Analysis of Honey Bee Hive Products as a Model for Monitoring Pesticide Usage in Agroecosystems

Abstract

Global food and nutritional security majorly rely on honey bees for pollination. Furthermore, honey bees (Apis mellifera), are considered as reliable biological indicators of environmental contamination because they pick up chemical pollutants in the air or in flowers as they search for food. As a result, the honey bee colony environment acts as a reservoir for a diversity of resources of floral origin and therefore analyzing hive products is more cost effective compared to monitoring individual crops. Effective methods for monitoring agrochemicals contamination in the environment can therefore be achieved by continuous analysis of honey bee products. We investigated pesticide residues in honey and pollen collected from honey bee hives in various agro-ecological zones across Kenya over a period of two years (September 2013 to August 2015) to determine the circulating organic chemical pollutants in the environment. A total of 36 pesticide residues were detected belonging to three chemical classes; insecticides (>50%) fungicides (27%) and herbicides (20%) with majority of the pesticides detected in pollen compared to honey. Although herbicides appeared to be the least prevalent, they were detected at the highest concentrations of up to 356 ppb in honey compared to insecticides which were detected at fairly low concentrations (0.1 to 53 ppb). Our findings highlight the need to create greater awareness of the ecological consequences of wide scale use of agro-chemicals in agriculture.

Full text

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
9
Analysis of Honey Bee Hive Products as a Model for Monitoring
Pesticide Usage in Agroecosystems
Janet Irungu* Ayuka T. Fombong Justus Kurgat Protus Mulati Juliette Ongus Kiatoko Nkoba
Suresh Raina
International Centre of Insect Physiology and Ecology (icipe), African Reference Laboratory for Bee Health,
P.O. Box 30772-00100, Nairobi, Kenya
Abstract
Global food and nutritional security majorly rely on honey bees for pollination. Furthermore, honey bees (Apis
mellifera), are considered as reliable biological indicators of environmental contamination because they pick up
chemical pollutants in the air or in flowers as they search for food. As a result, the honey bee colony environment
acts as a reservoir for a diversity of resources of floral origin and therefore analyzing hive products is more cost
effective compared to monitoring individual crops. Effective methods for monitoring agrochemicals contamination
in the environment can therefore be achieved by continuous analysis of honey bee products. We investigated
pesticide residues in honey and pollen collected from honey bee hives in various agro-ecological zones across
Kenya over a period of two years (September 2013 to August 2015) to determine the circulating organic chemical
pollutants in the environment. A total of 36 pesticide residues were detected belonging to three chemical classes;
insecticides (>50%) fungicides (27%) and herbicides (20%) with majority of the pesticides detected in pollen
compared to honey. Although herbicides appeared to be the least prevalent, they were detected at the highest
concentrations of up to 356 ppb in honey compared to insecticides which were detected at fairly low concentrations
(0.1 to 53 ppb). Our findings highlight the need to create greater awareness of the ecological consequences of wide
scale use of agro-chemicals in agriculture.
Keywords: Pesticide residues, honey bees (Apis mellifera), honey and pollen
1. Introduction
Agrochemicals are crucial to modern agriculture as they protect crops from pests and disease invasion thereby
boosting crop productivity that is much needed to meet the world food demands (József, 2013; Aktar, 2009).
Globally, millions of tons of pesticides are applied annually, but only a small fraction (<1%) effectively reaches
the target organisms, and the remainder is deposited either in the soil, atmosphere or water, contaminating the
environment and non-target organisms (József, 2013; Horrigan, 2002). In Africa, pesticides use represents less
than 5% of the total amount of pesticides used worldwide but many developing countries have large stockpiles of
obsolete pesticides, usually scattered over various sites (PAN UK, 2007, World Bank, 2013). These pesticides are
in deplorable state and are hazardous to both human and environmental health (Dinham, 2003; World Bank, 2013).
