Data from: Agricultural specialisation increases the vulnerability of pollination services for smallholder farmers
Data files
Jul 10, 2024 version files 2.78 MB
-
Plant_pollinator_visitation_data.xlsx
2.47 MB
-
Pollen_load_data.xlsx
305.10 KB
-
README.md
3.26 KB
Abstract
Smallholder farms make up 84% of all farms worldwide and feed two billion people. These farms are heavily reliant on ecosystem services and vulnerable to environmental change, yet under-represented in the ecological literature. The high diversity of crops in these systems makes it challenging to identify and manage the best providers of an ecosystem service, such as the best pollinators to meet the needs of multiple crops. It is also unclear whether ecosystem service requirements change as smallholders transition towards more specialised commercial farming – an increasing trend worldwide. Here, we present a new metric for predicting the species providing ecosystem services in diverse multi-crop farming systems. Working in 10 smallholder villages in rural Nepal, we use this metric to test whether key pollinators, and the management actions that support them, differ based on a farmers’ agricultural priority (producing nutritious food to feed the family versus generating income from cash crops). We also test whether the resilience of pollination services changes as farmers specialise on cash crops. We show that a farmers’ agricultural priority can determine the community of pollinators they rely upon. Wild insects including bumblebees, solitary bees, and flies provided the majority of the pollination service underpinning nutrient production, whilst income generation was much more dependent on a single species - the domesticated honeybee Apis cerana. The significantly lower diversity of pollinators supporting income generation leaves cash crop farmers more vulnerable to pollinator declines. Regardless of a farmers’ agricultural priority, the same collection of wild plant species (mostly herbaceous weeds and shrubs) were important for supporting crop pollinators with floral resources. Promoting these wild plants is likely to enhance pollination services for all farmers in the region.
Synthesis and applications: We highlight the increased vulnerability of pollination services when smallholders transition to specialised cash crop farming and emphasise the role of crop, pollinator, and wild plant diversity in mitigating this risk. The method we present could be readily applied to other smallholder settings across the world to help characterise and manage the ecosystem services underpinning the livelihoods and nutritional health of smallholder families.
https://doi.org/10.5061/dryad.0rxwdbs91
This submission includes two datasets: 1) a database of 10,975 plant-pollinator interactions; 2) data on the pollen loads of a range of different insect taxa recorded in the plant-pollinator surveys.
Data were collected in 10 smallholder farming villages (2400-3000 metres ASL; temperate climate) in Patarasi Rural Municipality of Jumla District, Nepal. Each study village comprised a cluster of 100-400 closely spaced households interspersed with small vegetable gardens and livestock enclosures. Village surroundings include many small (0.01–0.3 ha) arable fields and apple orchards as well as large areas of steep, heavily-grazed grassland pasture and native coniferous forest (see paper for more details).
Plant-pollinator interaction data were collected by conducting plant-pollinator visitation surveys every two weeks from 18 April to 4 November 2021 (spring to autumn) in a 600 x 600 metre sampling area centred on the midpoint of each study village. This area was divided into three habitat categories: village, crop, and semi-natural vegetation. In each of these habitats, we randomly located three replicate fixed survey plots of 60 x 60 metres (9 plots per village). Every two weeks, a 40-minute survey was conducted in each plot to record the interactions between plants (both crop and non-crop species) and flower-visiting insects. Insects were captured, pinned and identified to species or morphospecies.
Pollen load data were collected from a total of 1928 insect specimens, representing 18% of the total insect specimen collection. Insects were selected by stratified random sampling, to prioritise the sampling of taxa which were recorded visiting crops plants (rather than those exclusively visiting wild plants). The number of pollen grains on each individual insect specimen was quantified by swabbing the insect with glycerine jelly and counting the total number of pollen grains on each insect using light microscopy. We calculated the mean pollen carrying capacity of each insect taxon by taking a mean of the total number of pollen grains recorded across all replicates of the taxon.
Description of the data and file structure
In the plant-pollinator interaction dataset, each row represents an individual interaction, with details of the interaction including plant species, pollinator species, location, habitat etc. shown in each column. The tab labelled ‘Info’ within the Excel workbook provides further details on the different columns within the main datasheet.
In the pollen load dataset, the raw data (number of pollen grains per insect specimen) is shown in a worksheet called ‘all pollen data’. The data are then summarised at different taxonomic levels (mean number of pollen grains per species, genus, family and order) in subsequent worksheets. The ‘Info’ tab provides further details of the variables within each worksheet.
Code/Software
Analysis code and additional source data available from Zenodo via the following link: https://doi.org/10.5281/zenodo.12609151