Skip to content
Ecological Intelligence

Ecological evidence.
Independent perspective.
Explore Wild Lands ↗

Ecological Intelligence / Subject

pollinators

Peer-reviewed Latin America & Caribbean

A highly forested tree city still undersupplied specialist flower visitors

Evidence
Peer-reviewed
Geography
Campo Grande, Brazil
System
Urban street trees / floral-resource connectivity

WHAT THE EVIDENCE SAYS

A survey of 1,934 trees across 200 blocks found 72% of individual trees were non-native. Flowers and modeled resource connectivity favored small-insect and bee guilds, while some specialist visitors had fewer or more fragmented resources.

WHY IT MATTERS / OUR READ

Urban canopy coverage should be evaluated against phenology and the specific animals it supports, not simply total shade or tree counts.

CAVEATS / UNCERTAINTY

Floral morphology and spatial proxies do not directly establish pollen transfer or trends in pollinator population viability.

SOURCE

Urban Ecosystems — Silva de Carvalho et al. (2026)
View original evidence ↗

Campo Grande has year-round street-tree flowering, but its resource network favors generalized insect visitors.

Read the Wild Lands assessment →
Peer-reviewed Europe

Tiny field-edge strips changed where pollinator models placed habitat

Evidence
Peer-reviewed
Geography
Three agricultural landscapes, Netherlands
System
Agricultural landscape / pollinator habitat

WHAT THE EVIDENCE SAYS

Fine-resolution (10-meter) species-distribution models incorporated herbaceous linear landscape elements as cover or distance covariates. Compared with models lacking these features, inclusion improved predictions and reduced overprediction in fields and underprediction at edges; improvements were evaluated against independently collected field records as well as withheld training data.

WHY IT MATTERS / OUR READ

Restoration opportunity maps need to capture narrow field margins and ditches, not just broad land-cover classes; independent field validation matters when models guide scarce conservation funds.

CAVEATS / UNCERTAINTY

Model improvement is not proof that newly adding a strip will cause equivalent abundance gains. Transferability from three Dutch landscapes to other crops, climates or taxa remains to be tested.

SOURCE

Landscape Ecology — Moens et al. (2026)
View original evidence ↗

Dutch bee and hoverfly models improved when linear habitat elements were mapped instead of being averaged into cropland.

Read the Wild Lands assessment →
Peer-reviewed North America

For wetland pollinators, the hydrologic mosaic mattered more than the management label

Evidence
Peer-reviewed
Geography
Montezuma Wetlands Complex and Seneca Meadows Wetlands Preserve, central New York, USA
System
Restored emergent wetlands

WHAT THE EVIDENCE SAYS

Researchers sampled 38 unique restored emergent wetlands during June–September 2019 and 2020, collecting 9,046 bees representing at least 109 species. Sites included passive, partial-drawdown and full-drawdown management. Water depth, open-water extent, mudflat frequency and floral-resource measures explained plant and bee patterns better than treatment labels alone; management treatment by itself explained about 2% of bee-community variation, while sampling month explained 17%. Actively managed wetlands produced a strong late-season bloom pulse, while passive wetlands contributed other season-long floral resources.

WHY IT MATTERS / OUR READ

For managers, the useful unit is the wetland complex rather than one ‘best’ drawdown prescription. Maintaining nearby impoundments with different hydrologic states can create complementary bloom windows and habitat conditions. Monitor water depth, mudflat, floral frequency and invasive vegetation—not just whether a basin was labeled passive or drawn down.

CAVEATS / UNCERTAINTY

The design was observational and management treatments were uneven and sometimes rotated between years. Most collected bees were generalists, surrounding upland nesting habitat and landscape context were not fully measured, and treatment label alone explained little variation. Results support hydrologic heterogeneity and adaptive monitoring, not a universal drawdown schedule.

SOURCE

Wetlands Ecology and Management — primary study
View original evidence ↗

Across 38 restored central New York wetlands, water depth, open water, mudflat and floral resources explained bee and plant patterns better than simple drawdown categories; at least 109 bee species used the complex.