Moreover, the rapid increase in human population in African countries requires more food supply putting
a strain on agricultural land available for crop production (Naidoo, 2010; Williamson, 2008). In Kenya and other
African nations, most farmers are risk-averse and having small farm sizes strive to maximize their output by using
fertilizers and other pesticides (Aduol, 2005). A recent survey carried out in Kenya showed that farmers used a
dose above the recommended levels in their effort to reduce pest damage (Gitonga et al., 2010). These farmers
are accustomed to pesticide use in response to any signs of crop damage and most have little or no knowledge on
alternative pest-management approaches (Dinham, 2003; Lekei, 2014). As a result, the tendency to rely on
pesticides use to enhance agricultural output in Africa is on the rise. This dependency on pesticides threatens food
safety, causes health risks and environmental problems, and deepens the inequality between rich and poor African
farmers (Ngowi, 2007; Williamson, 2008; Oesterlund, 2014). Additionally, most farmers have limited knowledge
on pesticides and their widespread use of these pesticides in agriculture results in inappropriate use (Kimani, 1995;
Lekei, 2014). The subsequent slow degradation of some of these pesticides unfavorably affects the whole
ecosystem by entering into the food chain and polluting the air, soil and water (Asenso-Okyere, 2011; Lekei, 2014).
As a consequence, methods for monitoring pesticide residues circulating in African agro-ecosystems are required
to prevent their eventual toxicity to human health and the potential hazard to the conservation of the ecological
equilibrium. Conversely, analysis of trace pesticides in the environment across a large spatial area requires
laborious and expensive sample effort which is a major hurdle for most African countries.
Pollinators, particularly honey bees (Apis mellifera), are considered as reliable biological indicators
(Giorgio, 2003; Porrini, 2002) because they reveal the chemical contaminants in the environment which they
intercept in the air or in flowers as they search for food (Wallowork-Barber, 1992; Fernandez, 2002). Since honey
bees have great mobility to forage vast areas, they can be monitored cheaply for chemical pollutants by analyzing
their hive products. Honey bees are also well known to be highly susceptible to most chemicals and are typically
used as representatives of non-target beneficial insects by environmental agencies worldwide to measure toxicity
of pesticides during registration process (Desneux, 2007; Stoner, 2013; Solecki, 2006). Exposure of honey bees to

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
10
pesticides either leads to their sudden death, if the chemical is highly toxic or the doses are high or sub-lethal
effects which may compromise their immunity or foraging behavior and the return of the contaminated food to the
hive exposes the whole colony. The presence of these contaminants in honey bee food can then be detected using
appropriate analytical methods (Fernandez, 2002; Kevan 1999; Porrini et al. 2000). As a result, continuous analysis
of honey bees and their products can be employed as a method for monitoring fluctuations in pesticide exposure
and their levels in the environment at any given time. So far, information on the levels of pesticide residues in hive
products from Africa its environs are scanty.
Herein, pesticide residue analysis of honey bees and their products collected from different agro-
ecological zones in Kenya over a two year period (September 2013-August 2015) was performed as a first step to
monitor the extent of environmental contamination in Africa and also provide some insights on chemical residues
currently circulating in this country that could affect beneficial arthropods, the environment and the community at
large. A multi-residue analytical approach was employed using liquid chromatography-tandem mass spectrometry
(LC-MS/MS) and the acquired data was used to generate a map on distribution of various pesticides that are
currently present in the surveyed regions across Kenya.
1.1 Ethical consideration
Prior to undertaking the study in each of the selected agro-ecological zones, informed consents were obtained from
the owners of the honey bee colonies after explaining to them the background and the objectives of the study. The
participants in this study were mainly small holder farmers working individually or as part of a beekeepers
association group.
2.0. Materials and methods
2.1 Selection of study sites
Study sites were selected to reflect the major agro-ecological zones responsible for over 80% of food production
in Kenya, see Figure 1. The major food crops present in these regions were maize, beans, and various vegetable
and fruit crops with the exception of one site in Kiambu which consisted of a mixture of large scale farming
(horticultural, coffee, french beans and pineapple farms) and small scale farming containing maize and beans. In
each of these agro-ecological zones, eight apiaries spread apart (>10 km from each other) were randomly selected.
