Environmental Valuation, Human Impact, and Ecological Legacy

Knowledge Graph (1051 nodes, 6781 connections)

Research Primer

Background

Flowering plants in the subalpine and alpine meadows around the Rocky Mountain Biological Laboratory (RMBL) depend on insects, birds, and other animals to move pollen between flowers. In return, the animals receive food in the form of nectar (a sugary liquid) and pollen (a protein source). This exchange, called a plant-pollinator mutualism, is the foundation of wildflower reproduction in the Gunnison Basin and underpins the spectacular summer wildflower displays for which the region is famous. When pollinators decline or when the timing of flowering shifts, both wild plants and the animals that depend on them are affected.

Understanding these relationships requires several key ideas. Phenology refers to the seasonal timing of biological events, such as when snow melts, when flowers bloom, and when bees emerge from their nests. When plants and their pollinators no longer overlap in time, a phenological mismatch occurs, which can reduce reproduction for both partners. Pollinator foraging behavior describes the choices animals make about which flowers to visit, how long to stay, and how often to return. These choices are shaped by floral traits such as nectar volume and sugar concentration, floral morphology (whether a flower is open and simple or deep and complex), corolla length, and floral abundance. Researchers often summarize the entirety of these interactions in a community as a plant-pollinator network, a map of who visits whom that can be analyzed for properties such as connectance (how many of the possible interactions actually occur) and nestedness (whether specialists tend to visit subsets of the flowers that generalists also visit).

Finally, several pressures threaten these mutualisms in mountain landscapes. Climate change is shifting snowmelt earlier and altering growing seasons. Invasive plants such as yellow toadflax (Linaria vulgaris) and common dandelion (Taraxacum officinale) compete with native wildflowers for pollinator attention. Roads, dust, and habitat change add further stresses. Pollinator decline, the global reduction in bee and other pollinator populations, makes understanding these systems urgent for conservation and land management.

Foundational work

Early studies at RMBL helped establish how individual pollinator traits shape feeding behavior and, in turn, plant reproduction. Boggs (Boggs, 1988) showed that nectar feeding rates in Speyeria mormonia butterflies depend on sex, body size, age, and sugar concentration, with peak ingestion at sucrose concentrations between 30 and 40 percent. This work linked optimal foraging theory to the realities of nectar chemistry in the field and laid groundwork for understanding why pollinators prefer some flowers over others. Around the same period, regional ecological surveys (e.g., McAda et al., 1980) documented the distributions of organisms across the Upper Colorado River Basin that anchor later community-level studies.

Follow-up work in the 2000s expanded the scope to include floral signaling and the consequences of invasive species. Studies of nectar robbing in Linaria vulgaris (Newman, 2004) examined how cheaters affect mutualisms, and early bee-foraging work showed that Bombus appositus dramatically shifted its pollen loads from 24.9 percent invasive Linaria pollen in 2004 to 98.3 percent by 2006, demonstrating how quickly native bees can switch to invasive resources (Author, 2006). Together, these foundational studies established that pollinator behavior, floral rewards, and species composition jointly determine pollination outcomes.

Key findings

A central theme across RMBL studies is that pollinators are far from random visitors; they discriminate among flowers based on traits and rewards. Solitary bees in the East River Valley carried pollen loads that were significantly less diverse than the surrounding floral community, indicating non-neutral interactions and revealing preferences for flowers with wider corollas, shorter heights, and larger displays (Baker, 2021). Bumble bees showed similarly clear preferences: Bombus flavifrons visited only simple flowers, avoiding compound umbels whose nectar, though more concentrated, was harder to access (Steinmann, 2021). Bee body size also matters, with larger bees visiting more plant species and showing positive correlations between size and the height of anthers they could reach (Araujo & Forrest, 2009). Differences among pollinators translate into different pollination outcomes: Bombus nevadensis was the most effective pollinator of Delphinium barbeyi, depositing the most conspecific pollen and producing the highest seed set (Author, 2013), and Psithyrus insularis deposited less heterospecific pollen than Megachile on fireweed, making it a cleaner pollinator (Petroff, 2012).

At the community level, plant-pollinator networks at RMBL change dramatically across the season. Weekly networks built in Gothic from May to August 2013 showed that connectance was highest at the beginning and end of the season, tracking floral richness, and that 1448 interactions among 30 plants and 62 pollinators rearranged substantially from week to week (Cunningham, 2013). Floral display is a strong driver of visitation: larger displays significantly increased pollinator visits across native species, although nitrogen addition did not increase flower numbers or visitation in a Gothic restoration plot (Fresco, 2019). Specialized fly behavior also responds to community context. As flower diversity increased, fly foraging fidelity decreased (Author, 2012), and removing the most abundant Bombus species caused Diptera to increase their visits to generalist flowers (Author, 2011).

Nectar itself is a complex environment. Floral nectar in Delphinium barbeyi hosts yeast communities whose abundance increases through the season and is significantly higher in flowers exposed to pollinators, identifying pollinators as the main vectors of these microbes (Phillips, 2013). Pollen mass varies significantly across 35 plant species and declines from early to late season (Flores, 2021), meaning that the resource landscape pollinators face shifts dramatically over a few weeks. Invasive plants further reshape this landscape: most invasive plants offered more rewards than their native counterparts (Author, 2010), and dandelions outcompeted Ranunculus glaberrimus for pollinator visits while losing to Valeriana occidentalis (Salazar, 2023).