Read the Wild Lands assessment →
Peer-reviewed

Two native Hawaiian honeycreepers can compete for the same imperiled flower

Evidence
Peer-reviewed
Geography
Oʻahu, Hawaiʻi, USA
System
Island forest pollination network

WHAT THE EVIDENCE SAYS

Field cameras and playback experiments at an endangered native aster found asymmetric interactions between two native honeycreepers: the more dominant ʻapapane altered the timing and frequency of ʻamakihi flower visits. The study shows that competitive behavior among native pollinators can shape access to a scarce floral resource even in a system dominated by broader invasive-species pressures.

WHY IT MATTERS / OUR READ

Restoration planning in simplified island systems may need to retain interaction structure, not only species lists. Native species can compete as well as facilitate, and rare plant recovery may depend on how those interactions change around limited resources.

CAVEATS / UNCERTAINTY

The work centers on one plant and two bird species in a localized setting. Flower visitation is not the same as realized pollen transfer, seed set or plant population growth, so the demographic consequence remains uncertain.

SOURCE

Behavioral Ecology and Sociobiology — primary study
View original evidence ↗

Playback and camera monitoring on Oʻahu showed ʻapapane presence shifted ʻamakihi visitation timing and frequency at an endangered aster, revealing native–native interaction risk inside an already invaded system.

Read the Wild Lands assessment →
Peer-reviewed Global

Pollination networks do not become steadily more specialized toward the Equator

Evidence
Peer-reviewed
Geography
Global
System
Plant–pollinator networks

WHAT THE EVIDENCE SAYS

A global synthesis of more than 3,400 quantitative plant–pollinator networks and more than 110,000 interactions found no simple monotonic increase in specialization toward the Equator. Network- and pollinator-level specialization peaked around low northern latitudes, plant specialization showed a different asymmetric pattern, and temperature and precipitation often explained variation better than latitude, productivity or broad biodiversity gradients.

WHY IT MATTERS / OUR READ

The result weakens a familiar shorthand: latitude is not a sufficient proxy for interaction structure. For conservation and restoration, interaction networks may respond to climatic context and taxonomic composition in ways that species richness alone will not predict.

CAVEATS / UNCERTAINTY

The database combines networks collected with different methods, sampling effort and completeness. The analysis is observational and cannot by itself establish climate as the causal driver of specialization, and patterns differ among plants and pollinator groups.

SOURCE

Nature Ecology & Evolution — primary study
View original evidence ↗

A synthesis of more than 3,400 quantitative networks finds climate and taxon-specific responses explain global specialization better than a simple latitudinal gradient.

Read the Wild Lands assessment →
Peer-reviewed North America

Urban connectivity may work differently once organisms are allowed to evolve

Evidence
Peer-reviewed
Geography
100 metropolitan areas, United States
System
Urban landscapes and plant–pollinator metacommunities

WHAT THE EVIDENCE SAYS

Across 100 U.S. metropolitan regions, urban landscapes had lower heterogeneity and forest aggregation and higher aggregation of developed land than non-urban landscapes. A proof-of-concept spatial eco-evolutionary plant–pollinator model then showed that heterogeneity and connectivity can delay extinction, but trait evolution and mutualistic feedbacks change the size of those benefits.

WHY IT MATTERS / OUR READ

Green corridors and habitat networks are not static plumbing. If species interactions and traits change in fragmented urban landscapes, design performance may depend on how connectivity shapes both movement and adaptation over time.

CAVEATS / UNCERTAINTY

The empirical landscape analysis does not measure observed extinction outcomes, and the eco-evolutionary result comes from a proof-of-concept model rather than a post-occupancy design experiment. The 100-city U.S. sample may not generalize to other urban forms.

SOURCE

Landscape Ecology
View original evidence ↗

A 100-city landscape analysis paired with a plant–pollinator model suggests that heterogeneity and connectivity can delay extinction, while evolutionary feedbacks can erode some of that benefit.

