Spatio-temporal dynamics of insect communities in constructed and natural tidal marshes with distinct landscape positions
Data files
Apr 26, 2024 version files 118.36 KB
-
Biodiversity.csv
-
FloralCounts.csv
-
Herbivory.csv
-
README.md
-
Taxa.csv
Abstract
Recovery of species richness, community composition, and biogeochemistry in restored wetlands often fails to reach levels of reference wetlands. While the effects of restoration on plant and non-insect macroinvertebrate communities are relatively well-studied, much less is known about the impacts of restoration on insect communities in wetlands. The aim of this research was to determine if previously observed differences in biological structure between one reference and two constructed J. roemerianus-dominated tidal marshes extend to their insect communities. Sampling methods included pan trapping, line-transect netting, floral observations, floral clippings, and light trapping. All insect taxa and functional groups were identified to the lowest taxonomic level possible monthly from April – October 2021 and analyzed for taxa richness, abundance, and H′ diversity. Floral density and herbivory scars on J. roemerianus shoots were also measured during flowering and peak growing season, respectively. Results indicated that the reference marsh supported a more diverse insect community than the constructed marshes, although insect abundance and taxa richness were similar among sites. Additionally, temporal shifts in community composition, based on relative abundances of insect taxa and functional feeding groups, differed among marshes, likely reflecting differences in habitats in the surrounding landscapes of each site. By assessing the structural differences in insect communities between reference and constructed marshes, we can further understand the community composition of an understudied group of organisms, potentially improve restoration strategies, and support the health of tidal marshes.
README
This readme file was generated on 2023-07-05 by Emily Fromenthal | ||||||
GENERAL INFORMATION | ||||||
Title of Dataset: Beyond the Marsh: Tidal Marsh Landscape Position Influences Insect Community Structure | ||||||
Author/Principal Investigator Information | ||||||
Name: Emily Fromenthal | ||||||
Institution: University of Alabama | ||||||
Email:efromenthal@crimson.ua.edu | ||||||
Author/Associate or Co-investigator Information | ||||||
Name: Shelby Rinehart | ||||||
ORCID:0000-0001-9820-1350 | ||||||
Institution: University of Alabama & Drexel University | ||||||
Email: srinehart@ucdavis.edu OR sarinehart@ua.edu | ||||||
Author/Associate or Co-investigator Information | ||||||
Name: Jacob M Dybiec | ||||||
Institution: University of Alabama | ||||||
Email: jmdybiec@crimson.ua.edu | ||||||
Author/Associate or Co-investigator Information | ||||||
Name: Julia A Cherry | ||||||
Institution: University of Alabama | ||||||
Email: cherr002@ua.edu | ||||||
Date of data collection: 2021-04 through 2021-10 | ||||||
Geographic location of data collection: West Fowl River | Coden | Alabama | USA | |||
CON-1: 30.368 N | -88.152 W | |||||
CON-2: 30.367 N | -88.151 W | |||||
NAT: 30.368 N | -88.160 W | |||||
Information about funding sources that supported the collection of the data: | ||||||
The Society of Wetland Scientists | ||||||
University of Alabama | Department of Biological Sciences | |||||
SHARING/ACCESS INFORMATION | ||||||
Licenses/restrictions placed on the data: None | ||||||
Links to publications that cite or use the data: Please see the publication associated with these data in XXXXXXXX (doi: XXXXXX) | ||||||
Recommended citation for this dataset: | ||||||
Fromenthal | E | S. Rinehart | J.M. Dybiec | and J.A Cherry. Beyond the Marsh: Tidal Marsh Landscape Position Influences Insect Community Structure. Dryad | Dataset | https://doi.org/XXXXXXXXX |
DATA & FILE OVERVIEW | ||||||
File List: | ||||||
Taxa- count data for each insect taxon observed at study sites | ||||||
Biodiversity- total individuals | taxa richness | and Shannon-Weiner diversity (H') indeces for each quadrat | ||||
FloralCounts- total count | average count | standard deviation | and variance of Juncus roemerianus inflorescences | |||
Herbivory- percent area of herbivory damage on J. roemerianus shoots collected from each quadrat in each marsh. | ||||||
METHODOLOGICAL INFORMATION | ||||||
Description of methods used for collection/generation of data: See the publication associated with these data in XXXXXXXX (doi: XXXXXX) for details on methods. | ||||||
Methods for processing the data: See the publication associated with these data in XXXXXXXX (doi: XXXXXX) for details on methods. | ||||||
Instrument- or software-specific information needed to interpret the data: Microsoft Excel | ||||||
Environmental/experimental conditions: CON-1 and CON-2 are two constructed tidal marshes hydrologically connected via canal to the West Fowl River in Mobile County | Alabama. NAT is a reference marsh directly connected to the West Fowl River. All marshes are located in a sub-tropical estuary along the northern Gulf fo Mexico. | |||||
