Project Overview

Overview

In January 2016, Dr. Lane Foil at the Louisiana State University AgCenter was awarded an RFP-V grant of $1,847,459 to lead the GoMRI project entitled, “A Study of Horse Fly (Tabanidae) Populations and Their Food Web Dynamics as Indicators of the Effects of Environmental Stress on Coastal Marsh Health” consisted of approximately 10 research team members (including students).

 

Louisiana ranks first among the coastal states accounting for 43% of the total open estuary surface and volume of the Gulf of Mexico. The ecosystems of coastal marshes in general and especially in Louisiana are threatened due to multiple human-induced stressors and climate change. Oil spills in particular have added to the degradation of marshes in the recent past, causing acute death of plants and animals as well as long-term effects, such as increased erosion, persistence of oil-derived components in the food web and overall decrease in ecosystem services. The catastrophic explosion of the Deep Water Horizon drilling platform caused the largest man-made marine oil spill to date, and it is estimated that roughly 75 linear km of East Louisiana's salt marshes experienced moderate to heavy oiling. Acute oiling in the Gulf of Mexico subsequent to the Deepwater Horizon platform collapse created visible impact on highly prominent vertebrates, particularly birds along the coast. Within the tidal marshes of Louisiana, the PAH levels in subsurface water in the marsh locations remained high enough to have lethal effects on fish for up to two months, and it has been shown that high levels of oil were retained in the sediment of oiled marsh for much longer. Whether oil components remain in sediments of certain tidal marsh habitats and what effect these residues or degradation caused by oiling have on the invertebrate food web of vertebrates and invertebrates is now an important question. The aftermath of the oil spill provides unprecedented research opportunities for years to come. Tools must be developed not only to assess the impact of a particular oil spill but also to provide screening methods for time- and cost-efficient assessments of marsh health after future environmental insults to help guide remediation efforts. For future unpredicted insults on tidal marshes such as oil spills, techniques for rapid and intensive baseline sample collections with minimum impact on the fragile ecosystem will be needed. The measures of these techniques also could be used to monitor the health of the ecosystems during the process of land restoration and creation of marshland, e.g. by river diversions, etc. To address those needs, we propose to build on our previous studies and experiences to develop affordable and efficient measures of invertebrate food webs. Our previous studies began in May 2010, and we selected the saltmarsh greenhead (T. nigrovittatus) as an entomological model for coastal ecology for our studies. This species is native to and tightly bound to specific coastal marsh habitats. The larvae of T. nigrovittatus are apex predators that are dependent upon the invertebrate food web around them in the mud of the tidal marshes. Adult tabanids that are active in the spring and summer developed as larvae in the marshes over the past 9-12 months while adults active in the fall developed during the summer. During a two-year period beginning immediately after the spill, we conducted studies on the abundance of greenhead horse fly populations at four locations (Grand Bayou and Grand Isle which were oiled and Cypremort Point and Cameron which remained pristine). Horse fly abundance estimates showed severe crashes of adult tabanid populations as well as reduced numbers of larvae recovered from the soil in oiled areas. We developed microsatellite primers to genotype 10 polymorphic microsatellite loci of T. nigrovittatus, and these genetic markers were used to compare population size and genetic structure of adult horse flies from pristine and oiled locations. Microsatellite genotyping of six pristine and seven oiled populations detected genetic bottlenecks in five of the oiled populations in association with fewer breeding parents, reduced effective population size, lower number of family clusters and fewer migrants among populations. We propose to follow up that study with a longitudinal population genetic study of horse flies as bioindicators of marsh health and recovery. In 2011, we conducted surveys of tabanid larvae and their surrounding sediment, and specimens were archived to initiate studies on establishing the food web within the sediments. We propose to use those and future collections for metagenetic analyses to compare the micro- and meiofauna community in larval guts to that in the immediate soil environment where tabanid larvae are either present or absent. Based on this knowledge we will develop a time- and cost- efficient PCR-based diagnostic method to differentiate between healthy and biologically depleted marsh soil for use in intensive sampling. The development of PCR diagnostic tools for food web detection fits our reductionist approach to develop and verify tools that can be used by coastal ecologists to evaluate tidal marsh health.