Current frontier

Early work at RMBL through the 1990s and 2000s focused on individual species, foraging rules, and reward chemistry. Studies from 2010 to 2019 increasingly assembled these pieces into community-scale networks. Since 2020, the field has shifted toward understanding how networks change in time and how they will respond to climate change. Recent network studies have examined within-day variation, finding that different pollinator species are active at different times, with subtle shifts in nestedness, connectance, and specialization through the day (Gascon, 2022), and crepuscular networks dominated by moths, which differ markedly between Colorado and California sites (Syskine, 2020). Arrowsmith (Arrowsmith, 2023) used structural equation modeling to show that temperature variation and plant community dissimilarity directly drive network rewiring, while geographic distance does not, and that the consequences of pollinator loss for seed production depend on which species and traits remain.

The most recent work expands the cast of pollinators and the chemical complexity of nectar. Syrphid flies with denser hair carry significantly more pollen, and every specimen examined carried enough pollen for fruit production, suggesting these flies are more important than previously recognized (Devora, 2025). Quantification of ethanol in commercial bumble bee honey and in nectar of native wildflowers including Mertensia ciliata, Ipomopsis aggregata, and Delphinium barbeyi has opened a new chemical-ecology frontier (Bustamante, 2025). Climate-driven phenological work continues to mature: experimental early-snowmelt plots reduced conspecific pollen deposition in Geum triflorum but not in Claytonia lanceolata or Delphinium nuttallianum, indicating that responses are species-specific (Sosa Antunez, 2024). Mason bee transplant experiments along elevational gradients are beginning to disentangle how season length and diet breadth limit reproductive success at high elevations (Tripathy, 2023).

Open questions

Major uncertainties remain about how RMBL plant-pollinator networks will reorganize as snowmelt continues to advance and summer temperatures rise. Which species pairs are most vulnerable to phenological mismatch, and which can rewire their interactions to compensate? How do non-bee pollinators such as syrphid flies and moths contribute to pollination across the full daily and seasonal cycle, and how should they be incorporated into conservation planning? The roles of nectar microbes, ethanol, and other chemical features in shaping pollinator choice and plant fitness are only beginning to be quantified at natural concentrations. Finally, the cumulative impacts of invasive plants, road dust, and management decisions on long-term wildflower reproduction in the Gunnison Basin remain poorly resolved, especially over the multi-decadal timescales relevant to mountain ecosystems.

References

Araujo & Forrest (2009). Is there a positive correlation between bee size and size parameters of the flowers the bees visit for pollen to provision their nests? →

Arrowsmith (2023). Biotic and abiotic drivers of plant-pollinator interaction rewiring. →

Author (2006). Are native bees picky enough? The effect of an invasive plant, Linaria vulgaris, on the foraging of native bee pollinators. →

Author (2010). Nectar and pollen rewards of invasive plants and their native congeners. →

Author (2011). Diptera behavioral response to targeted Bombus removal. →

Author (2012). Effects of floral diversity and density on fly floral foraging fidelity in subalpine meadows. →

Author (2013). Comparing Bombus pollinator efficacies in Delphinium barbeyi reproductive success. →

Baker (2021). Comparing pollen distribution on pollinators to floral community composition in the East River Valley. →

Boggs (1988). Rates of nectar feeding in butterflies: effects of sex, size, age, and sugar concentration. Functional Ecology. →

Bustamante (2025). Quantifying ethanol in bumble bee honey and Colorado rocky mountain wildflowers. →

Cunningham (2013). Exploring within season temporal variation in plant-pollinator interactions in sub-alpine meadows using weekly pollination interaction networks. →

Devora (2025). Functional morphology of pollen capture in Syrphid flies: The effects of pile density and body region. →

Flores (2021). Variation of Pollen Mass Across Species, Habitat Types, and Time. →

Fresco (2019). The effect of nutrient availability on floral display and pollinator interactions. →

Gascon (2022). Within Day Temporal Variation Within Plant-Pollinator Networks. →

McAda, Berry, Phillips (1980). Distribution of Fishes in the San Rafael River System of the Upper Colorado River Basin. The Southwestern Naturalist. →

Newman (2004). The ecology of pollination and nectar robbing in Linaria vulgaris in the Colorado Rocky Mountains. TSpace. →

Petroff (2012). Single-visit pollination efficacy of Psithyrus insularism and Megachile on Chamerion augustifolium. →

Phillips (2013). Nectar dwelling yeasts and their effects on pollinator preference. →

Salazar (2023). Do dandelions compete with native plants for pollinator visits? →

Sosa Antunez (2024). How does early snowmelt affect pollen deposition on spring wildflowers? →

Steinmann (2021). Nectar Variation in Simple and Compound Flowers; and its Effects on Plant Pollinator Interactions. →

Syskine (2020). Flying by night: Comparing crepuscular pollinator networks across two sites in Western North America. →

Tripathy (2023). How season length and diet breadth limit mason bee success across an elevational gradient. →

Species (4) →

Concept (21) →

Place (15) →

Stakeholder (13)

Author (1) →

Document (5) →

Dataset (2) →