The choice of the apiaries from each of these sites was based on the number and the strength of active colonies
present.
Figure 1: Map of Kenya highlighting study sites from different agro-ecological zones (Macharia, 2004)

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
11
2.2 Sample collection and storage
Sample collection was performed over a period of 2 years (September 2013 to August 2015). From the eight
selected apiaries, five colonies were randomly selected for sample collection. Samples (honey and pollen) were
collected from the apiaries in each of the sites provided in Figure 1 except for Isiolo, Laikipia, and Nairobi where
samples were collected from only two apiaries per site due to limited number of available apiaries. Samples
collection was conducted during different seasons, Nov-Dec (short rains), Feb-March (start of long rains) and July-
Aug (dry season). A total of 261 honey samples and 322 pollen samples were collected during the study period.
All samples were immediately stored in either liquid nitrogen or cooler boxes and transported to the laboratory
where they were stored at -80
O
C until further analysis.
2.3 Chemicals and reagents
All pesticide standards were of high purity (>90%) and were obtained from Sigma-Aldrich (Chemie GmbH,
Germany) and Dr Ehrenstorfer GmbH (Augsburg, Germany). These standards were stored according to
manufacturer’s recommendations until use. Pesticide stock solutions were prepared in acetonitrile at 1µg/mL and
stored in amber screw-capped glass vials at -20
o
C until analysis.
2.4 Sample preparation
Samples of the same matrix from each apiary, which constituted of 5 colonies, were pooled and prepared following
the QuEChERS method (Anastassiades et al., 2013).
Briefly, from each pooled sample matrix, 5g of honey or 3g
of pollen were weighed into a 50ml falcon tube and 10ml of water added and the mixture was homogenized.
Acetonitrile (10ml) plus a mixture of QuEChERS salts were added. The samples were vortexed for 1 min and
centrifuged at 4,200 rpm for 5 min. For quality control monitoring, a blank sample was spiked with a mixture of
pesticides of interest at limit of quantification (0.1 µg kg
-1
) and was processed with a neat blank matrix (non-spiked
blank) along with the other samples of each batch of each sample matrix.
Clean-up procedure was performed by taking 1.0 ml of aliquot into 2 ml eppendorf tube and cleaned
using dispersive solid phase extraction, packed with 150 mg MgSO
4
, 50 mg PSA. Pollen samples were additionally
cleaned using graphitized carbon to remove excess pigment. The cleaned extracts were centrifuged and filtered
through hydrophilic PTFE 0.2 µm. The final extract was diluted at 1:1 (v/v) with water before transferring into
auto-sampler vial for LC-MS/MS.
2.5 LC-MS/MS instrumentation
Analysis was performed using an ultra high performance liquid chromatography (UHPLC) Agilent 1290 series
coupled to a 6490 model triple quadrupole mass spectrometer (Agilent technologies) with an ifunnel JetStream
electrospray source operating in the positive ion mode. Nitrogen was used both as a nebuliser and as the collision
gas. Data acquisition and processing was performed using Mass Hunter Data Acquisition; Qualitative and
Quantitative analysis software (Agilent Technologies, Palo Alto, CA, v.B.06 and v.B.07).
The chromatographic separation was performed on a Rapid Resolution reverse phase column-C18 1.8
µm, 2.1 x 150 mm column (Agilent Technologies). A gradient elution at a flow rate of 0.4 mL/min was used with
water and acetonitrile each containing 5 mM ammonium formate in 0.1% formic acid as mobile phase A and B
respectively.
2.6 Sample Analysis
A multi-residue approach, using LC-MS/MS for screening, was adapted to search for chemical contaminants
against 102 pesticides that were chosen based on the information obtained from the farmers and local agrochemical
stores. Data analysis was carried out by monitoring two transition ions where possible for each targeted analyte as
per LC-MS/MS criteria for residue analysis provided in SANCO document (SANCO, 2013). The most dominant
transition ion was used for quantification whereas the second most intense ion was used as a qualifier for
confirmation purposes. To generate calibration curves used for quantification, matrix-matched calibration
standards were prepared at seven calibration levels covering 0.05, 0.1, 1, 10, 25 and 50 parts per billion (ppb),
including the zero point in blank extracts of the respective matrices. The resulting calibration curves were used to
determine the method’s limit of quantification (LOQ) and limits of detection (LOD). The LOQ was set as the
minimum concentration that could be quantified with acceptable accuracy and precision.
