Pollen source variation between agricultural areas and forest edges: a case study of the red dwarf honeybee (Apis florea Fabricius) in the Rarh region, West Bengal, India

Introduction

Honeybees forage on the flowers of a wide range of plant species to collect floral rewards, including nectar and pollen. Specifically, pollen serves as the sole natural source of protein for the development of honeybee brood (Liolios et al., 2015; Pang et al., 2022). In addition, it fulfils the bees’ requirements for lipids, sterols, amino acids, carbohydrates, polyphenols, vitamins, pigments, antioxidants, and minerals (Campos et al., 2008; Rzepecka-Stojko et al., 2015; Alshallash et al., 2023; Mărgăoan & Cornea-Cipcigan, 2024). Consequently, the survival of a honeybee colony is highly dependent on the pollen it collects. Pollen is required daily, as honeybees prefer freshly stored grains (Carroll et al., 2017) and tend to store only small amounts of pollen in the hive, sufficient for a few months (Layek et al., 2016). This limited reserve makes colonies particularly vulnerable to sudden fluctuations in environmental pollen availability (Pernal & Currie, 2001; Requier et al., 2017), underscoring the crucial role of pollen in honeybee colony survival. Gradually, various landforms are being converted into built-up areas for human habitation and agricultural land to meet rising food demand, leading to a decline in forested areas. Changes in landscape use patterns can significantly influence the foraging behaviour of honeybees (Steffan-Dewenter & Kuhn, 2003; Danner et al., 2016). Moreover, pollen foraging patterns can affect colony distribution, survival, and the spatial extent of pollination services (Joseph et al., 2020). Consequently, assessing honeybees' pollen collection patterns across different landscapes is essential for both humanity and global food security (Garibaldi et al., 2011; Van der Sluijs et al., 2016). Nonetheless, the extent to which resource distribution and habitat composition influence pollen collection at the landscape scale remains poorly understood. Understanding pollen foraging patterns is, however, invaluable for agricultural landscape planning, as it supports the development of optimised landscape designs that enhance pollination services, promote sustainable biodiversity, and improve crop yields (Brosi et al., 2008). 

Identifying the floral resources utilised by honeybees is essential for developing conservation strategies and optimising the ecosystem services associated with pollination (Aguiar et al., 2013; Melin et al., 2018). Palynological analysis of pollen loads is a reliable and widely employed method for determining the pollen sources of a bee species (Campos et al., 2008; Barth et al., 2010). It also provides insights into the foraging preferences of honeybees, ecological interactions (between plants and pollinators, as well as among different pollinator species), and the floral composition of the vegetation surrounding the colony. The red dwarf honeybee (Apis florea Fabricius) is native to Asia, with its natural range extending from Southeast Asia to the Persian Gulf, encompassing several countries, including India, Pakistan, Sri Lanka, China, Malaysia, and Thailand. This honeybee species is an important pollinator of many wild and cultivated plants (Abrol, 2010), and plays a vital role in conserving biodiversity and sustaining food security. Numerous researchers have conducted palynological studies to characterise the pollen foraging behaviour of this honeybee species (e.g., Kalpana et al., 1990; El Shafie et al., 2002; Suwannapong et al., 2013; Layek et al., 2015a, 2026). However, the pollen foraging patterns of this species remain poorly understood in West Bengal (e.g., Layek et al., 2015a, b, 2026). Information on how landscape composition (vegetation) affects the collection of floral resources by red dwarf honeybees is scarce. In particular, the extent to which landscapes influence pollen collection by honeybee species remains to be determined. However, this information is crucial for conserving native honeybee species in a changing landscape.

In this study, we aimed to uncover the impact of landscape composition (especially entomophilous flora) on pollen host selection of red dwarf honeybee (Apis florea); for this, we identified the polleniferous plants utilised by Apis florea across two distinct landscapes (agricultural and forest areas) through palynological analysis of pollen loads in the Rarh region of West Bengal. Specifically, we sought to address the following research questions: (1) Which plant species provide pollen for Apis florea? (2) Which are the key polliniferous plants for this honeybee species? (3) Does the palyno-diversity of their pollen diet vary between landscapes? We hypothesised that landscape composition may influence the pollen diets of honeybees and that agriculture-prone and forest-edge zones both have the potential to provide honeybees with pollen of different botanical compositions. 

