5dd31996-6689-4fe3-8b24-d8af99ee0628 https://cloud.gbif.org/africa/resource?r=insect_pollinators_in_rwandan_agroecosystems Insect pollinators in Rwandan agroecosystems Elie Sinayitutse Nature Rwanda

Musanze Northern RW

+250788988150 sinayitutse@naturerwanda.org Elie Sinayitutse

Musanze Northern RW

Theodette Imanizabera

Musanze Norther RW

Jean Claude Dusabimana

Musanze Northern RW

Jean de Dieu Nsenganeza

Kigali Kigali City RW

Christelle Suavis Iradukunda

Kigali Kigali City RW

2026-01-13 eng Pollinators in Rwanda remain poorly studied despite their significant economic value, estimated to contribute over US $100 million annually to crop production. While some research has been conducted on honeybees, little is known about other pollinators such as butterflies, beetles, moths, and others. A major gap exists in comprehensive baseline data on pollinator diversity and distribution, making it difficult to design effective conservation strategies. We used sweep nets and pan traps to sample insect pollinators along 1 km transects in agricultural areas across 5 agroecosystems. Surveys documented pollinator abundance, diversity, and their associations with crop flowering stages and habitat structure. A total of 99 insect species were recorded, with Apis mellifera emerging as the predominant species across all sites, indicating that these agroecosystems continue to support essential pollination services. Hymenoptera dominated other orders, reflecting their critical role in crop pollination. Across all agroecosystems, pesticide application was consistently identified as the primary threat, with notably reduced pollinator abundance and richness in areas of frequent and recent pesticide use, signaling ecosystem stress. Pollinator presence varied with crop flowering stages and habitat structure, highlighting the importance of temporal and spatial factors in pollinator community dynamics. We observed that reducing pesticide pressure, raising awareness on pollinator importance and strengthening sustainable pest management are essential for pollinator conservation and agricultural resilience. Insect pollinators agroecosystems pesticides pollinator conservation sustainable agriculture GBIF Dataset Type Vocabulary: http://rs.gbif.org/vocabulary/gbif/dataset_type_2015-07-10.xml To the extent possible under law, the publisher has waived all rights to these data and has dedicated them to the Public Domain (CC0 1.0). Users may copy, modify, distribute and use the work, including for commercial purposes, without restriction. Creative Commons Zero v1.0 Universal https://spdx.org/licenses/CC0-1.0.html CC0-1.0 http://naturerwanda.org This dataset encompasses pollinator surveys conducted across five distinct agroecosystems in Rwanda. Sampling sites included: Rwakigeri village in Cyinzovu cell, Kabarondo sector, Kayonza district, specifically at the KWIIP farming project around Rugozi dam; Rugunga village in Biryogo cell, Gashora sector, Bugesera district, in the agricultural areas surrounding Lake Mirayi; Musenyi village in Murama cell, Kinyinya sector, Gasabo district, along the marshland of the Mbonwa-Rufigiza River; Rugarama village in Kibali cell, Byumba sector, Gicumbi district; and Muyebe village in Ruhango cell, Rongi sector, Muhanga district, located opposite to Busaga Forest. 29.202 30.795 -1.395 -2.713 2025-11-25 2025-12-06 Insect species, identified to the genus and species level species Eristalis arbustorum genus Afreumenes genus Anterhynchium species Anterhynchium natalense genus Anthrenus genus Apis species Apis mellifera genus Asarkina genus Belenois species Belenois creona genus Belonogaster genus Bibylia genus Bicyclus genus Borbo species Borbo fatuellus genus Brachicoma genus Brachymeria genus Catopsilia species Catopsilia florella genus Ceratina genus Ceriana genus Ceryx genus Cetonia genus Chlorocala genus Chrysomya species Chrysomya chloropyga genus Chrysotoxum genus Colletes genus Colotis species Colotis antevippe species Colotis euippe genus Condylostylus genus Ctenochares genus Cyana species Danaus chrysippus genus Decapotoma species Episyron histrio genus Episyrphus species Eretis umbra genus Eristalinus species Eristalinus taeniops genus Eristalis genus Eupeodes genus Eurema species Eurema hecabe genus Galerucinae genus Glyphodes genus Halictus species Hyalites hiponina genus Hypolimnas genus Hypotrigona genus Isodontia genus Junonia species Junonia hierta species Junonia oenone species Junonia sophia species Junonia terea genus Lasioglossum species Leptosia alcesta genus Longitarsus species Lucilia sericata genus Megachile species Megachile combusta genus Melanagromyza genus Musca species Musca domestica genus Mylabris species Neptis serena genus Pachnoda genus Papilio genus Paragus genus Promecesse genus Psyllobora genus Ropalidia genus Sarcophaga genus Scaeva genus Sphaerocoris genus Sphaerophoria species Sphaerophoria scripta genus sphex species Spialia diomus genus Synagris species Synagris analis species Syritta pipiens genus Syrphus species Syrphus ribesii species Tenebrio molitor species Thyreus nitidulus genus Trigona genus Vespula genus Xanthogramma genus Xylocopa species Xylocopa caffra species Xylocopa flavorufa species Xylocopa inconstans species Xylocopa nigrita species Xylocopa virginica species Ypthima albida species Zizina antanossa This dataset provides baseline information on the diversity and distribution of insect pollinators across five studied agroecosystems in Rwanda. Standardized surveys were conducted along 1 km transects using sweep nets and pan traps to record insect pollinators, alongside contextual observations on land use, crop types, flowering stages, habitat