3.0 Results and Discussion
Pesticide residues in pollen and honey (or its concentrate, nectar) are likely to account for most of the chemical
contaminant exposures to honey bees and may represent most of the potential risks concerns since bees rely on
honey and pollen to meet majority of their nutritional requirements. Results from this study indicate that among
the two hive products, pollen contained approximately 90% of the pesticide residues detected while 50% were
detected in honey. A total of 36 pesticide residues were detected of which 5 were only found in honey, 18 in pollen,

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
12
and 13 in both matrices as shown in Table 1. Previous studies that investigated chemical residues in hive products
have also reported a similar trend leading to some researchers concluding that pollen is the most contaminated
hive product (Mullin et al., 2010). A recent study conducted in Kenya evaluated two hive products (bees wax and
pollen) and found only four pesticides (1-naphthol, chlorothalonil, chlorpyrifos and fluvalinate) at very low
concentrations below 50 ppb (Muli et al., 2014). In that study, the highest concentration was found in 1-naphthol
(119 ppb), a metabolite for carbayl and naphthalene, but it was only detected in one of the 15 sites investigated.
From the current study, only two (chlorpyrifos and chlorothalonil) of the four pesticides were detected, see Table
1. The other two were possibly not detected due to the fact that the site where 1-naphthol was previously detected
was not visited in the current study whereas the highly lipophilic nature of fluvalinate is incompatible with the
analytical approach employed in this study. Regarding the big difference between the total numbers of residues
detected from the previous study (4) and current study (36), one plausible explanation could be due to the fact that
more sites were visited (45 sites) compared to the 15 sites that were visited previously, and perhaps the seasonal
variations between the two study periods.
Figure 2 represents a summary of the prevalence (%) of all the various chemical classes detected in Kenya
while Figure 3 illustrates how these chemical residues are distributed across the country. Overall, it appears that
insecticides are the most prevalent pesticides (>50%, including neonicotinoids and acaricides) followed by
fungicides (27%) and herbicides (20%), see Figure 2, implying that pests are the major threat to most agricultural
crops in the study areas investigated whereas herbicides are the least frequently used.
Figure 2: Representative example showing prevalence of different pesticide classes used by Kenyan farmers

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
13
Table 1: Concentration of detected pesticides in Kenya and effect on honey bees
Pesticide Residue Category Mean Conc. in honey (ppb)
Mean Conc. in pollen (ppb)
Acetamiprid Insecticide 0.5 N/D
Aldicarb Insecticide 1.19 27.1
Atrazine Herbicide 356.7 23.5
Azoxystrobin Fungicide 30.8 N/D
Bupirimate Fungicide N/D 0.47
Carbaryl Insecticide 0.31 0.4
Carbendazim Fungicide 0.47 4.64
Chlorpyriphos Insecticide 0.52 19.5
Chlorpyrofos methyl Insecticide N/D 4.14
Clofentazine Insecticide N/D 0.83
Chlorothalonil Fungicide N/D 0.71
Cymiazole Insecticide 0.19 0.33
Cymoxanil Fungicide 1.03 0.14
Cyproconazol Fungicide 7.04 0.46
Diazinon Insecticide 1.14 4.12
Dichlorvos Insecticide 0.6 1.47
Diflubenzuron Insecticide N/D 0.73
Dimethoate Insecticide 1.19 N/D
Epoxyconazole Fungicide N/D 0.36
Etofeprox Insecticide N/D 3.23
Febuconazole Fungicide N/D 0.17
Fenazaquin Insecticide N/D 2.55
Flutriafol Fungicide 0.81 N/D
Hexaconazole Fungicide N/D 1.56
Hexythiazox Insecticide N/D 0.94
Imidacloprid Insecticide 0.42 2.19
Malathion Insecticide N/D 52.9
Metalaxyl Fungicide N/D 0.67
Metribuzin Herbicide 52.6 3.55
Oxamyl Insecticide 0.96 5.54
pirimiphos methyl Insecticide N/D 2.95
Propamocarb Fungicide N/D 0.58
Propiconazole Fungicide N/D 0.5
Pyraclostrobin Fungicide N/D 0.71
Thiamethoxam Insecticide N/D 29.4
Triadimefon Fungicide 0.89 N/D
The same pattern was observed when the distribution of these pesticides across the country was evaluated