Describe any quality-assurance procedures performed on the data: | ||||||
General QA/QC done by all co-authors. | ||||||
People involved with sample collection | processing | analysis | and/or submission: | |||
Emily Fromenthal was involved in sample collection | processing | analysis | and submission. | |||
Shelby Rinehart was involved in sample collection | analysis | and submission. | ||||
Jacob M Dybiec was involved in sample collection and analysis. | ||||||
Julia A Cherry was involved in analysis and submission. | ||||||
DATA-SPECIFIC INFORMATION FOR: Taxa | ||||||
Number of variables: 86 | ||||||
Number of cases/rows: 146 | ||||||
Missing data codes: No data missing. | ||||||
Specialized formats or other abbreviations used: N/A. | ||||||
Variable List: | ||||||
Marsh-indicates the marsh (CON1 | CON2 | or NAT) that data was collected from | ||||
Month- month that data was collected | ||||||
Method- method used to collect data (Pan | Net | Light | FC) | |||
Quadrat- indicates the replicate quadrat (i.e. | CON1-1 | CON1-2 | etc.) that data was collected from | |||
Variables E-CH (5-86) represent counts of indiviual taxa identified to the lowest possible taxa (family | in most cases). | |||||
DATA-SPECIFIC INFORMATION FOR: Biodiversity | ||||||
Number of variables: 5 | ||||||
Number of cases/rows: 12 | ||||||
Missing data codes: No missing data. | ||||||
Specialized formats or other abbreviations used: | ||||||
H'- Shannon-Wiener diversity index; calculated using the formula H^'= -_(i=1)^Rp_i ln p_i | ||||||
Variable List: | ||||||
Marsh- indicates the marsh (CON1 | CON2 | or NAT) that data was collected from | ||||
Quadrat- indicates the replicate quadrat (i.e. | CON1-1 | CON1-2 | etc.) that data was collected from | |||
Total Individuals- total count of individual insects per quadrat across all sampling strategies. | ||||||
Taxa Richness- number of unique taxa identified per quadrat across all sampling stratagies. | ||||||
H'- Shannon-Wiener diversity calculated for each quadrat across all sampling stratagies. | ||||||
DATA-SPECIFIC INFORMATION FOR: FloralCounts | ||||||
Number of variables: 4 | ||||||
Number of cases/rows: 37 | ||||||
Missing data codes: No missing data. | ||||||
Specialized formats or other abbreviations used: None | ||||||
Variable List: | ||||||
Marsh- indicates the marsh (CON1 | CON2 | or NAT) that data was collected from | ||||
Quadrat- indicates the replicate quadrat (i.e. | CON1-1 | CON1-2 | etc.) that data was collected from | |||
Replicate- inducates which sub-sample from each quadrat is associated with each observation | ||||||
Floral count- the number of flowering J. roemerianus shoots in each observation. | ||||||
DATA-SPECIFIC INFORMATION FOR: Herbivory | ||||||
Number of variables: 7 | ||||||
Number of cases/rows: 12 | ||||||
Missing data codes: No missing data. | ||||||
Specialized formats or other abbreviations used: N/A | ||||||
Variable List: | ||||||
Quadrat- indicates the replicate quadrat (i.e. | CON1-1 | CON1-2 | etc.) that data was collected from. | |||
Marsh- notes which tidal wetland site the sample was collected from. | ||||||
Herbivory (sq inch)- area of insect herbivory damage/scars in square inches | ||||||
Herbivory (cm2)- area of insect herbivory damage/scars per cm2 | ||||||
Total area (sq inch)- total size (area) of J. roemerianus shoots in square inches | ||||||
Total area (cm2)- total size (area) of of J. roemerianus shoots in cm2. | ||||||
% Herbivory- the percent area of J. roemerianus shoots with insect herbivory damage |
Methods
Study Site Description
The study sites included one reference (i.e., natural) tidal marsh (NAT, -88.160’ N, 30.368’ W) and two 34-year-old, constructed tidal marshes (CON-1, -88.152’N, 30.368’W, and CON-2, -88.151’ N, 30.367’ W) located near the mouth of the West Fowl River in Mobile County, AL, U.S.A . All three marshes are located within 1 km of each other, with NAT directly connected to West Fowl River and CON-1 and CON-2 hydrologically connected to NAT via a canal created at the time of marsh construction. All three sites experience diurnal microtides of approximately 0.26 m. The two constructed marshes were built in 1987 as a mitigation site through the harvesting and excavation of pine savannah in long, parallel strips that lowered the elevation to near sea level. After the initial excavation of the sites, canals were created though the center of each marsh to connect them to the main canal, which links to a tidal creek originating in the natural marsh. The canals and tidal creeks in these marshes are lined with a narrow (< 1 m) band of S. alterniflora, but the marsh platforms of all three marshes are dominated by J. roemerianus. The platform at NAT also contains an assemblage of other salt marsh plants, such as S. alterniflora, S. patens and Distichlis spicata as sub-dominates (Fromenthal, pers. obs.).