structure, and potential threats.The dataset documents 99 insect pollinator species representing 5 orders and 31 families. Hymenoptera was the most dominant order (99 records), followed by Diptera (59 records) and Lepidoptera (53 records). At the family level, Apidae was the most abundant (69 records), followed by Syrphidae (39 records) and Nymphalidae (24 records). The Western Honey Bee (Apis mellifera) was the most frequently recorded species across all surveyed areas. Preliminary findings indicate pesticide application as the primary threat influencing pollinator abundance and richness, with notably reduced diversity in areas of frequent and recent pesticide use. Pollinator presence varied with crop flowering stages and habitat structure, highlighting the importance of temporal and spatial factors in community dynamics. This dataset establishes an essential ecological baseline to support future research, conservation planning, and sustainable agricultural management in Rwanda. Pollinators play a critical role in sustaining agricultural productivity, biodiversity, and ecosystem resilience, yet their diversity, distribution, and ecological roles remain poorly documented in Rwanda. Existing research has focused primarily on honeybees, with comparatively little attention given to other pollinator groups such as butterflies, beetles, moths, and flies. As a result, there is a major lack of comprehensive baseline data needed to inform effective conservation planning and sustainable agricultural management across the country.Recent studies underscore both the importance and vulnerability of pollinators in Rwanda. Insect pollinators are estimated to contribute more than USD 100 million annually to national crop production, highlighting their significant economic value (Yocgo et al. 2023). However, this value remains underappreciated due to limited ecological and spatial data. Pollinator populations are increasingly threatened by habitat loss, pesticide use, and climate change, yet the specific impacts of these pressures across Rwanda's agroecosystems are not well understood. Research conducted in Musanze District (Ndayiragije 2021) and Huye District (Uwingabire et al. 2021) has provided localized insights, but such studies are spatially restricted and insufficient to guide nationwide conservation or policy responses.In addition to data gaps, pollinator conservation is not yet well integrated into agricultural and environmental policies, and public awareness of pollinator importance remains low. Many farming landscapes lack pollinator-friendly habitats, and farmers often have limited guidance on conservation-compatible practices, increasing the risk of declining pollination services and food insecurity, particularly for smallholder farmers.This dataset was generated as part of a broader project designed to address these gaps by establishing baseline data on insect pollinator diversity, distribution, and threats across five key agroecosystems in Rwanda. The data support analyses of pollinator occurrence patterns in relation to land use, crop types, flowering stages, habitat structure, and management practices, and can be used to inform future research, habitat restoration, awareness-raising initiatives, and policy development. By making these data openly available through GBIF, the dataset contributes to national and global efforts to understand pollinator decline and to promote evidence-based strategies for sustaining pollination services and agricultural resilience. This dataset follows the GBIF Sampling Event data model and documents insect pollinator observations across five agroecosystems in Rwanda. It consists of a Sampling Event core and an associated Occurrence extension, reflecting the hierarchical field sampling design.The Sampling Event core includes different records corresponding to occurrences at sampling stations along transects, with each event identified by a unique eventID. Event-level information includes sampling date, protocol, geographic location, eventRemarks and other necessary information based on the Sampling Event Datasets.The Occurrence extension contains species-level records linked to the core via eventID. Each record represents a single insect pollinator species detected during a sampling event and includes a unique occurrenceID, taxonomic identification, and individual counts. Multiple occurrence records may be associated with a single sampling event, reflecting the detection of multiple species at the same sampling station.Data were collected using standardized 1 km transect surveys (Popic et al., 2012) employing two complementary methods: sweep netting and pan traps (Toler et al., 2005; Roulston et al., 2007; Grundel et al., 2011). All files are provided in Darwin Core–compliant formats and can be analyzed using standard biodiversity data tools and software, including spreadsheet applications, R, and GIS platforms, without the need for specialized software. We gratefully acknowledge the JRS Biodiversity Foundation for their generous financial support, which enabled the implementation of this study. We thank the Rwanda Environment Management Authority (REMA) for providing the recommendation letter that facilitated the successful execution of this work. We extend our sincere appreciation to Jean de Dieu Nsenganeza, and Theodette Imanizabera for their valuable contributions to field data collection. We further acknowledge the support of local administrative authorities and community members for their guidance and for granting access to their farms and study sites. The dataset will be updated as required. asNeeded Elie SINAYITUTSE Nature Rwanda Head of Species and Habitat Conservation