Read the Wild Lands assessment →
Peer-reviewed Europe

Heat during pupation kept killing bumblebees after emergence

Evidence
Peer-reviewed
Geography
Controlled laboratory experiment, Germany
System
Pollinator developmental ecology

WHAT THE EVIDENCE SAYS

In vitro Bombus terrestris pupae exposed to cyclic 37°C or 38°C heat had lower emergence than the 34°C treatment, and surviving adults showed elevated post-emergence mortality and more deformities at the highest treatment. Effects therefore persisted beyond the developmental exposure window.

WHY IT MATTERS / OUR READ

Heat risk can carry across life stages. Monitoring only immediate survival during an extreme event can miss delayed demographic costs in pollinators and other insects.

CAVEATS / UNCERTAINTY

This is a controlled experiment on one model bumblebee species. The imposed thermal regimes simplify field conditions, and individual survival effects do not directly quantify colony- or population-level consequences.

SOURCE

Communications Biology
View original evidence ↗

EvidencePeer-reviewedGeographyControlled laboratory experiment, GermanySystemPollinator developmental ecology WHAT THE EVIDENCE SAYS In vitro Bombus terrestris pupae exposed to cyclic 37°C or 38°C heat had lower emergence than the

Read the Wild Lands assessment →
Peer-reviewed North America

For monarch habitat on rangelands, the grazing regime changed milkweed more consistently than it changed every floral resource

Evidence
Peer-reviewed
Geography
Central North Dakota, USA
System
Northern Great Plains rangeland

WHAT THE EVIDENCE SAYS

Comparing modified twice-over rest-rotation, patch-burn and season-long grazing in 2021–2022, researchers found higher milkweed abundance under modified rest-rotation. Floral resources showed a strong year-by-regime interaction, including a drought-driven collapse and rebound in patch-burn pastures. Adult and juvenile monarch abundance correlated positively with milkweed.

WHY IT MATTERS / OUR READ

The result makes disturbance timing and drought context part of pollinator-habitat management rather than treating grazing systems as fixed labels.

CAVEATS / UNCERTAINTY

Two years at one research landscape cannot establish a universal grazing prescription or demographic benefit to monarch populations. Milkweed abundance and observed monarch counts are not equivalent to population growth.

SOURCE

Frontiers in Ecology and Evolution
View original evidence ↗

A two-year North Dakota comparison found more milkweed under modified rest-rotation grazing, while drought created a strong year-by-regime response in flowers.

Read the Wild Lands assessment →
Peer-reviewed North America

Prairie seed mixes performed best across several services at intermediate grass-to-forb ratios

Evidence
Peer-reviewed
Geography
Iowa, USA
System
Tallgrass prairie restoration plots

WHAT THE EVIDENCE SAYS

An Iowa experiment established prairie mixes across 11 grass-to-forb seed ratios and followed them for four years. Plant richness, flower production, volunteer biomass and pollinator and butterfly visitation showed nonlinear responses, with several outcomes strongest at intermediate grass-to-forb ratios. Net primary productivity did not show a significant relationship with the seed ratio.

WHY IT MATTERS / OUR READ

Seed-mix composition is a design variable, not just a species list. An intermediate functional balance may support multiple restoration objectives at once rather than forcing managers to choose between floral resources, resistance to unwanted colonists and standing productivity.

CAVEATS / UNCERTAINTY

The results come from one experimental restoration setting over four years. Optimal ratios will depend on species identities, soils, climate and management, and a null productivity response does not guarantee equal forage or habitat function.

SOURCE

Ecological Applications · Original evidence
View original evidence ↗

Four years of Iowa restoration plots found a hump-shaped response in richness, flowering, pollinator visits and invasion resistance—without a detectable productivity penalty.

