Skip to main content
Dryad logo

Efficacy of cover crops for pollinator habitat provision and weed suppression

Citation

Gage, Karla et al. (2021), Efficacy of cover crops for pollinator habitat provision and weed suppression, Dryad, Dataset, https://doi.org/10.5061/dryad.80gb5mkp3

Abstract

Pollinator declines have been documented globally, but little information is available about native bee ecology in Midwestern US agriculture. This project seeks to optimize pollinator support and weed suppression in a 3-year crop rotation with a fallow growing season. During fallow, one of five cover crop treatments (T1: crimson, red, and ladino clover and Bob oats [Trifolium incarnatum, T. pratense, T. repens, Avena sativa]; T2: crimson clover and oats; T3: red clover and oats; T4: ladino clover and oats; T5: no cover crop; T6/control: winter wheat [Triticum aestivum]) was seeded in one-half of 25 agricultural fields, while wheat was left unharvested in the other half as a comparison. Treatments that provide season-long floral resources support the greatest bee diversity and abundance (T1), and treatments with red clover support declining Bombus species (T1 and T3). Late-season floral resources may be important, yet limited (T1 and T4), and some species of agricultural weeds provide floral resources. Floral diversity may be less important than flower abundance or timing for pollinator diversity (T1 – T4). Weed diversity was greatest in the no cover crop treatment (T5), least in winter wheat (T6), and intermediate in cover crop treatments (T1 – T4) with no differences in weeds of economic concern. Wheat suppresses weeds but does not provide floral resources for pollinators. These results may also be applicable to marginal lands taken out of cultivation or field margin pollinator plantings in a typical corn-soybean rotation. Floral resource availability across the landscape is critical to maintain pollinator diversity.

Methods

Data collection and processing is outlined in detail in Bryan et al., Efficacy of cover crops for pollinator habitat provision and weed suppression, in the journal Environmental Entomology. Methodology is also included in the usage notes section below. 

Usage Notes

COMMUNITY DATA: The first two files listed below contain community data for bee species and weedy plant species, respectively. The first column is ‘Ordcode’ and corresponds to the given environmental data files (Bee_ENV.csv, Weeds 2017 ENV.csv) and site and treatment information, as well as survey timings for species occurrence. The other columns contain species codes. In the bee dataset, species names are composed of the first three letters of the genus name and the first three letters of the species name. In the weeds community dataset, the species names are 5-letter codes, which typically represent the first three letters of the genus name and the first two letters of the species name, following the European and Mediterranean Plant Protection Organization (EPPO) coding system (https://www.eppo.int). For weed species sampling, a chevron-patterned transect was designed in ArcGIS prior to conducting surveys. Transects consisted of 20 points, spaced 10-20 m apart. Researchers navigated to each point and sampled community data using a 0.5 x 1 m quadrat. Species and percent cover were measured at each plot.

FILE: Bee_community_1s_removed.txt

(Bray-Curtis dissimilarity was used to ordinate samples, since it is one of the most robust ordination methods. Survey rounds where one sample had zero taxa were removed for multivariate analysis. The dataset used in analysis is provided here with some omitted samples.)

FILE: Weeds 2017.txt

ENVIRONMENTAL DATA: The environmental datafiles provided for multivariate analysis, which correspond to the community data files, are: Bee_ENV.csv, Bee ENV_remove0_1s_removed.csv, and Weeds 2017 ENV.csv.

FILE: Bee_ENV.csv

Ordcode – This is the code provided to relate the community and environmental datasets for multivariate analysis.

Field – Fields were located at Crab Orchard National Wildlife Refuge in Marion, Illinois, USA. A map is available in Bryan et al. (accepted) Environmental Entomology. Field sizes ranged from 3.82 to 9.48 ha and averaged 5.36 ha. Fields were divided in half; one half was planted in the control treatment, Triticum aestivum L. (winter wheat), and the other half was planted in one of five clover cover treatments. Treatments were replicated 5 times and completely randomized across the Refuge land. Treatments were planted immediately prior to soybean harvest in September 2016.

Method – This is the method of bee specimen capture: ‘Net’ or pan ‘Trap’.

Treatment – This is abbreviated ‘T’. Numbers in parentheses represent the seeding rates of each cover species in kg/ha. Fields were assigned one of 6 treatments: T1 - Clover 3-Way and Bob Oats [Trifolium incarnatum (7.8), T. repens (4.5), T. pratense (1.7), and Avena sativa (16.8)], T2 - Crimson Clover and Bob Oats [T. incarnatum (16.8) and A. sativa (22.0)], T3 - Red Clover and Bob Oats [T. pratense (11.0) and A. sativa (22.0)], T4 - Ladino Clover and Bob Oats [T. repens (4.5) and A. sativa (22.0)], T5 - No Cover Crop [None], T6 - Control: Winter Wheat [Triticum aestivum (112.0)].

NestedT – This variable groups the field halves, T1-5 or T6, for analysis. This allows comparison of the paired treatment with the control.

Round – Six surveys or ‘rounds’ were performed. Bee sampling occurred approximately every three weeks from mid-April to late-August 2017.

CorT – This denotes control ‘C’ or treatment ‘T’.

Field.CT – This is a combination of ‘Field’ code and ‘C’ or ‘T’ notation.

S – Richness, or number of species, per sample.

N – Number of individuals per sample.

d – Diversity. ‘****’ here and throughout notes that species and individuals were too few to calculate.