 

Research Highlights

As of December 31, 2019, this project’s research resulted in 2 peer-reviewed publications, 25 scientific presentations and 6 datasets being submitted to the GoMRI Information and Data Cooperative (GRIIDC), which are/will be made available to the public. The project also engaged 4 Master’s level students over its award period. Significant outcomes of this project’s research according to GoMRI Research Theme are highlighted below. All is Theme 3:   Objective 1(population abundance and genetics): After the spill, population census and genetic studies of T. nigrovittatus showed massive population crashes in both adult and larval numbers in oiled locations compared to unoiled locations in 2010 and 2011. Severe genetic bottlenecks, loss of breeders, shrinking family sizes and reduced migration rates provided combined proof for a loss of effective population size and a change in population structure in oiled areas. In 2016, adult horse fly trap catches in formerly oiled areas, had numerically increased compared to the severely reduced population numbers in oiled areas in 2010 and 2011. Numbers of adults caught near oiled areas in Plaquemines Parish were not significantly different from those caught in the unaffected areas of Cameron Parish. However, seasonal fluctuations in tabanid abundance and variance in trap catches, being by nature a snapshot in time, likely contributed to the fact that we did not obtain statistical significance in all comparisons between oiled and non-oiled regions. Although increases in adult numbers in formerly oiled areas were small, the incidence rate of detecting larvae in the marsh soil rose dramatically in the Grand Bayou location. Previously, we found only 1 larva in 8 sediment samples at Grand Bayou while in 2016 we found 14 larvae in 16 samples (0.9 larvae per sample). Previously we found 1-3 larvae per sample in at Cypremort Point and Rockefeller Wildlife Refuge while in 2016 the average was 0.8 and 1.2 larvae per sample. Since the larvae are cannibalistic, finding approximately 1 per sample would be expected in a productive habitat. With fly numbers on the rise, formerly detected severe genetic bottlenecks in oiled populations have disappeared. Migration into oiled areas began to replenish formerly depleted horse fly populations in impacted regions.

Parameters of family structure that had been impacted by the oil spill (number of breeding parents, effective population size, number of family clusters) have rebounded to levels similar to or exceeding those in non-oiled control areas. Although many species on higher trophic levels still suffer in the aftermath of the oil spill, it is encouraging to report signs of recovery at the level of a top arthropod predator of the invertebrate community in salt marshes. By using both abundance and population genetics as measures, we were able to demonstrate that the greenhead horse fly is a valuable bioindicator of spartina marsh health as a top arthropod predator of the invertebrate community in salt marshes. Two manuscripts were published in the Nature journal Scientific Reports and presented at multiple scientific meetings.

Husseneder, C., Donaldson, J. R., and Foil, L. D. 2016. Impact of the 2010 Deepwater Horizon oil spill on population size and genetic structure of horse flies in Louisiana marshes. Nature Scientific Reports 6:18968; DOI: 10.1038/srep18968

Husseneder, C., Park J., & Foil, L. D. (2018) Recovery of horse fly populations in Louisiana marshes following the Deepwater Horizon oil spill. Sci. Rep., 8, DOI: 10.1038/s41598-018-31442-1

Foil, L., K. Thomas, and C. Hussenender. 2019. Invertebrate Population and Community Studies for Assessing Coastal Gulf of Mexico Environmental Health in the Aftermath of the Deepwater Horizon Oil Spill. Proc. The Gulf of Mexico Workshop on International Research. Ed. Larry McKinney, Mark Besonen, Kim Withers. Pp. 111-121

 

In year 3 and 4, we began sampling over a larger geographic scale than our original study with the purpose of establishing baselines of population structures using genotype by sequencing techniques. A genomic library was prepared for DNA samples extracted from different horse flies using the Genotype-By-Sequencing preparation protocol, and the samples were successfully sequenced on 1 lane of Illumina HiSeq at the UNH Hubbard Center for Genome Studies (a collaboration with Kelley Thomas, RFPV grant recipient). A pilot study was conducted in order to prototype a bioinformatics pipeline for a larger scale population genomics study across a much larger number of individuals. The pilot was successful, providing high quality sequencing reads which were able to be demultiplexed, and successfully call and identify SNP based loci. A total 375 T. nigrovittatus from 4 unique localities in coastal Louisiana that were collected from 2016 and 2017 covering two different seasons (early and late). We additionally added populations of T. nigrovittatus and T. conterminus from Massachusetts. We successfully extracted and quantified genomic DNA from 425 individuals, and the samples were sent for GBS library construction and sequencing on 2-lanes of Illumina HiSeq at the UNH Hubbard Center for Genome Studies. In year 4, The GBS-SNP-CROP pipeline (https://github.com/halelab/GBS-SNP-CROP/tree/master/tutorial) was used for SNP(Single Nucleotide Polymorphism) and Indel (inserts and deletions) genotyping. The multilocus genotype profile for each individual (homo- or heterozygote) is currently processed in a variety of downstream applications. Exploratory distance based population genetic analyses showed that East and West Louisiana tabanids are genetically similar, but Massachusetts samples are clearly distinct. Moreover, GBS analysis revealed the possible existence of two species in the dataset. The tabanids from Massachusetts fall into two groups that likely represent T. nigrovittatus and T. conterminus. Only a few individuals from Louisisana group with the putative T. conterminus clade. Genetic substructure is currently described in depth with further downstream analyses, e.g. fastSTRUCTURE. Populations of the T. nigrovittatus complex of flies are found along the coastal estuaries of North America from Nova Scotia to Texas. In our studies that showed the impact of the spill in Louisiana, we used 4 locations and 2 were impacted. In future applications, the population genetics study described above may help in addressing coastal environmental insults in a much broader fashion.