NM11ST MUSANZE Musanze Northern RW

+250788988150 sinayitutse@naturerwanda.org http://www.naturerwanda.org https://orcid.org/my-orcid?orcid=0000-0002-4848-0416 Jean Claude DUSABIMANA Nature Rwanda Executive Director

NM11ST MUSANZE Musanze Northern RW

+250788869445 dusabimana@naturerwanda.org http://naturerwanda.org The project followed a systematic workflow to ensure data quality and GBIF compliance. Initial steps involved developing data collection materials by consulting existing GBIF datasets to identify relevant fields and standards, followed by designing custom data sheets aligned with Darwin Core terminology. These data sheets were deployed in KoboToolbox to enable easy digital data capture in the field. Prior to fieldwork, the survey team received comprehensive training on standardized data collection methodologies, species identification, and proper use of digital tools, and underwent a field trial practice. Following data collection, extensive data cleaning was conducted to align records with GBIF dataset standards, including taxonomic verification, coordinate validation, and formatting consistency checks. The final step involved preparing and publishing the cleaned dataset to GBIF, making it accessible for biodiversity research and conservation planning. The study was conducted across five districts in Rwanda: Kayonza (Kabarondo sector, Cyinzovu cell, Rwakigeli village), Bugesera (Gashora sector, Biryogo cell, Rugunga village), Gasabo (Kinyinya sector, Murama cell, Musenyi village), Gicumbi (Byumba sector, Kibali cell, Rugarama village), and Muhanga (Rongi sector, Ruhango cell, Muyebe village). These sites were selected to represent variation in agroecological conditions, elevation, and agricultural practices across different provinces of Rwanda. Insect pollinators were sampled using standardized transect methods along 1 km transects (Popic et al., 2012) at each site. Two complementary sampling techniques were employed to capture a broad range of pollinator taxa: sweep netting and pan traps (Toler et al., 2005; Grundel et al., 2011). Sweep netting involved actively collecting insects along the transect, targeting pollinators visiting flowers and foraging in vegetation. Pan traps, consisting of colored bowls filled with soapy water, were strategically placed along transects to passively capture flower-visiting insects. Surveys were conducted during daylight hours (between 6:00 a.m. and 3:00 p.m.) when pollinator activity is highest, ensuring consistent sampling effort across all sites and enabling comparability of abundance and diversity metrics. To enhance species detection, opportunistic observations will complement structured surveys (Fuster et al., 2020). This involves documenting pollinator visitations encountered outside formal observation periods in the same study area, valuable for detecting rare species, unusual interactions, or taxa underrepresented in fixed-time sampling. Rigorous quality control measures were implemented to ensure data accuracy and reliability, following established protocols for insect pollinator surveys. Only insects observed visiting flowers or engaging in pollination-related behaviors were recorded to capture pollination services. When field identification proved difficult, flower-visiting insects were captured with hand nets, preserved in alcohol (70% ethanol) (O'Connor et al., 2019) for later morphological identification. Specimens were stored in labeled tubes or glassine envelopes, recording date, site, plant species, capture time, and collector identity (Uwizelimana et al., 2022). Specimens collected via sweep netting and pan traps were carefully preserved and transported to the laboratory for taxonomic identification by qualified entomologists. Double-counting was prevented by implementing systematic sampling protocols along transects, ensuring that the same individual was not recorded multiple times during active collection. Pan traps were placed at fixed intervals along transects and monitored for standardized periods to maintain consistency (Toler et al., 2005; Roulston et al., 2007; Grundel et al., 2011). Insects incidentally captured but not associated with pollination activities were excluded from the dataset to maintain focus on pollination-relevant observations. All specimens were identified to the lowest taxonomic level possible, with voucher specimens retained for verification and future reference, adhering to standardized practices in pollinator biodiversity surveys (Popic et al., 2012). This approach ensures comparability across sites and seasons while