Materials and methods

Experimental site

The study was carried out in the Rarh region (in Bankura district) of West Bengal, India, over two consecutive years, from 2023 to 2024 (Fig. 1). The Rarh region of West Bengal lies in the western part of the state, extending over districts such as Bankura, Purulia, Paschim Bardhaman, and parts of Birbhum. It is characterised by undulating uplands, lateritic soils, sparse vegetation, and a relatively dry climate compared to the humid Gangetic plains. The region forms a transitional zone between the Chota Nagpur Plateau and the alluvial plains of eastern West Bengal. The Rarh region, including Bankura district, has different landscape compositions, including forest, forest edge, and agricultural zones. Sampling sites were selected across two types of landscapes: (i) agricultural areas and (ii) forest areas. The agricultural sampling sites included Golakpur, Lalbandh, Mandi, Mamra, Rajpur, and Saldaha, while the forest sites comprised Bankadaha, Benachapra, Dhabani, Kharkharee, Koniary, Lachna, Majuria, Panchmura, and Telerajpur. The agricultural landscapes were primarily characterised by crop fields containing species such as Coriandrum sativum L., Cucumis sativus L., Luffa acutangula (L.) Roxb., Luffa aegyptiaca Mill., Momordica charantia L., Oryza sativa L., Rhamphospermum nigrum (L.) Al-Shehbaz, and Sesamum indicum (L.).

Fig 1. The map shows study sites within the Rarh region of West Bengal, India.

In contrast, the forest areas were dominated by tree species, including Acacia auriculiformis A. Cunn. ex Benth., Eucalyptus tereticornis Sm., Shorea robusta C. F. Gaertn., and Terminalia chebula (Retz.). Pedologically, the soils of the study sites are dominated by lateritic and red soils, with alluvial soils occurring in the eastern riverine plains. These soils are generally acidic, low in fertility, and sandy loam in texture, with laterization as the dominant soil-forming process in this humid tropical region. The region has a tropical climate with clearly defined seasons: summer (April–June), monsoon (July–August), post-monsoon or autumn (mid-September–mid-October), and winter (December–January). Summers are hot, with daytime temperatures ranging from 35 to 42 °C, whereas winters are relatively cool, with daytime temperatures between 7 and 15 °C. The study area receives approximately 1,400 mm of annual rainfall, the majority of which falls during the monsoon season.

Honeybee species

This study focused on the red dwarf honeybee, Apis florea, a species belonging to the family Apidae. This honeybee constructs single-comb, open nests on the branches of trees and shrubs, as well as on man-made structures (Layek & Karmakar, 2020). It exhibits migratory behaviour, and its occurrence (colony density) varies seasonally, reaching its peak from winter to summer (Layek & Karmakar, 2020).  

Pollen load collection

We collected corbicular pollen samples by directly capturing returning pollen foragers. With appropriate safety measures (using a bee veil), we approached the nests closely during foraging periods — in the morning (08:00–10:00 h), around noon (11:30–12:30 h), and in the afternoon (15:30–17:00 h). On each sampling day, 10–15 pollen foragers were captured using forceps, and the two corbicular pollen loads from each forager were transferred into a glass vial. FAA solution was then added to the vial, and the samples were preserved for subsequent analysis. Pollen samples were collected throughout the year from both agricultural and forest areas. In total, 806 pairs of pollen loads were collected (427 pairs of loads from agricultural areas and 379 pairs of loads from forest edges), comprising 20–49 pairs of loads per month for each vegetation zone (Table 1). 