as shown in Figure 3. This figure illustrates the distribution of pesticides in the two hive matrices (honey and
pollen) from different agro-ecological zones in Kenya. It was noted that the concentration and the number of
pesticides detected increased with altitude and agricultural intensification. When the various insecticides detected
were further examined, organophosphates (31%) and carbamates (33%) appeared to be the most commonly used
throughout Kenya as illustrated in Figure 4. This could be attributed to their slow degradation whereas the 1% of
synthetic pyrethroids detected could be attributed to their fast degradation from the environment. This observation
is in line with studies done by Gambarcorta et al, 2005, which showed that after spraying, pesticides degrade by
first order kinetics resulting in a decrease in their residue levels over time. Unfortunately, in most developing
countries, farmers’ knowledge about pesticides and available alternatives is still remarkably limited and short-term
cost considerations still remain an important factor in poor farmers’ choices of pesticides. Cheap pesticides that
usually have long degradation periods and present a high risk to users, the public or the environment, often continue
to be used in place of less hazardous but more expensive alternatives with shorter degradation time.

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
14
Figure 3: Pesticide distribution in various agro-ecological zones in Kenya
Figure 4: Distribution of different insecticides used in Kenya
Although herbicides appeared to be the least prevalent compared to the other chemical classes, the highest
concentration detected from all pesticides originated from atrazine (356 ppb) as shown in Table 1, which is a broad
spectrum herbicide used for weed control in maize, sugarcane and pineapple farms in the surveyed regions.
Herbicides are generally known to be non-toxic to bees and therefore the levels detected would unlikely pose any
hazardous risk to them. Unfortunately no maximum residue limits (MRL) have been set for this chemical in honey
or other apicultural products hence its impact to human health requires further investigation. On the contrary,
insecticides which seemed to be the most prevalent were detected at the lowest concentrations ranging from 0.1 to
53 ppb, see Table 1. This chemical class is known to be highly toxic to honey bees. Of particular interest are the
Frequency,
Organophos
phate, 31,
31%
Frequency,
Carbamate,
33, 33%
Frequency,
Neonicotino
id, 22, 22%
Frequency,
Acaricides,
11, 12%
Frequency,
Benzamide,
1, 1%
Frequency,
Pyrethroid ,
1, 1%

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
15
neonicotinoids, in this case, acetamiprid, imidacloprid and thiamethoxam, which were detected at various
concentrations in this study. Specifically thiamethoxam, a broad spectrum systemic insecticide used in control of
sucking insects in flowers, vegetables, leaf miner in coffee and for maize and beans seed treatment, was detected
at 29 ppb as depicted in Table 1. This level could be hazardous to both humans and honey bees since the MRL
acceptable for this chemical in apicultural products that would not pose a health concern is almost 3-fold lower
(10ppb) than the concentration reported in this study. This pesticide is also known to be highly toxic to honey bees
with an oral lethal dose that would kill 50% of the test population (LD
50
) at 0.005 µg/bee (US EPA). A closer look
at the data obtained from all sites revealed that only one site, Kiambu, contained levels above the set MRL for this
compound whereas the other regions where this pesticide was detected contained very low levels (data not shown).
This was not suprising considering that Kiambu region was the only site where both small scale and large scale
farming of horticultural, french beans, pineapple and coffee farming was practiced and most of the farmers from
this region indicated that they use actara or thiamethoxam in their farms to control pests in their coffee and
horticultural farms. In addition, the wet season sampling in this region coincided with periods of heavy pesticide
spraying and application.
