Honey bee colony disorder in crop areas: the role of pesticides and viruses

Transcription

1 oney bee colony disorder in crop areas: the role of pesticides and viruses Noa SIMONELSO Noa Simonelso,2, Gilles San Martin3, Etienne Bruneau, LaureAnne Minsart, Coralie Mouret, Louis autier 3 Centre Apicole de Recherche et d Information 2 3 Utrecht University Centre wallonde Recherches Agronomiques

2 First objective of the study oney bee Colony disorders Virus Pesticides (acaricides, insecticides, fongicides, herbicides)

3 Second objective of the study oney bee Colony isorders Agricultural Land Use

4 oney bee disorders? Include several unspecific clinical symptoms for which no causal agent can be identified:.dead and disappeared colonies : 2.weak colonies 3.queen loss 4.problems linked to brood and not related to any known disease iscard :. pesticides (acute intoxication) 2. Varroa destructor 3. starvation

5 () Link between disorders and virus pesticide residues

6 Strategy The condition of honey bee colonies was followup 3 times : 2 before and after the winter with the help of questionnaires and clinical visits done by beekeeping technicians At each visit, samples were collected for residues and viral analysis: wax, honey, beebread, honeybees On basis of field observations, set up of two groups ( with disorders and healthy group), to constitute the group of colonies whose samples will be analyzed Compare the content of virus and residues between the two groups

7 Selection of beekeepers and colonies Need more than 300 colonies (prevalence rate of disorder ~ 030% and statistical analysis with model requires n~30) Request of voluntary beekeepers to participate in the study, fulfilling 3 criteria: ) have a minimum of 5 production colonies per apiary in June 20; 2) regularly follow up the health status and development of their colonies; 3) apply good beekeeping practices: monitor the varroa infestation level and carry out a varroa control Five welldeveloped colonies selected from each apiary CARI , Gembloux AgroBio Tech ,

8 Selection of beekeepers and colonies July 20 (n=330 colonies & 66 apiaries )

9 ata collection Apiaries were visited by beekeeping technicians and paper questionnaires were filled at each visit : () midjuly to midaugust 20, (2) midseptember to midoctober 20, (3) March to April 202 Specific questionnaires for beekeeper, apiary and colony Clinical inspection of the colony Information on the queen Varroa treatment Colony status (possible clinical symptoms): mortality, weakness, brood problems, queen loss, diarrhoea, deformed wings, Level of reserves

10 Matrix sampling 3 Sample periods: () midjuly to midaugust 20, (2) midseptember to midoctober 20, (3) March to April 202 oney bees 00 alive bees from brood frames Beebread, honey, wax cutting of frames : 0x0cm or 5x20cm (00 cm²)

11 Apiaries with matrices and fully completed forms April 202 (n=73 colonies & 35 apiaries )

12 73 colonies Selection according the availability oney bees / honey / beebread Period & 2 (n=07) Selection of samples ) Colony treated against varroa 2) Food reserves before and after the winter 3) dronelaying queens Clustering Symptoms (n=75) ) 2) 3) 4) Clustering Colony condition (n=75) mortality weakening brood problem queen loss ) 2) 3) Selection for analyses 25 colonies with disorders 29 healthy colonies Virus & pesticides residues analyses Food reserve before the winter : B/B2/B3 Year of creation Queen color

13 Virus analyses Quantitative RTPCR by the National Bee Unit laboratory, Food and Environment Research Agency (Sand utton, York, UnitedKingdom).. Black Queen Cell Virus (BQCV) 2. Chronic Bee Paralysis Virus (CBPV) 3. Acute Bee Paralysis Virus (ABPV) 4. eformed Wings Virus (WV) 5. Sac Brood Virus (SBV)

14 Pesticide analyses Eurofins Chemiphar NV, Brugge, Belgium and analysed by SOFIA Gmb Chemisches Labor für Softwareentwicklung und Intelligente Analytik, Berlin, Germany. Two multiresidues methods (SF46 and SF50) were used searching pesticides residues : 99 residues in wax (54 samples), 93 residues in beebread (08 samples) 96 residues in honey (07 samples one sample did not contain enough matrix)

15 Pesticide analyses Active ingredient or metabolite* 2,4 Abamectin Acetamiprid Aldicarb Alphacypermethrin Amitraz Azoxystrobin Bentazone Benthiavalicarb Betacyfluthrin Bifenthrin Boscalid Captan Carbaryl Chlorpyriphos Chlorpyriphosmethyl Chlorothalonil Clothianidin Coumaphos Coumaphos oxon * Coumaphos phenolic* Cyazofamid Cyfluthrin Cymoxanil Cypermethrin Cyproconazole Cyprodinil T eltamethrine ichlorpropp ifenoconazole iflubenzuron imethenamidp imethoate LOQ (mg/kg) Class I I I I A F F I I F F I I I F I A A A F I F I F F I I F I I Wax Beebread oney insecticides acaricides & metabolites 4 fungicides 4 herbicides synergist