 

Objective 2 (food web analysis/marsh health assessment):

After the Deepwater Horizon explosion but prior to oil landfall, we chose four locations along the Louisiana coast that were predominantly Spartina marshes. As predicted, the two locations in Western Louisiana were not affected by oil while areas within both of the Eastern locations (Grand Isle and Grand Bayou) were oiled. In 2011 as part of an RFPIII project, sediment samples were taken from transects that were approximately 5 m long from each site at the high tide mark. Tabanid larvae were collected from five to seven marsh substrate samples taken using a flotation technique to separate tabanid larvae from the substrate. In 2016, 19 larvae collected in 2011 and stored in 95% ethanol were morphologically identified as tabanids and dissected under sterile conditions. DNA was extracted from the gut contents and DNA quantity (>10ng) and purity (260/280 ratio of

~1.8) was measured using NanoDrop® ND-1000, Spectrophotometer. DNA amplification and sequencing was done at a sequencing facility with a procedure developed for diversity high-throughput amplicon sequencing. Data analysis was then carried out using QIIME accessible on a server provided by Hubbard Center for Genome Studies, University of New Hampshire. The goal of this student project was to identify the food web of tabanid larvae which are top predators in the marsh. The logic is that a healthy food web is needed for a predator to develop and that food web should be a good indicator of marsh health. The gut contents from larvae as well as their surrounding environment examined for food web analysis identified 2654 species from all kingdoms with 18S rRNA genes (animalia, planta, protista, fungi, archaea). We found insect and fungi to be predominantly present in the tabanid gut contents and sediments.

 

In December 2017, the student successfully completed the exam and thesis defense. The data from the thesis will be reformatted and submitted for publication. The thesis has been published in the LSU electronic thesis database.

Bhalerao, Devika Rajeev, "Determining Bioindicators for Coastal Tidal Marsh Health Using the Food Web of Larvae of the Greenhead Horse Fly (Tabanus nigrovittatus)" (2018). LSU Master's Theses. 4377. https://digitalcommons.lsu.edu/gradschool_theses/4377

 

Subsequently, we made collections of sediment samples and insects that were the basis for research projects for 3 M.S. graduate students in years 3 and 4. The student projects are in parallel with the first student’s project on food web identification. The objectives of each of the 3 projects address 1) biodiversity of invertebrates in sediment samples along the tidal gradient at sheared marsh edges compared to unsheared banks of previously oiled islands 2) distribution of invertebrates in sediment samples collected from spartina marsh “islands” in different salinity zones 3) comparing native insect communities along clines of salinity and other environmental measures within spartina estuaries:

 

For 1) the study on biodiversity of invertebrates in the sediment of previously oiled islands in Barataria Bay, metabarcoding techniques were used to investigate the soil biodiversity in sheared sites paired with nearby intact sites. Nematodes dominated the samples, followed by annelids, arthropods, and molluscs. We also did visual confirmation of specimens in aliquots and the predominance of nematodes was confirmed. This is consistent with numerous other studies of salt marsh meiofauna. Sheared sites were found to have lower levels of diversity and different community structures than intact sites. This was likely due to the loss of plant root associated meiofauna in the sheared sites.

 

For 2) distribution of invertebrates in sediment samples collected from spartina marsh “islands” in different salinity zones, soil core samples were taken at the marsh edge, 1m, 5m, and 10m inland from the shoreline at each of the sites for each salinity location in each bay for a total of 72 soil cores. Metabarcoding techniques were used to identify the relative presence of hexapod taxa OTUs detected in the samples collected from each of the 3 salinity sites at each Bay. Collembola, Diptera, and Hemiptera OTUs were present at all sites. OTUs matching to the genus Tabanus were present in all samples. Low salinity sites contained the greatest richness of OTUs at 19, representing 15 Families from 9 Orders, followed by 12 OTUs representing 10 Families from 4 Orders at mid- salinity sites and 8 OTUs representing 8 Families from 3 Orders at high salinity sites. As expected, results of 18S metabarcoding of isolated organic material from soil sediment samples reflected a high proportion of ground dwelling hexapods than did the sweep net samples, in particular the Collembola.