Palynological analyses

The vials containing pollen samples were vigorously shaken to homogenise the solution. The samples were then centrifuged at 3,000 rpm for 5 minutes. After removing the supernatant, 2 mL of glacial acetic acid was added to the pollen sediment and stirred thoroughly. The mixture was again centrifuged at 3,000 rpm for 5 minutes. Following the removal of the supernatant, the resulting pollen sediment was treated with an acetolysis mixture of acetic anhydride and concentrated sulphuric acid in a 9:1 ratio (Erdtman, 1960). The mixture was well stirred, and the tube was placed in a water bath at 100 °C for 3 minutes. The acetolysed pollen sample was then centrifuged at 3,000 rpm, after which the sediment was rinsed with distilled water and centrifuged twice more. A small portion of the pollen sediment was transferred to the centre of a glass slide along with a drop of glycerine jelly. The jelly containing the pollen was gently warmed to melt it, then covered with a coverslip, and the edges were sealed with nail polish. The prepared slides were examined under a light microscope (Primo Star, Zeiss), and pollen micrographs were captured (with objective lenses of 20x [200x magnification], 40x [400x magnification] and 100x [1000x magnification]). Pollen grains were identified by comparing their morphological characteristics (including shape, size, aperture features, and exine ornamentation) with reference pollen slides prepared from the pollen of entomophilous plants found at the study sites. In most cases, identification was carried out to the species level (i.e., pollen species) for pollen obtained from bee pollen loads; however, when species-level identification was not possible, a pollen-type approach was adopted (Joosten & Klerk, 2002).  We estimated the relative abundance (RA) (also referred to as occurrence) of each pollen species or type based on the total number of corbicular pollen loads recorded over the study period (Layek et al., 2026). 

RA (%) = ni / N × 100

ni represents the number of pollen loads counted for the i pollen type, while N denotes the total number of pollen loads analysed.

The monthly percentages of the pollen types obtained were calculated. The pollen types were then categorised into five groups (with slight modification of Layek et al., 2026): (1) very frequent (>30%), (2) frequent (>20–30%), (3) moderate frequent (>10–20%), (4) less frequent (3–10%), and (5) rare (<3%). Regarding the landscape, we recorded the number of pollen types obtained from analysing bee pollen loads for each month. Month-wise pollen diversity was also calculated using the Shannon–Weaver diversity index (H') (Shannon & Weaver, 1949), and the evenness of pollen collection — that is, the extent to which different pollen types were equally represented — was assessed using Pielou’s (1977) evenness index (J') as follows: 

Here, pᵢ denotes the proportion of each pollen type within a given month, ln represents the natural logarithm, and S is the total number of pollen types recorded in that month. The value of H' ranges from 0 to greater than 1, with higher values indicating greater diversity. The value of J' ranges from 0 to 1, reflecting resource use from highly uneven (heterogeneous) to completely even (homogeneous).

Field surveys

Field surveys were conducted at regular intervals (approximately every 10–15 days) across the vegetation of the study sites. These surveys were used to prepare reference pollen slides from bee-visited entomophilous plants and to identify the collected pollen loads, primarily to species level. For each honeybee-visited plant species, we recorded the growth habit, inflorescence type, and floral traits, including flower size, shape, and colour. Flowers were classified into five size categories based on the length of floral parts: very small (<5 mm), small (5–15 mm), medium (>15–30 mm), large (>30–50 mm), and very large (>50 mm) (Layek et al., 2026). Flower shapes were categorised following Dafni (1992), with minor modifications. For capitula and umbels, the overall size and shape of the inflorescence were considered rather than individual flowers. Flower colours were recorded according to human visual perception.

Statistical analyses

We performed descriptive statistics to estimate the mean and standard deviation of the datasets. Within each group, we first assessed the key assumptions for parametric tests: normality (using the Shapiro–Wilk test), homoscedasticity (via scatter plots), homogeneity of variances (using Levene’s test), and the absence of outliers (through box plots). If the assumptions were not satisfied, the non-parametric Mann-Whitney U test was used to compare means between groups (for example, number of pollen types, values of the pollen diversity index H’ and equitability index J’ in agricultural areas versus forest edges). The obtained p-values were considered significant at the 0.05 level. All analyses were carried out using SPSS software (version 26.0).