16 Statistics ESCRIPTIVE ANALYSES Linear model (two way ANOVA) >dependent variable: Ct or number of residues >explanatory variables: visit (first or second), group (with disorder or healthy) or type (fungicide or insecticide/acaricide) and their interaction RELATIONSIP BEE COLONY ISORERSPOTENTIAL STRESSORS GLM models (with binomial distribution and logit link function) >dependent binary variable: group (with disorder or healthy) & 2) >explanatory variables: total number of fungicides, insecticidesacaricides, viruses (visits >random effect : apiary Likelihood Ratio (LR) Tests to evaluate the significance of the explanatory variables

17 (2) Link between disorders and land use

18 Spatial analysis Geographical Coordinates of Apiaries Land Parcel Identification System (CAP) elimitation of «buffer» around apiary :,5 km Calculation of crop and grassland surfaces Statistical analysis issorder ~ surf. crops * surf. grassland

19 Spatial analysis around 3 apiaries Grasslands (ha) Crops (ha) 40A B C

20 Statistics RELATIONSIP BEE COLONY ISORERAPIARY ENVIRONMENT GLM models (with binomial distribution and logit link function) >dependent binary variable: group (with disorder or healthy) >explanatory variables: surface of grasslands and crops sensu lato (cereals, potatoes, beet, oilseed rape, maize, flax,...) and fruit, vegetable, fodder, horticulture productions >random effect: apiary Likelihood Ratio (LR) Tests to evaluate the significance of the explanatory variables

21 Main results

22 Observed disorders n=29 healthy n=24 with disorders (+ 5A4 mortality)

23 etected virus (National Bee Unit, UK) Molecular analysis (RTPCR) on 53 samples 3 dominant virus : WV, BQCV, SBV No significant difference between with disorders and healthy : WV : p = 0.73 BQCV : p = SBV : p = 0.54

24 Content of residues ** With disorders vs. healthy Significantly more fungicides residues (p = 0.008), 2 times more! No significant difference for acaricides/insecticides (p = 0.402) NS

25 Fungicide residues

26 Insecticide /acaricide residues

27 isorders ~ virus*pesticides Significant positive correlation between the number of fungicides residues and the disorder No significant correlation between viral charge or residues of insecticidesacaricides No interactions >fungicides*virus >insecticidesacaricides*virus >fungicides*insecticidesacaricides LR df p(>chisq) totfungicides totinsaca totvirus totfungicides:totvirus totinsaca:totvirus totfungicides:totinsaca

28 isorder ~ agricultural land use isorder probability : with the surface of crop ( LR = 8.052, df =, p = ) with the surface of grassland (LR = 4.527, df =, p = 0.000) Crops Grasslands => Apiaries surrounded by a crop environment have more probability to have a disorder!

29 Conclusions There is a significant positive correlation between the appraisal of bee disorders and ) the fungicides residues 2) the surface of crops around apiaries These results questioned the: General acknowledgement of the nontoxicity of fungicides on honey bee colonies Compatibility between the intensive agricultural model and pollinators wellbeing

30 Thanks to Beekeeping technicians & beekeepers Project funded by Wallonie Service public de Wallonie irection générale opérationnelle de l Agriculture, des Ressources naturelles et de l Environnement (GO3)

31

32 Carbaryl Zoxamide s.d s.d. 39 Mean Range s.d. Trifloxystrobin Thiophanatemethyl 24 Thiacloprid 28 Terbuthylazine Tebufenozide 3. Tebuconazole 3. Taufluvalinate Pyrimethanil Pyraclostrobin Propamocarbe Piperonyl butoxide Iprodione Indoxacarb Fenpropimorph 4 Cyprodinil 25 Coumaphos 28 Chlorothalonil 0.29 Chlorpyriphos 0.29 Mean Range 8 Quantified Captan 6 Residues amount (mg/ kg) etected LOQ (mg/kg) Quantified Number of sampl es (n=07) etected oney Residues amount (mg/ kg) < LO Boscalid Mean Range LOQ (mg/kg) Quantified LOQ (mg/ kg) etected Amitraz Group Number of samples (n= 08) < LO Active ingredient Bee bread Residues amount (mg/kg) < LO Wax Number of samples (n= 54)

33 ierarchical clustering? Why : in order to build groups with the lowest variation within the group and the highest variation between group (see Zuur et al. 2007) ow : Ward aggregation method