Previous studies at these sites demonstrated that NAT and CON marshes differ in their biological structure and ecosystem function. For instance, NAT had greater above- and belowground plant biomass, more soil organic matter and carbon content, and higher rates of denitrification and dissimilatory nitrate reduction to ammonium (DNRA) than CON marshes, although fluxes of carbon dioxide were similar between plots with similar vegetative cover in NAT and CON marshes. These observed differences in biogeochemical pathways may affect nutrient stocks and the palatability and quality of plant biomass. Changes in soil chemistry can also affect the timing and intensity of flowering in salt marsh plants. Therefore, these sites were deemed appropriate locations in which to examine whether observed differences in their plant community structure, soils, and nutrient processing rates translate into observed differences in insect communities.
Insect Sampling
To compare insect communities between reference and constructed marshes, we conducted insect surveys in randomly selected J. roemerianus stands at each marsh over the plant growing season (April – September 2021). Four replicate quadrats (approximately 10 m2) were established per marsh in J. roemerianus stands to collect insects using a variety of replicated survey methods, including pan trapping, line-transect netting, and flower collection, as described below (n = 4 per marsh). Within one randomly selected site per marsh, we also collected insects via light trapping (n = 1 per marsh), which sampled insects from a larger area than the replicated quadrats used for other sampling methods.
Replicated Insect Sampling Methods
To account for pollinators and other flying insect taxa, we established 12 colored pan traps in each of the four quadrats per marsh. These pan traps (30 mL; 5 cm diameter) were placed on platforms positioned at the local average inflorescence height of the surrounding J. roemerianus stand to ensure that they were at an optimal height for attracting pollinators (Montgomery et al., 2021). The platforms were made of PVC pipes with open, 36 x 30 x 8 cm plastic storage containers secured on top, in which the 12 pan traps were placed (Montgomery et al., 2021). Sets of pan traps included four traps of each color that were painted either blue, yellow, or white to attract a variety of insect taxa. They contained a solution of ~2.5 cm of water mixed with several drops of unscented dish soap, which was prepared as described in Montgomery et al. (2021). Once per month, we deployed the pan traps in the field for 20 – 24 hours. Upon collection, insects were preserved in 70% ethanol to later be identified and archived in the lab at the University of Alabama according to Boyer et al. (2020).
In addition to pan traps, we conducted line-transect netting once per month in each of the four quadrats at each marsh (n = 4 per marsh). This form of sampling primarily targeted flying insect taxa of various functional feeding groups, as well as herbivores present on J. roemerianus shoots. Within each quadrat, we established one transect (5 m long) running parallel to the canal and the tidal creek in the constructed and reference marshes, respectively. Around mid-day, we swept each transect for 5 minutes using a 30-cm ringed, fine-meshed net suitable for retaining macroinvertebrates (BIOQUIP Student Insect Net). Upon collection, all insects were preserved temporarily in Ziploc bags containing a paper towel soaked in 70% ethanol before being transferred to scintillation vials filled with 70% ethanol. Upon return to the lab, insects were identified and pinned or otherwise preserved (Boyer et al., 2020).
Light Trap Sampling Method
To attract night-flying insects across all functional feeding groups, we conducted light trapping monthly in one, randomly selected quadrat per marsh using an ultraviolet light (BIOQUIP’s Night Collecting Light, DC, 12 Volt, 15 Watt Blacklight). Light traps have been shown to attract flying insects from distances as far as 40 m, and they capture many more individuals per trap than the other sampling approaches in this study. To maximize the effectiveness of each trap and for logistical reasons, we deployed only one light trap per marsh (n = 1 per marsh), for a total of three light traps overall. Light trapping was conducted in the same quadrat in each marsh from month-to-month. This light trap apparatus consisted of a platform that supported a white plastic container (36 x 30 x 8 cm) holding approximately 2 L of the same soap and water solution used in the pan traps (Montgomery et al., 2021). A suspended UV light was positioned above each plastic container to attract night-flying insects. Light traps remained in the field overnight, and insects were collected the following morning. Upon collection, they were transferred into 50-mL conical centrifuge tubes (VWR Inc.) containing 70% ethanol for subsequent identification, preservation, and storage in the lab.