5. Conclusion
A total of 36 pesticides were detected across the four agro-ecological zones in Kenya belonging to a wide array of
chemical classes with insecticides as the most predominantly (>50%) used pesticides followed by fungicides
whereas herbicides were the least frequently used in the study areas. Although herbicides appeared to be the least
prevalent, they were detected at the highest concentrations of up to 356 ppb in honey compared to insecticides
which were detected at fairly low concentrations (0.1 to 53 ppb). Our findings highlight the need to create greater
awareness of the ecological consequences of wide scale use of agro-chemicals in agriculture. Further
investigations are needed to determine the effect of the detected pesticides on Africa’s agro-ecosystem, consumers
and honey bee health over time and their potential synergic effects.
Acknowledgement
The authors would like to thank Prof. Baldwyn Torto for his tremendous help and useful comments during the
manuscript write-up, colleagues from African Reference Laboratory Bee Health (ARLBH) at icipe for their
support. Special thanks to James Ng’ang’a, Fiona Mumoki, Isabella Nyamoita, Ada, M. Achieng and various
beekeepers in Kenya for their support during sample collection. This work has been supported financially by the
European Union grant number DCI-FOOD-2011/023-520.
References
1. József P, Károly P, János N. (2013) “Pesticide productivity and food security. A review,” Agron. Sustain.
Dev., 33, 243–255.
2. Aktar MW, Sengupta D, Chowdhury A. (2009) “Impact of pesticides use in agriculture: their benefits and
hazards,” Interdiscip Toxicol., 2, 1–12.
3. Horrigan L, Lawrence RS, Walker, P. (2002). “How sustainable agriculture can address the environmental
and human health harms of industrial agriculture,” Environmental Health Perspectives, 110, 5.
4. World Bank. Obsolete Pesticide Stockpiles: An Unwanted Legacy of the African Landscape.
http://www.worldbank.org/en/news/feature/2013/08/05/obsolete-pesticide-stockpiles-an-unwanted-legacy-
of-the-african-landscape. Accessed Nov. 2013.
5. PAN UK. (2007). Hazardous pesticides and health impacts in Africa. PAN UK Food &Fairness, London.
http://www.panuk.org. Accessed Nov 2015.
6. Dinham B. (2003) “Growing vegetables in developing countries for local urban populations and export
markets: problems confronting small-scale producers,” Pest Manag Sci, , 59, 575-582.
7. Naidoo S, London L, Rother HA, Burdorf A, Naidoo RN, Kromhout H. (2010) Pesticide safety training and
practices in women working in small-scale agriculture in South Africa. Occup Environ Med, 67, 823-828.
8. Williamson S, Ball A, Pretty J. (2008) Trends in pesticide use and drivers for safer pest management in four
African countries. Crop Protection, 27, 1327.
9. Oduol JBA, Tsuji M. (2005) “The effect of farm size on agricultural intensification and resource allocation
decisions: Evidence from smallholder farms in Embu district, Kenya.” J. Fac. Agr. Kyushu Univ., 50, 727-
742.
10. Gitonga W., Macharia J., Mungai A., Njue H., Kanja D. and Olweny H. (2010) “Cotton Production, Constr
aints, and Research Intervention in Kenya,” Government of Kenya Report.
11. Lekei EE, Ngowi AV, London L. (2014). Farmers’ knowledge, practices and injuries associated with pesticide
exposure in rural farming villages in Tanzania,” BMC Public Health, 14, 389.
12. Ngowi AV, Mbise TJ, Ijani AS, London L, Ajayi OC. (2007) “Pesticides use by smallholder farmers in
vegetable production in Northern Tanzania,” Crop Prot, 26, 1617-1624.

Journal of Environment and Earth Science www.iiste.org
ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online)
Vol.6, No.8, 2016
16
13. Kimani VN, Mwanthi MA. (1995) “Agrochemical exposure and health implication in Githunguri location,
Kenya,” East Afri Med J., 72, 531-5.
