. (2007) Presence ofNosema ceranae in French honey bee colonies, J.Apic. Res. 46, 127–128.

Chen Y.P., Huang Z.Y. (2010) Nosema ceranae, anewly identified pathogen of Apis mellifera in theU.S.A. and Asia, Apidologie 41, this issue.

Cox-Foster D.L., Conlan S., Holmes E.C., Palacios G.,Evans J.D., Moran N.A., Quan P.-L., Briese S.,Hornig M., Geiser D.M., Martinson V., vanEn-gelsdorp D., Kalkseitn A.L., Drysdale L., Hui J.,Zhai J., Cui L., Hutchison S., Simons J.F., EgholmM., Pettis J.S., Lipkin W.I. (2007) A metagenomicsurvey of microbes in honey bee colony collapsedisorder, Science 318, 283–287.

Currie R.W., Gatien P. (2006) Timing acaricidetreatments to prevent Varroa destructor (Acari:Varroidae) from causing economic damage tohoney bee colonies, Can. Entomol. 138, 238–252.

de Miranda J.R., Fries I. (2008) Venereal and verti-cal transmission of deformed wing virus in honeybees (Apis mellifera L.), J. Invertebr. Pathol. 98,184–189.

de Miranda J.R., Genersch E. (2010) Deformed wingvirus, J. Invertebr. Pathol. 103, S48–S61.

de Miranda J., Cordoni G., Budge G. (2010) The acutebee paralysis virus – Kashmir bee virus – Israeliacute paralysis virus complex, J. Invertebr. Pathol.103, S30–S47.

Decourtye A., Devillers J., Cluzeau S., Charreton M.,Pham-Delègue M.H. (2004) Effects of imidaclo-prid and deltamethrin on associative learning inhoney bees under semi-field and laboratory con-ditions, Ecotoxicol. Environ. Saf. 57, 410–419.

Decourtye A., Lacassie E., Pham-Delègue M.H.(2003) Learning performances of honey bees(Apis mellifera L.) are differentially affected byimidacloprid according to the season, Pest Manag.Sci. 59, 269–278.

Delaplane K.S., Hood W.M. (1999) Economic thresh-old for Varroa jacobsoni Oud. in the southeasternUSA, Apidologie 30, 383–395.

Desneux N., Decourtye A., Delpuech J.M. (2007) Thesublethal effects of pesticides on beneficial arthro-pods, Annu. Rev. Entomol. 52, 81–106.

Ellis J., Evans J.D., Pettis J. (2010) Reviewing colonylosses and Colony Collapse Disorder in the UnitedStates, J. Apic. Res. 49, 134–136.

Fries I., Camazine S., Sneyd J. (1994) Population dy-namics of Varroa jacobsoni: a model and a review,Bee World 75, 5–28.

Fries I., Ekbohm G., Villumstad E. (1984) Nosemaapis, sampling techniques and honey yield, J.Apic. Res. 23, 102–105.

Fries I., Hansen H., Imdorf A., Rosenkranz P. (2003)Swarming in honey bees (Apis mellifera) andVarroa destructor population development inSweden, Apidologie 34, 389–398.

Genersch E. (2005) Development of a rapid and sensi-tive RT-PCR method for the detection of deformedwing virus, a pathogen of the honey bee (Apis mel-lifera), Vet. J. 169, 121–123.

Giray T., Kence M., Oskay D., Döke M.A., KenceA. (2010) Colony losses survey in Turkey andcauses of bee deaths, Apidologie, in press, DOI10.1051/apido/2009077.

Gisder S., Aumeier P., Genersch E. (2009) Deformedwing virus (DWV): viral load and replication inmites (Varroa destructor), J. Gen. Virol. 90, 463–467.

Gnädinger F. (1984) Auswinterungsverluste beiBienenvölkern in Baden, Allg. DeutscheImkerzeitung 18, 297–299.

Goodwin R.M., Taylor M.A., Mcbrydie H.M., CoxH.M. (2006) Drift of Varroa destructor -infestedworker honey bees to neighbouring colonies, J.Apic. Res. 45, 155–156.

Greatti M., Milani N., Nazzi F. (1992) Reinfestationof an acaricide-treated apiary by Varroa jacobsoniOud., Exp. Appl. Acarol. 16, 279–286.

Higes M., Garcia-Palencia P., Martin-Hernandez R.,Meana A. (2007) Experimental infection ofApis mellifera honey bees with Nosema ceranae(Microsporidia), J. Invertebr. Pathol. 94, 211–217.

Higes M., Martín-Hernández R., Botías C., GarridoBailón E., González-Porto A.V., Barrios L., delNozal M.J., Bernal J.L., Jiménez J.J., GarcíaPalencia P., Meana A. (2008) How natural infec-tion by Nosema ceranae causes honey bee colonycollapse, Environ. Microbiol. 10, 2659–2669.