Read the Wild Lands assessment →
Peer-reviewed Europe

Pollinator habitat can look adequate in spring and still run short of food in June

Evidence
Peer-reviewed
Geography
Intensive arable landscapes, France
System
Agricultural landscapes and pollinator habitat

WHAT THE EVIDENCE SAYS

Across three growing seasons and 10 intensively managed agricultural landscapes, researchers recorded 5,612 flower–pollinator interactions across eight landscape-element types. Landscape-scale nectar and pollen peaked in April and declined through August, with the largest apparent supply-demand gap in June.

WHY IT MATTERS / OUR READ

Habitat amount is not the same as resource continuity. Pollinator planning may need to ask what is flowering when, and whether different field margins, fallows, verges, hedgerows or crop resources complement one another through the season.

CAVEATS / UNCERTAINTY

The study estimates floral-resource production and observed pollinator use rather than directly measuring population growth or survival. Its proposed June management focus should be calibrated to local crop systems and phenology outside the study landscapes.

SOURCE

Agriculture, Ecosystems & Environment · Original evidence
View original evidence ↗

Three years across 10 French arable landscapes show nectar and pollen supply peaks early, then declines as pollinator demand continues.

Read the Wild Lands assessment →
Peer-reviewed North America

For forest pollinators, a mosaic of fire histories may matter more than one “optimal” burn schedule

Evidence
Peer-reviewed
Geography
Fire-managed forests, southeastern United States
System
Conifer forest and prescribed-fire landscape

WHAT THE EVIDENCE SAYS

In a fire-managed forest landscape, the number of distinct fire histories around sampling locations was positively associated with bee and hover-fly richness and with butterfly and hover-fly abundance, although informative spatial scales differed among taxa. Fire frequency, roads, canopy openness and tree composition also showed taxon-specific relationships.

WHY IT MATTERS / OUR READ

For managers, the useful signal is not “burn more” or a universal return interval. Different pollinator groups appear to respond to different aspects and spatial scales of disturbance, so retaining a landscape mosaic of burn histories and forest conditions may support a broader community than optimizing for a single average treatment.

CAVEATS / UNCERTAINTY

This is observational evidence from one southeastern U.S. system, not a randomized burn-prescription experiment. Relationships differed by taxon and spatial scale, and correlated roads and forest structure complicate simple causal interpretation.

SOURCE

Global Change Biology · Original evidence
View original evidence ↗

A southeastern U.S. study found bees, butterflies and hover flies responded differently to fire history, roads and forest structure—arguing for heterogeneity rather than one-size-fits-all disturbance.

Read the Wild Lands assessment →
Peer-reviewed Europe

Europe — Bumblebee habitat loss becomes clearer when climate is isolated from other pressures

Evidence
Peer-reviewed
Geography
Europe
System
Bumblebee and pollinator habitat

WHAT THE EVIDENCE SAYS

Researchers compared factual and counterfactual climate scenarios from 1901–2019 for 47 European bumblebee species. Climate change reduced community-level ecological suitability by about 5% on average and by as much as 19% locally, with pronounced losses in southern and lowland central Europe. Higher-altitude gains in Alpine and Boreal regions partially offset losses elsewhere.

WHY IT MATTERS / OUR READ

Pollinator decline is often discussed as one blended outcome of land use, pesticides and climate. Separating one pressure clarifies where climate is already reshaping suitability and also shows why apparent gains in some regions should not be used to erase losses in others.

CAVEATS / UNCERTAINTY

The attribution depends on ecological niche models, historical occurrence records and reconstructed climate scenarios. Habitat suitability is not the same as observed abundance, reproductive success or population viability.

SOURCE

Nature Climate Change · Original evidence
View original evidence ↗

Counterfactual modelling attributes an average 5% reduction in community-level ecological suitability to climate change across Europe, with local losses up to 19%.

Read the Wild Lands assessment →

Keep exploring

Follow the systems. Find the connections.

WildlifeLandWaterOceansClimateCitiesOne HealthGovernanceMethods / Monitoring
Browse the archive →