Due to the large number of insects captured by the light traps, we used a subsampling method to estimate the composition of insects collected using a grid system, similar to that described by Doğramaci et al. (2010).Insect samples were poured into a pan (28 x 20 cm; length x width) that was evenly divided into 16 cells (7 x 5 cm; length x width). All cells were first scanned by eye and then under a stereo microscope for rare insects (MOTIC SMZ-140 and Meiji Techno EMZ-8TR). Then, four randomly selected cells within each sample were pooled together for identification. Lastly, the total number of insects of each identified taxa were multiplied by four to estimate the overall abundance of insect taxa within the 16-cell pan.
In addition to the insect surveying methods described above, inflorescences were collected once per month during the flowering period of J. roemerianus in April, May, and June in each quadrat (n = 4) to account for inconspicuous or minute taxa. Near mid-day, the inflorescences were clipped and quickly dropped into jars containing 70% ethanol (Montgomery et al., 2021). The inflorescences were later analyzed for insects using stereo microscopes (MOTIC SMZ-140 and Meiji Techno EMZ-8TR). As with all other insects collected, they were archived in the lab after identification.
Timed floral counts also were conducted monthly in each quadrat per marsh during May, June, and July (Boyer et al., 2020; Montgomery et al., 2020). Timed floral counts consisted of observing and filming J. roemerianus inflorescences within a 0.25 m2 area for a total of 5 minutes per quadrat during each month, with observed insect visitations recorded in a field notebook. Insects captured on film were identified to the lowest taxonomic level possible. Identification of insect taxa was largely successful, as subsequent film analysis revealed that several insect taxa were observed visiting J. roemerianus inflorescences during floral surveys, including soldier beetles (family Cantharidae) during May and June. Thrips belonging to family Phlaeothripidae and suborder Terebrantia were also found on inflorescences across the entire sampling period. While most insects visiting inflorescences were identified, small flying insects could not be visually identified to taxa. Identified taxa were included in abundance counts for the months in which surveys were performed.
Floral Counts and Herbivory
To quantify the density of J. roemerianus inflorescences potentially available as a resource to insect pollinators, we counted inflorescences in all four quadrats of each marsh in June. Inflorescence counts were recorded within each quadrat (n = 4 per marsh) by randomly selecting four smaller, sub-quadrats (0.25 m2) per quadrat. These smaller sub-samples were averaged to determine inflorescence counts for each quadrat (n = 4 per marsh).
Lastly, herbivore damage was quantified by clipping J. roemerianus stems at the soil surface within four randomly selected 0.25 m2 sub-plots in each of the quadrats per marsh in September. Upon collection, clippings were bagged and brought back to the lab for analysis, where they were then photographed and analyzed for percent surface area of herbivory damage using ImageJ software. Both halves of the cylindrical shoots were photographed and analyzed using ImageJ software (https://imagej.nih.gov/ij/, 1997-2018) to determine their total area and the area scarred by herbivore grazing. The percent surface area of herbivory was quantified by dividing the area affected by herbivory by the total area of shoots. By measuring damage in September, this approach captured the cumulative extent of herbivory throughout the growing season.
Habitat Classifications
To explore the possibility of differences in resource availability across marshes, we quantified the relative area of habitats at each marsh by creating a land use/land cover (LULC) map using the ESA WorldCover 10m v100 dataset in Google Earth Engine. Around each sampled quadrat (n = 4 per marsh), we defined a 250 m radius in which to characterize habitat types. Habitats included open water, herbaceous wetland, marsh grassland, forest, developed area, and barren. All quadrats were combined for each marsh and the percent area of each habitat type was determined.
Literature cited
Boyer, K. J., Fragoso, F. P., Dieterich Mabin, M. E., & Brunet, J. (2020). Netting and pan traps fail to identify the pollinator guild of an agricultural crop. Scientific Reports, 10(1), 13819. https://doi.org/10.1038/s41598-020-70518-9
Doğramaci, M., DeBano, S. J., Wooster, D. E., & Kimoto, C. (2010). A method for subsampling terrestrial invertebrate samples in the laboratory: estimating abundance and taxa richness. Journal of Insect Science, 10(25), 1–17. https://doi.org/10.1673/031.010.2501
Montgomery, G. A., Belitz, M. W., Guralnick, R. P., & Tingley, M. W. (2021). Standards and best practices for monitoring and benchmarking insects. Frontiers in Ecology and Evolution, 8. https://doi.org/10.3389/fevo.2020.579193
Montgomery, I., Caruso, T., & Reid, N. (2020). Hedgerows as ecosystems: Service delivery, management, and restoration. Annual Review of Ecology, Evolution, and Systematics, 51(1), 1–6. https://doi.org/10.1146/annurev-ecolsys-012120-100346
Usage notes
All data files are usavle in excel.