Higes M., Martín-Hernández R., Garrido-Bailón E.,González-Porto A.V., García-Palencia P., MeanaA., del Nozal M.J., Mayo R., Bernal J.L. (2009)Honey bee colony collapse due to Nosema cer-anae in professional apiaries, Environ. Microbiol.Rep. 1, 110–113.

Higes M., Martin R., Meana A. (2006) Nosema cer-anae, a new microsporidian parasite in honey beesin Europe, J. Invertebr. Pathol. 92, 93–95.

Highfield A.C., El Nagar A., Mackinder L.C.M., NoëlL.M.-L.J., Hall M.J., Martin S.J., Schroeder D.C.(2009) Deformed wing virus implicated in over-wintering honey bee colony losses, Appl. Environ.Microbiol. 75, 7212–7220.

Horn H. (2009) Die wichtigsten Trachtpflanzen undihre Sortenhonige, Teil 1: Raps. Allg. DeutscheImkerzeitung 43, 13–15.

Page 20: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

20 E. Genersch et al.

Iqbal J., Müller I. (2007) Virus infection causes spe-cific learning deficits in honey bee foragers, Proc.R. Soc. B 274, 1517–1521.

Johnson R.M., Ellis M.D., Mullin C.A., Frazier M.(2010) Pesticides and honey bee toxicitiy –U.S.A., Apidologie 41, this issue.

Johnson R.M., Evans J.D., Robinson G.E., BerenbaumM.R. (2009) Changes in transcript abundance re-lating to colony collapse disorder in honey bees(Apis mellifera), Proc. Natl. Acad. Sci. USA 106,14790–14795.

Karise R. (2007) Foraging behaviour and physi-ology of bees: impact of insecticides, PhD-thesis, Estonian University of Life Sciences, Tartu,Estonia, 123 p.

Klee J., Besana A.M., Genersch E., Gisder S., NanettiA., Tam D.Q., Chinh T.X., Puerta F., Ruz J.M.,Kryger P., Message D., Hatjina F., Korpela S.,Fries I., Paxton R.J. (2007) Widespread dispersalof the microsporidian Nosema ceranae, an emer-gent pathogen of the western honey bee, Apis mel-lifera, J. Invertebr. Pathol. 96, 1–10.

Klein A.-M., Vaissiere B.E., Cane J.H., Steffan-Dewenter I., Cunningham S.A., Kremen C.,Tscharntke T. (2007) Importance of pollinators inchanging landscapes for world crops, Proc. R. Soc.B 274, 303–313.

Liebig G. (2001) How many varroa mites can be toler-ated by a honey bee colony? Apidologie 32, 482–484.

Maori E., Lavi S., Mozes-Koch R., Gantman Y.,Edelbaum O., Tanne E., Sela I. (2007) Isolationand characterization of IAPV, a dicistrovirus af-fecting honey bees in Israel: evidence for intra-and inter-species recombination, J. Gen. Virol. 88,3428–3438.

Maori E., Paldi N., Shafir S., Kalev H., Tsur E., GlickE., Sela I. (2009) IAPV, a bee-affecting virus as-sociated with Colony Collapse Disorder can be si-lenced by dsRNA ingestion, Insect Mol. Biol. 18,55–60.

Martin-Hernandez R., Meana A., Garcia-Palencia P.,Marin P., Botias C., Garrido-Bailon E., Barrios L.,Higes M. (2009) Effect of temperature on the bi-otic potential of honey bee microsporidia, Appl.Environ. Microbiol. 75, 2554-2557.

Martin S.J. (2001) The role of Varroa and viralpathogens in the collapse of honey bee colonies:a modelling approach, J. Appl. Ecol. 38, 1082–1093.

Martin-Hernandez R., Meana A., Prieto L., SalvadorA.M., Garrido-Bailon E., Higes M. (2007)Outcome of colonization of Apis mellifera byNosema ceranae, Appl. Environ. Microbiol. 73,6331–6338.

Meixner M.D., Illies I., Büchler R., Wallner K. (2009)Pesticide use in rape seed culture - are residues inhoney unavoidable? Apidologie 49, 669.

Moncharmont F.X.D., Decourtye A., Hennequet-Hantier C., Pons O., Pham-Delègue M.H. (2003)Statistical analysis of honey bee survival afterchronic exposure to insecticides, Environ. Toxicol.Chem. 22, 3088–3094.

Moritz R.F.A., de Miranda J., Fries I., Le Conte Y.,Neumann P., Paxton R.J. (2010) Research strate-gies to improve honey bee health in Europe,Apidologie 41, this issue.

Morse R.A., Calderone, N.W. (2000) The value ofhoney bee pollination in the United States, BeeCulture 128, 1–15.