Results

Pollen types in pollen loads

Each pair of pollen loads (collected from a single returning pollen forager) was monofloral in origin. A total of 57 pollen types, belonging to 23 plant families, were identified from the analysis of 806 pairs of corbicular pollen loads (Table 2, Fig. 2). Thirty-seven pollen types originated from agricultural areas, while 41 were from forest edges. The most represented plant families were Cucurbitaceae (8 pollen types), Fabaceae (8 pollen types), Anacardiaceae (5 pollen types), Asteraceae (4 pollen types), and Lamiaceae (4 pollen types). The families contributing the highest proportions of pollen loads were Apiaceae (6.66%), Arecaceae (7.82%), Brassicaceae (8.44%), Cucurbitaceae (22.20%), Fabaceae (12.28%), and Myrtaceae (9.80%). In agricultural areas, the most abundant pollen types were Rhamphospermum nigrum (RA = 11.24%), Cucurbita maxima (RA = 9.84%), Luffa aegyptiaca (RA = 9.13%), Coriandrum sativum (relative abundance, RA = 7.96%), and Cucumis sativus (RA = 7.72%). In contrast, in forest edges, the dominant pollen types were Eucalyptus tereticornis (RA = 11.60%), Acacia auriculiformis (RA = 9.76%), and Terminalia arjuna (RA = 6.06%). On a monthly basis, in agricultural areas, pollen types with higher frequencies (classified as very frequent and/or frequent) included Acacia auriculiformis, Coccinia grandis, Cucumis sativus, Cucurbita maxima, Helianthus annuus, Luffa aegyptiaca, Rhamphospermum nigrum, Sesamum indicum, and Trichosanthes curcumina (Table 3). In forest edges, the more frequent pollen types were Acacia auriculiformis, Eucalyptus tereticornis, Pithecellobium dulce, Rhamphospermum nigrum, Terminalia arjuna, and Ziziphus mauritiana (Table 4).

Month-wise, the number of pollen types ranged from 6 to 11 (8.33 ± 1.44) in agricultural areas and from 5 to 11 (7.08 ± 2.27) at forest edges (Table 5). In agricultural areas, the lowest number of pollen types was recorded in September and October compared with other months. At forest edges, a greater number of pollen types were observed in March, April, May, and August. The number of pollen types did not differ significantly between the two landscape types (agricultural areas and forest edges) (Table 5). Monthly values of the Shannon–Weaver diversity index (H’) ranged from 1.71 to 2.15 (1.92 ± 0.14) in agricultural areas and from 1.50 to 2.19 (1.75 ± 0.28) at forest edges; Pielou’s evenness index (J’) ranged from 0.85 to 0.96 (0.91 ± 0.04) in agricultural areas and from 0.86 to 0.94 (0.91 ± 0.03) at forest edges. Landscape-wise, the values of both H’ and J’ showed no variation.

Characteristics of honeybee-visited plants

The red dwarf honeybee species visited a wide range of plants (56 species identified at the species level based on pollen types), exhibiting considerable diversity in plant origin, growth habit, inflorescence type, and floral traits such as size, shape, and colour. They foraged on both wild (75%) and cultivated plants, including agronomic–horticultural (23.21%) and ornamental species (1.79%) (Supplementary Table 1). Among the visited plants, the majority were trees (53.57%), followed by herbs, shrubs, and climbers. The honeybees showed a preference for a variety of inflorescence types, with cymes (25.00%), racemes (14.28%), solitary (14.28%), panicles (12.50%), and spikes (12.50%) being the most common. Flower sizes ranged from small to very large, although small (55.36%) and medium-sized (26.79%) flowers were the most frequently visited. Flower shapes were also diverse, with dish-shaped (28.57%), brush-shaped (19.64%), and stellate forms (17.86%) being the most prevalent. In terms of flower colour, white (23.21%), cream (21.43%), and yellow (21.43%) were the most commonly visited colours.  

Discussion

Considering the composition of individual bees’ pollen loads (each pair collected by a single forager), all were monofloral in botanical origin, indicating that the foragers exhibited marked floral constancy during their foraging trips. Floral constancy is a well-documented phenomenon among various honeybee species, including Apis cerana (Raj et al., 1993; Mondal et al., 2023), Apis dorsata (Layek et al., 2023), Apis florea (Layek et al., 2015a,b), and Apis mellifera (Grüter et al., 2011; Layek et al., 2020). The constancy of flower visitation strongly influences a pollinator’s efficiency and its contribution to ecosystem services. The honeybee species collected pollen from a wide range of plant species in both agricultural areas and forest edges, indicating its polylectic nature, which has also been recognised by other researchers (e.g., Layek et al., 2015a,b, 2020). Such polylectic behaviour enhances the nutritional quality of pollen diets, improves adaptability across diverse habitats, and supports colony sustainability.

In the present study, Anacardiaceae, Cucurbitaceae, and Fabaceae were recorded as having the highest number of pollen types, likely due to their abundance in the study area and honeybees' preference for their flowers. The principal pollen-contributing families were Apiaceae, Arecaceae, Brassicaceae, Cucurbitaceae, Fabaceae, and Myrtaceae. The significance of these families as major pollen sources for honeybees is well recognised globally (Andrada & Tellería, 2005; Layek et al., 2015a,b, 2020; Alves & Santos, 2017; Mondal et al., 2023). Many crop species in Apiaceae, Brassicaceae, and Cucurbitaceae were cultivated within the study areas, resulting in their dominance in the pollen spectra. In contrast, the families Arecaceae, Fabaceae, and Myrtaceae mainly include wild and cultivated trees that were common within the study zones, accounting for their substantial representation in the collected pollen loads.