J' – Pielou’s evenness

H'(loge) – Shannon-Weiner diversity index

eH – the effective number of species

1-Lambda' – Simpson index

FA.S – Floral abundance richness

FA.N – Floral abundance number of individuals

FA.J – Floral abundance evenness

FA.H – Floral abundance diversity

FA.eH – Floral abundance effective number of species

Apimel – Abundance of honey bees

 

FILE: Bee ENV_remove0_1s_removed.csv

Variables are coded in the same way as the file above. The difference is that this file corresponds to the bee community file provided, with the same survey rounds removed.

FILE: Weeds 2017 ENV.csv

Ordcode - This is the code provided to relate the community and environmental datasets for multivariate analysis.

Field – Fields were located at Crab Orchard National Wildlife Refuge in Marion, Illinois, USA. A map is available in Bryan et al. (accepted) Environmental Entomology. Field sizes ranged from 3.82 to 9.48 ha and averaged 5.36 ha. Fields were divided in half; one half was planted in the control treatment, Triticum aestivum L. (winter wheat), and the other half was planted in one of five clover cover treatments. Treatments were replicated 5 times and completely randomized across the Refuge land. Treatments were planted immediately prior to soybean harvest in September 2016.

Treatment – This is abbreviated ‘T’. Numbers in parentheses represent the seeding rates of each cover species in kg/ha. Fields were assigned one of 6 treatments: T1 - Clover 3-Way and Bob Oats [Trifolium incarnatum (7.8), T. repens (4.5), T. pratense (1.7), and Avena sativa (16.8)], T2 - Crimson Clover and Bob Oats [T. incarnatum (16.8) and A. sativa (22.0)], T3 - Red Clover and Bob Oats [T. pratense (11.0) and A. sativa (22.0)], T4 - Ladino Clover and Bob Oats [T. repens (4.5) and A. sativa (22.0)], T5 - No Cover Crop [None], T6 - Control: Winter Wheat [Triticum aestivum (112.0)].

CorT – This denotes control ‘C’ or treatment ‘T’.

Field.CT – This is a combination of ‘Field’ code and ‘C’ or ‘T’ notation.

Round – There were 3 rounds or survey periods (early April, mid-June, and late July in 2016 and 2017.

Survey Year  - 2016 or 2017. Surveys in 2016 were taken while files were in a soybean rotation and were intended as a baseline comparison for the weed community data. Surveys in 2017 occurred while the fields were planted into the clover cover treatments.

S – Richness, or number of species, per sample.

J' – Pielou’s evenness

H'(loge) – Shannon-Weiner diversity index

eH – the effective number of species

faS – Floral abundance richness

faJ' – Floral abundance evenness

faH'(loge) – Floral abundance diversity

faeH – Floral abundance effective number of species

faSumCover – Floral abundance sum of cover

AVESA – Cover of Avena sativa L.

TRFIN – Cover of Trifolium incarnatum L.

TRFPR – Cover of Trifolium pratense L.

TRFRE – Cover of Trifolium repens L.

TRZAW – Cover of Triticum aestivum L.

 

FILE: Floral Abund.csv

Sample – This corresponds to ‘Ordcode’ in other files.

S – Richness, or number of species, per sample.

N – Number of individuals per sample.

J' – Pielou’s evenness

H'(loge) – Shannon-Weiner diversity index

e^H' – the effective number of species

Field – Fields were located at Crab Orchard National Wildlife Refuge in Marion, Illinois, USA. A map is available in Bryan et al. (accepted) Environmental Entomology. Field sizes ranged from 3.82 to 9.48 ha and averaged 5.36 ha. Fields were divided in half; one half was planted in the control treatment, Triticum aestivum L. (winter wheat), and the other half was planted in one of five clover cover treatments. Treatments were replicated 5 times and completely randomized across the Refuge land. Treatments were planted immediately prior to soybean harvest in September 2016.

Treatment – This is abbreviated ‘T’. Numbers in parentheses represent the seeding rates of each cover species in kg/ha. Fields were assigned one of 6 treatments: T1 - Clover 3-Way and Bob Oats [Trifolium incarnatum (7.8), T. repens (4.5), T. pratense (1.7), and Avena sativa (16.8)], T2 - Crimson Clover and Bob Oats [T. incarnatum (16.8) and A. sativa (22.0)], T3 - Red Clover and Bob Oats [T. pratense (11.0) and A. sativa (22.0)], T4 - Ladino Clover and Bob Oats [T. repens (4.5) and A. sativa (22.0)], T5 - No Cover Crop [None], T6 - Control: Winter Wheat [Triticum aestivum (112.0)].

NestedT – This variable groups the field halves, T1-5 or T6, for analysis. This allows comparison of the paired treatment with the control.

Round – Six surveys or ‘rounds’ were performed. Floral abundance was measured at the same time as bee sampling, approximately every three weeks from mid-April to late-August 2017.

CorT – This denotes control ‘C’ or treatment ‘T’.

Field.CT – This is a combination of ‘Field’ code and ‘C’ or ‘T’ notation.

 

ANALYSIS FILES:

Bees PERMANOVA – This file contains the R code for the bee multivariate analysis.

Floral Abund – This file contains the R code to analyze floral abundance.

Weeds PERMANOVA – This file contains the R code for the weeds multivariate analysis.

Funding

Service First Authority (43 U.S.C. 1703, revised by Public Law 113-76), CFDA Program 15.650, Award: Cooperative Agreement Award F16AC01016