14. Oesterlund, AH, Thomsen JF, Deogratias K, Sekimpi JM, Apio Racheal, EJ. (2014) “Pesticide knowledge,
practice and attitude and how it affects the health of small-scale farmers in Uganda: a cross-sectional study,”
African Health Sciences, 14, 420-433.
15. Asenso-Okyere K, Catherine C, Kwaw SA. (2011) “Interactions between health and farm-labor productivity,”
International Food Policy Institute, Washington, DC.
16. Giorgio C, Bettina M. (2003) “Honey bees as bioindicators of environmental pollution,” Bulletin of
Insectology, 56 (1), 137-139.
17. Porrini C, Ghini S, Girotti S, Sabatini AG, Gattavecchia E, Celli G. (2002) “Use of honey bees as bioindicators
of environmental pollution in Italy,” In: Deveillers J, Pham-Delegue MH (eds) Honey bees: the environmental
impact of chemicals. Routledge-Taylors & Francis Group, London, pp 186–247.
18. Wallowork-Barber AK, Ferenbaugh RW, Gladney ES. (1982) The use of honey bees as monitors of
environmental pollution. American Bee Journal, 122 (11), 770-772.
19. Fernandez M, Pico Y, Manes J. (2002) “Analytical methods for pesticides residue determination in bee
products,” J. Food Protect., 65, 1502.
20. Desneux N., Decourtye A, Delpuech J. (2007) “The sublethal effects of pesticides on beneficial arthropods,”
Annu. Rev. Entomol., 52, 81‐106.
21. Stoner KA, Eitzer BD (2013) “Using a hazard quotient to evaluate pesticide residues detected in pollen trapped
from honey bees (Apis mellifera) in Connecticut,” PLoS One 8: e77550.
22. Solecki R, Davies L, Dellarco V, Dewhurst I, van Raaij M, Tritscher A. (2005) “Guidance on setting of acute
reference dose (ARfD) for pesticides,” Food and Chemical Toxicology, 43, 1569‐1593.
23. Kevan PG (1999) “Pollinators as bioindicators of the state of the environment: species, activity and diversity,”
Agriculture, Ecosystems & Environment, 74 (1-3), 373-393.
24. Anastassiades M, Lehotay SJ, Stajnbaher D, Schenck FJ (2003) “Fast and easy multiresidue method
employing acetonitrile extraction/partitioning and “dispersivc solid phase extraction” for the determination
pesticide residues in produce,” J. AOAC Int. 86, 412.
25. Macharia P (2004) Country overview. Kenya Soil Survey.
http://www.apipnm.org/swlwpnr/reports/y_sf/z_ke/ke.htm#menu (accessed on 21
st
July, 2016)
26. SANCO/12571/2013. “Guidance document on analytical quality control and validation procedures for
pesticide residues analysis in food and feed,” Available online:
http://ec.europa.eu/food/plant/plant_protection_products/guidance_documents/docs/qualcontrol_en.pdf
(accessed on 14 April 2015).
27. Mullin CA, Frazier M, Frazier JL et al. (2010) “High levels of miticides and agrochemicals in North American
apiaries: implications for honey bee health,” PloS ONE 5(3) Article ID e9754.
28. Gambacorta G, Faccia M, Lamacchia C, Di Luccia A and La Notte E. (2005) “Pesticide residues levels in
tomatoes grown in an open field,” Food Control., 16, 629-632.
29. Muli E, Patch H, Frazier M, Frazier J, Torto B, Baumgarten T, Kilonzo J, Kimani JN, Mumoki F, Masiga D,
Tumlinson J, Grozinger C (2014) “Evaluation of the distribution and impacts of parasites, pathogens, and
pesticides on honey bee (Apis mellifera) populations in East Africa,” PLOSone 9, 4
30. US EPA Pesticide Ecotoxicity Database of the Office of Pesticide Programs, Ecological Fate and Effects
Division. Available: http://www.ipmcenters.org/Ecotox/. Accessed 2016 March 8