Fig. 2. Photomicrographs of pollen types obtained from pollen loads of Apis florea in Bankura district, West Bengal, India. (A–B) Anacardium occidentale, (C) Borassus flabellifer, (D) Cocos nucifera, (E) Coriandrum sativum, (F) Eucalyptus tereticornis, (G) Helianthus annuus, (H–I) Neolamarckia cadamba, (J) Phoenix sylvestris, (K) Psidium guajava, (L–M) Rhamphospermum nigrum, (N–O) Ricinus communis, (P) Sesamum indicum, (Q–R) Terminalia arjuna, (S–T) Ziziphus mauritiana. Scale bars = 10 µm.

The mother plants that produced (and provided to honeybees) pollen grains differed between agricultural areas and forest edges. In agricultural areas, a large proportion of pollen originated from crops such as Coriandrum sativum, Cucumis sativus, Cucurbita maxima, Luffa aegyptiaca, and Rhamphospermum nigrum.

In the present study, Anacardiaceae, Cucurbitaceae, and Fabaceae were recorded as having the highest number of pollen types, likely due to their abundance in the study area and honeybees' preference for their flowers. The principal pollen-contributing families were Apiaceae, Arecaceae, Brassicaceae, Cucurbitaceae, Fabaceae, and Myrtaceae. The significance of these families as major pollen sources for honeybees is well recognised globally (Andrada & Tellería, 2005; Layek et al., 2015a, 2020; Alves & Santos, 2017; Mondal et al., 2023). Many crop species in Apiaceae, Brassicaceae, and Cucurbitaceae were cultivated within the study areas, resulting in their dominance in the pollen spectra. In contrast, the families Arecaceae, Fabaceae, and Myrtaceae mainly include wild and cultivated trees that were common within the study zones, accounting for their substantial representation in the collected pollen loads. The mother plants that produced (and provided to honeybees) pollen grains differed between agricultural areas and forest edges. In agricultural areas, a large proportion of pollen originated from crops such as Coriandrum sativum, Cucumis sativus, Cucurbita maxima, Luffa aegyptiaca, and Rhamphospermum nigrum. In contrast, at forest edges, a greater share came from tree species including Acacia auriculiformis, Eucalyptus tereticornis, and Terminalia arjuna.

The importance of several of these species (e.g., Acacia auriculiformis, Eucalyptus tereticornis, Rhamphospermum nigrum, and Terminalia arjuna) as valuable pollen sources for honeybees has been documented in earlier studies (Laxmikant & Devendra, 2014; Layek et al., 2015a). However, the identification of Cucurbita maxima as an important pollen source for honeybees is newly reported from West Bengal.

Month-wise, the number of pollen types remained largely consistent across the two landscape types, agricultural areas and forest edges. In agricultural areas, monoculture is unlikely to be extensive, allowing honeybee species to collect pollen from a wide range of plant species. This may positively reflect the benefits of nutritionally diverse pollen diets and the conservation of wild honeybee populations. The month-wise values of H’ and J’ were also comparable between the two landscapes (in agricultural areas: H’ = 1.92 ± 0.14, J’ = 0.91 ± 0.04; in forest edges: H’ = 1.75 ± 0.28, J’ = 0.91 ± 0.03). These values are consistent with those reported in other studies on social bees (e.g., Sosa-Najera et al., 1994; Novais et al., 2013; Layek et al., 2021). The consistently high H’ values suggest a considerable diversity of bee-visited plants. The diversity of pollen types depends on the availability of floral resources, the flowering duration of utilised plant species, and the foraging behaviour of the bees. Moreover, the high values of Pielou’s evenness (J′) index indicate limited dominance by any single or a few pollen types and a nearly equal representation of all pollen types within the pollen spectra. Such a diversity of bee forages and reduced dominance of any particular floral resource within a landscape could enhance bee activity and diversity (Moreira et al., 2015) and may contribute to improved honeybee survival and honey production.