maintaining the scientific rigor necessary for baseline ecological assessments. Elie SINAYITUTSE https://orcid.org/my-orcid?orcid=0000-0002-4848-0416 author Jean Claude DUSABIMANA https://www.linkedin.com/in/jean-claude-dusabimana-8992692a/ reviewer This dataset provides comprehensive baseline data on insect pollinator diversity, distribution, and threats across five key agroecosystems in Rwanda, where pollinator information remains critically limited despite their substantial economic contribution to national crop production. Standardized surveys were conducted along 1 km transects using sweep nets and pan traps at five distinct sites: Kayonza district (KWIIP farming project around Rugozi dam), Bugesera district (around Lake Mirayi), Gasabo district (marshland along Mbonwa-Rufigiza River), Gicumbi district (Rugarama village), and Muhanga district (opposite Busaga Forest). These sites represent diverse agricultural landscapes across different ecological zones, capturing variation in habitat structure, crop types, and management practices. A total of 99 insect pollinator species were recorded, representing 5 orders and 31 families. Hymenoptera was the most dominant order (99 records), followed by Diptera (59 records) and Lepidoptera (53 records). At the family level, Apidae dominated with 69 records, followed by Syrphidae (39 records) and Nymphalidae (24 records). The Western Honey Bee (Apis mellifera) emerged as the most frequently recorded species across all surveyed areas, indicating that these agroecosystems continue to support essential pollination services. Pesticide application was consistently identified as the primary threat, with notably reduced pollinator abundance and richness in areas of frequent and recent pesticide use, signaling ecosystem stress. Pollinator presence varied significantly with crop flowering stages and habitat composition, highlighting the importance of temporal and spatial factors in community dynamics. The dataset follows the GBIF Sampling Event data model and is provided in Darwin Core–compliant formats. It establishes an essential ecological baseline to support future research, conservation planning, sustainable agricultural management, and policy development in Rwanda, contributing to national and global efforts to understand pollinator decline and promote evidence-based conservation strategies. The project was funded by the JRS Biodiversity Foundation This dataset presents data from the first phase of a nationwide pollinator assessment conducted across five strategically selected sites in Rwanda. The study covered one site in each of five districts representing distinct agroecological zones: Kayonza District (Kabarondo Sector) and Bugesera District (Gashora Sector) in the Eastern Province; Muhanga District (Nyamabuye Sector) in the Southern Province; Gicumbi District (Byumba Sector) in the Northern Province; and Gasabo District (Kinyinya Sector) in the City of Kigali. These sites were purposively selected due to their significant contribution to national agricultural production and their contrasting climatic and topographic conditions, which together provide a representative overview of Rwanda’s farming systems. Kayonza and Bugesera districts are located in the Eastern Province, characterized by relatively low elevations (approximately 1,200–1,500 m above sea level) and semi-arid to sub-humid climatic conditions. Annual rainfall in these areas is generally lower than the national average, typically ranging between 800 and 1,000 mm, with pronounced dry seasons. Agriculture in these districts is dominated by mixed cropping systems and irrigated farming, making them particularly relevant for assessing pollinator dynamics under water-limited and pesticide-intensive conditions. Muhanga District, situated in the Southern Province, lies within Rwanda’s central plateau at moderate elevations (approximately 1,700–2,200 m). The area experiences sub-humid climatic conditions, with annual rainfall commonly ranging between 1,200 and 1,400 mm. Farming systems in Muhanga are diverse and include mixed crops, horticulture, and agroforestry practices, providing an important contrast to the drier eastern landscapes. Gicumbi District in the Northern Province is characterized by higher elevations (approximately 1,800–2,200 m) and a cooler, wetter climate, with annual rainfall often exceeding 1,300 mm. The district supports intensive smallholder agriculture, including perennial crops and mixed farming systems, and represents highland agroecosystems where climatic conditions and land-use patterns