 

For 3) distribution of invertebrates above ground, a survey was conducted using monthly collections over a 1.5 year period using sweep net transect sampling as well as targeted horse fly collections. Trips were made to Terrebone and Barataria bays where there were a total 18 sites between these locations. These locations were selected because they can be discretely divided into low-salinity marsh, mid-salinity marsh, and high-salinity marsh based on data from the Coastwide Reference Monitoring System (CRMS) (lacoast.gov). The collected insects were identified and inventoried. There were significant differences in family-level abundances among salinity levels. At low salinities, 61 families were collected with the most abundant families being Chironomidae and Chloropidae followed by the Delphacid and Cicadellid hoppers. At-mid salinity sites 64 total families were collected. There was little change in the amount of Chironomids and Chloropids present but there was an increase in the number of Delphacids and plant bugs in the family Miridae. Only 39 families were collected from high salinity sites, but again many of the families were present in high abundance. Delphacid and Mirid abundances were still high in addition to Ulidiids, Ceratapogonids, and Blissids. Ten families were found to be bioindicators of different habitats. Three families were found to be indicators for low salinity sites, one family was found to be an indicator for mid, one family was found to be an indicator for high salinities, two families were found to be indicators for the combination of low and mid salinities, and three families were found to be indicators for the combination of mid and high salinities. Over the course of the year, 15 plant species were collected; 12 at low salinity sites, 7 at mid, and 4 at high. While species composition varied between salinity levels, Spartina species had the highest total ground cover at each salinity level. Other notable plant species are the sedge Schoenoplectus americanus and the legume Vigna luteola at low salinities and the grass Distichlis spicata and sedge Bolboschoenus robustus at mid salinities. We also conducted adult tabanid collections, there were three trapping sites at each of the different salinity zones in both bays. Relative abundance of T. nigrovittatus was consistent among salinity zones with a clear bivoltine occurrence at all locations. T. acutus was found to be repuscular/nocturnal in all 3 salinity zones while T. hinellus was found in low abundance in diurnal periods.

We established the importance of the tabanid species Tabanus nigrovittatus as a bioindicator of marsh health in previous studies in high saline localities of coastal Louisiana. In this study, we examined the prospect of using other species of tabanids from different salinity zones as additional bioindicators. The two other species that were found in the extended survey were shown to be anautogenous (require a blood meal to develop the first batch of eggs; T. nigrovittatus is autogenous) by dissection techniques; one of the species was found to be nocturnal and the other was found in low numbers.

Therefore, the two other species are of little value as bioindicators that would be both prominent and easily captured. We did find that T. nigrovittatus was equally represented in all of the salinity zones studied which adds value to the use of this species as a bioindicator of marsh health. These studies are the subject of 3 master’s theses being prepared currently and publication of the material will follow. An example of the recent presentations made regarding the progress of these studies is provided below.

 

Davis, D. M.; Husseneder, C.; Foil, L.D. Describing the Spatial Occurrence of Horse Fly Species along the Louisiana Coastline Entomological Society of America South Eastern Branch Conference 3/3/2019 - 3/6/2019 Mobile, AL

 

Aker, B.; Husseneder, C; Foil, L. Establishing Inventories of Insects along Salinity Gradients in Louisiana's Coastal Marshes. 2019 Entomological Society of America South Eastern Branch Conference 3/3/2019 - 3/6/2019 Mobile, AL

 

Patrick Rayle, Claudia Husseneder, Lane D. Foil Comparison of intertidal metazoan biodiversity between sheared and intact margins of previously oiled estuarine islands. 25th Biennial Coastal Estuarine Research Conference 11-3-20 - 11-7-20 Mobile, Alabama

 

Darius Davis. Occurrence of Horse Fly Species along the Louisiana Coastline 10th Annual Graduate Student Symposium 10-25-2019 Baton Rouge, La

 

Patrick Rayle, Claudia Husseneder, Lane D. Foil Comparison of intertidal metazoan biodiversity between sheared and intact margins of previously oiled estuarine islands Rayle, P 10th Annual Graduate Student Symposium 10-25-2019 Baton Rouge, La

 

Ben Akers, Claudia Husseneder, Lane Foil Biodiversity Analysis of Insect Communities in Relation to Plant Communities and Underlying Salinity Gradient in Coastal Louisiana Tidal Marshes 10th Annual Graduate Student Symposium. 10-25-2019 Baton Rouge, La

 

Lane Foil, Ben Aker, Patrick Rale, Darrius Davis, and Claudia Husseneder. Establishing/Reestablishing Baselines for Invertebrate Populations in Louisiana Tidal Marsh Estuaries. 2020 Gulf of Mexico Oil Spill and Ecosystem Science Conference Feb