Oldroyd B.P. (2007) What’s killing American honeybees? PLoS Biology 5, e168.

Pettis J.S., Delaplane K.S. (2010) Coordinatedresponses to honey bee decline in U.S.A.,Apidologie 41, this issue.

Pham-Delègue M.H., Decourtye A., Kaiser L.,Devillers J. (2002) Behavioural methods to assessthe effects of pesticides on honey bees, Apidologie33, 425–432.

Potts S.G., Roberts S.P.M., Dean R., Marris G., BrownM.A., Jones H.R., Neumann P., Settele J. (2010)Declines of managed honey bees and beekeepersin Europe, J. Apic. Res. 49, 15–22.

Renz M., Rosenkranz P. (2001) Infestation dynam-ics and reinvasion of Varroa destructor mites inhoney bee colonies kept isolated and in groups,Apidologie 32, 492–494.

Rosenkranz P., Kirsch R., Renz R. (2006) Populationdynamics of honey bee colonies and varroatolerance: a comparison between Uruguay andGermany, In: 7th Encontro sobre Abelhas, USP,Ribeirão Preto, Brazil.

Shen M., Yang X., Cox-Foster D., Cui L. (2005) Therole of varroa mites in infections of Kashmir beevirus (KBV) and deformed wing virus (DWV) inhoney bees, Virology 342, 141–149.

Siede R., König M., Büchler R., Failing K., ThielH.-J. (2008) A real-time PCR based survey onacute bee paralysis virus in German bee colonies,Apidologie 39, 650–661.

Stoltz D.B., Shen X.-R., Boggis C., Sisson G. (1995)Molecular diagnosis of Kashmir bee virus infec-tion, J. Apic. Res. 34, 153–160.

Strange J.P., Sheppard, W.S. (2001) Optimum tim-ing of miticide applications for control of Varroadestructor (Acari: Varroidae) in Apis mellif-era (Hymenoptera: Apidae) in Washington State,USA, J. Econ. Entomol. 94, 1324–1331.

Tentcheva D., Gauthier L., Zappulla N., Dainat B.,Cousserans F., Colin M.E., Bergoin M. (2004)Prevalence and seasonal variations of six beeviruses in Apis mellifera L. and Varroa destruc-tor mite populations in France, Appl. Environ.Microbiol. 70, 7185–7191.

Topolska G., Gajda A., Hartwig A. (2008) Polishhoney bee colony losses during the winter of 2007/2008, J. Apic. Sci. 52, 95–104.

Page 21: The German bee monitoring project: a long term …The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies* Elke Genersch1

The German bee monitoring project 21

vanEngelsdorp D., Evans J.D., Donovall L., Mullin C.,Frazier M., Frazier J., Tarpy D.R., Hayes J., PettisJ.S. (2009a) “Entombed Pollen”: A new conditionin honey bee colonies associated with increasedrisk of colony mortality, J. Invertebr. Pathol. 101,147–149.

vanEngelsdorp D., Evans J.D., Saegerman C., MullinC., Haubruge E., Nguyen B.K., Frazier M., FrazierJ., Cox-Foster D., Chen Y., Underwood R., TarpyD.R., Pettis J.S. (2009b) Colony collapse disorder:a descriptive study, PLoS One 4, e6481.

vanEngelsdorp D., Hayes J. Jr., Underwood R.M.,Pettis J. (2008) A survey of honey bee colonylosses in the U.S., fall 2007 to spring 2008, PLoSONE 3, e4071.

vanEngelsdorp D., Underwood R., Caron D., Hayes J.(2007) An estimate of managed colony losses inthe winter of 2006-2007: A report commissionedby the apiary inspectors of America, Am. Bee J.147, 599–603.

Williams I.H. (1994) The dependences of crop produc-tion within the European Union on pollination byhoney bees, Agric. Zool. Rev. 6, 229–257.

Yang X., Cox-Foster D.L. (2005) Impact of an ectopar-asite on the immunity and pathology of an inverte-brate: Evidence for host immunosuppression andviral amplification, Proc. Natl. Acad. Sci. USA102, 7470–7475.

Yue C., Genersch E. (2005) RT-PCR analysis ofDeformed wing virus in honey bees (Apis mellif-era) and mites (Varroa destructor), J. Gen. Virol.86, 3419–3424.

Yue C., Schröder M., Bienefeld K., Genersch E.(2006), Detection of viral sequences in semen ofhoney bees (Apis mellifera): Evidence for ver-tical transmission of viruses through drones, J.Invertebr. Pathol. 92, 93–96.

Yue C., Schröder M., Gisder S., Genersch E. (2007)Vertical transmission routes for Deformed wingvirus of honey bees (Apis mellifera), J. Gen. Virol.88, 2329–2336.