Regarding categorisation of honeybee-visited plants, it was observed that the red dwarf honeybee species primarily visited wild plants, as well as some agri-horticultural and ornamental species. Within the selected zones, particularly in agricultural areas, the honeybee species collected pollen from several crop plants (e.g., Allium cepa, Coriandrum sativum, and Rhamphospermum nigrum). The significance of agri-horticultural and ornamental plants as sources of floral rewards for honeybees is well established (e.g., Abrol, 2010; Layek et al., 2015a; Gupta et al., 2020). Hence, in addition to wild plants, a diverse range of cultivated species plays a crucial role in sustaining native honeybee populations. Among the bee-visited plants, trees were the most frequently foraged. Our findings corroborate previous studies on honeybees (e.g., Layek & Karmakar, 2020; Layek et al., 2020), which suggest that foraging on tree flowers may offer greater security and reduced disturbance for the foragers. The honeybee species exploited a variety of inflorescence types, with cyme, raceme, solitary, panicle, and spike flowers being the predominant types. These inflorescence types may be more attractive to, or more readily available around, honeybee colonies. Most bee-visited plants possessed small or medium-sized flowers. Although a single small flower may contain a limited amount of pollen, pollen host selection in honeybees is governed by multiple factors, including flower availability, pollen quantity and quality, rather than the pollen content of individual flowers (Pernal & Currie, 2001; Layek et al., 2020; Lan et al., 2021). The red dwarf honeybees visited flowers of diverse shapes, with a predominance of dish-, brush-, and stellate-shaped flowers. In brush-shaped flowers, bees can easily gather pollen from numerous exposed stamens (Nates-Parra & Rodríguez, 2011), while dish- and stellate-shaped flowers provide a stable landing platform for collecting floral rewards. The bee-visited flowers exhibited a variety of colours, with white, cream, and yellow being the most common. This finding is consistent with previous studies on honeybees (e.g., Layek et al., 2020) and stingless bees (e.g., Vossler, 2012; Obregón & Nates-Parra, 2014). Considering the wide range of characteristics of bee-visited plants, it may be concluded that the red dwarf honeybee species is a generalist forager. Such generalist behaviour enables honeybees to exploit a broad spectrum of plants to meet their nutritional requirements and enhance the sustainability of their colonies (Kaluza et al., 2017).

Conclusion

The present study demonstrated that the red dwarf honeybee (Apis florea) foraged on a wide variety of pollen sources across both agricultural areas and forest edges. Certain plant families, such as Apiaceae, Arecaceae, Brassicaceae, Cucurbitaceae, Fabaceae, and Myrtaceae, were consistently dominant within the pollen spectra. In agricultural areas, a substantial proportion of pollen loads originated from crop plants (e.g., Coriandrum sativum, Cucumis sativus, Cucurbita maxima, Luffa aegyptiaca, and Rhamphospermum nigrum). In contrast, at forest edges, the majority were derived from trees (e.g., Acacia auriculiformis, Eucalyptus tereticornis, and Terminalia arjuna). The number and diversity of pollen types did not differ significantly between the two landscapes. Based on the characteristics of the bee-visited plants, the honeybee species was identified as a generalist forager, collecting floral resources from a wide variety of flower types. These findings highlight the ecological adaptability of Apis florea in utilising available floral resources across diverse habitats and emphasise the importance of maintaining heterogeneous landscapes to meet foraging requirements and conserve native honeybee populations.

Author's Contributions

Suman Kumar De: methodology, investigation, formal analysis, data curation, writing – original draft. Ujjwal Layek: investigation, formal analysis, writing, review & editing. Prakash Karmakar: conceptualization, supervision, writing- review and editing.

Author's Information

Funding

This research work did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data Availability Statement

All data supporting the findings of this study are available in the paper or in the supplementary files.

Acknowledgments

 We thank the authorities of Vidyasagar University for providing the necessary laboratory facilities for light microscopic studies. We are also thankful to Trisha Bhandari (research scholar, Department of Botany & Forestry, Vidyasagar University) for helping us prepare reference pollen slides.

Ethics Approval and Consent to Participate

This article does not contain any studies with human participants performed by any of the authors.

Consent for Publication

Not applicable

Conflict of Interest

The authors declare that they have no known competing interests.

 Generative AI statement

The authors declare that no Gen AI was used in the creation of this manuscript.