differ markedly from lowland regions. Gasabo District (Kinyinya Sector), located in the City of Kigali, occupies an intermediate elevation range (approximately 1,500–1,800 m) and experiences moderate rainfall of around 1,000–1,200 mm annually. The area reflects a peri-urban agricultural landscape, where farming activities coexist with urban expansion, offering insights into pollinator dynamics in human-dominated environments. The study was designed as a stratified, observational field survey to document insect pollinator diversity, distribution, and associated environmental conditions across representative agroecosystems in Rwanda. The overall project is structured to be implemented in three main phases, each aligned with different flowering seasons, in order to capture seasonal variation in pollinator activity and crop–plant–pollinator interactions. The first phase, which forms the basis of this dataset, targeted five purposively selected sites across different districts and provinces, chosen to represent variation in climate, elevation, land use, and agricultural practices. Subsequent phases are intended to expand temporal coverage and strengthen understanding of seasonal dynamics in pollination services. Within each site, surveys were conducted using standardized 1 km transect sampling along agricultural areas. Insect pollinators were collected and observed using two complementary methods: sweep netting and pan traps (Popic et al., 2012; Toler et al., 2005; Roulston et al., 2007; Grundel et al., 2011). This consistent sampling effort enables comparability across sites and seasons while capturing a broad range of insect pollinator taxa with different foraging behaviors and habitat preferences. The study design is grounded in the premise that improved empirical evidence on pollinator diversity, seasonal dynamics, and threats is essential for informing policy response, conservation advocacy, and sustainable agricultural strategies. 2026-01-09T17:28:44.132+00:00 dataset Sinayitutse E, Imanizabera T, Dusabimana J C, Nsenganeza J D D, Iradukunda C S (2026). Baseline Assessment of Insect Pollinators in Agroecosystems of Rwanda: a baseline perspective for pollinator conservation. Version 1.6. Nature Rwanda. Samplingevent dataset. https://cloud.gbif.org/africa/resource?r=insect_pollinators_in_rwandan_agroecosystems&v=1.6 Fuster F., Kaiser-Bunbury N.C, and Traveset A, 2020. Pollination effectiveness of specialist and opportunistic nectar feeders influenced by invasive alien ants in the Seychelles https://doi.org/10.1002/ajb2.1499 Uwingabire, Z., Gallai, N., & Kephaliacos, C. (2021). Vulnerability analysis of food production and nutrient consumption on pollinators decline: The Case of Smallholder Farm Households in Huye District in Southern Rwanda. In 17th GLOBELICS INTERNATIONAL CONFERENCE “Innovation Systems and Sustainable Development: New Strategies for Growth, Social Welfare and Environmental Sustainability (p. 28). https://hal.science/hal-04842083/document Ndayiragije, P. S. (2021). Assessment of Aboveground Insects Associated Within Farm Fields of Musanze District in Rwanda: Abundance and Diversity. Yocgo, R. E. E., Hitimana, I., Hakizimana, M., & Birachi, E. A. (2023). Insect pollinators can unlock an annual monetary value of more than US $100 million from crop production in Rwanda. Scientific Reports, 13(1), 20108. Toler R.T., Evans W.E and, & Tepedino V.J. 2005. Pan-trapping for bees (Hymenoptera: Apiformes) in Utah's West Desert: the importance of color diversity the pan-pacific entomologist 81(3/4):103–113, (2005). Popic T.J, Wardle G.M., Davila Y.C, 2012. Flower-visitor networks only partially predict the function of pollen transport by bees. https://doi.org/10.1111/j.1442-9993.2012.02377.x Uwizelimana J.D, Nsabimana D. and Wagner T. 2022. Diversity and distribution of Fruit- feeding butterflies (Lepidoptera: Nymphalidae) in Nyungwe National Park, Rwanda. DOI: 10.1111/aje.12997 Grundel R., Frohnapple J.K., Jean R., and Pavlovic B.N. 2011. Effectiveness of Bowl Trapping and Netting for Inventory of a Bee Community. Environmental Entomology, Volume 40, Issue 2, 1 April 2011, Pages 374–380, https://doi.org/10.1603/EN09278 O'Connor R.S., Kunin W.E., Garratt M.P.D, Potts G.S, Roy E.H, Andrews C, Jones M. C, Peyton M.J., Savage J, Harvey M.J, Morris R.K, Roberts S. P. M., Wright I., J. Vanbergen, and Carvell C. 2018. Monitoring insect pollinators and flower visitation: The effectiveness and feasibility of different survey methods. Methods Ecol Evol. 2019; 10:2129–2140. 5dd31996-6689-4fe3-8b24-d8af99ee0628/v1.6.xml