Spartina Expt 2013 Manuscript #1

Rachael E. Blake, Jill A. Olin


                S. alterniflora is an important foundation species exposed to multiple stressors.  Importance of Spartina marshes, especially in Louisiana. 

Oil spill happened in 2010. Oil impacts from Mendelssohn and Lin.  Oil spill is just one source of the stressors. 

Herbivore impacts well-studied and may be especially strong in Gulf Coast area, from Silliman, Don Strong, other insect folks. 

Foodweb marsh papers. 

We wanted to look at plant responses as well as responses in the herbivores that have the potential to "cascade up the food web".   <- re-word this last bit
Specifically, we wanted to address these questions:
1) What are the  physiological and biomass impacts of oiling and herbivory on Spartina alterniflora?
2) Are the multi-stressor effects additive, synergistic, or antagonistic?
3) Do oiling effects "cascade up the food web"?   <- This needs to be re-worded!!!!!


                 Experimental Design

                To examine the impacts of multiple chemical and herbivore stressors on the important saltmarsh foundation species, Spartina alterniflora (hereafter Spartina), we conducted a factorial mesocosm experiment in a greenhouse at Louisiana State University (LSU), Baton Rouge, LA, during May - July 2013.  Each mesocosm consisted of one 19 L plastic bucket with a drain hole (plugged with a rubber stopper) near the bottom, inside which a 13.66 cm diameter, 30 cm deep flower pot rested on a brick (to improve drainage).  Clusters of 7 - 8 mature stems and associated root material of Spartina were planted in each flower pot in a substrate comprised of 2.85 L potting soil and 0.95 L sieved estuarine mud collected from the Louisiana Universities Marine Consortium lab in Cocodrie, LA (LUMCON).  All mesocosms were enclosed by a supported mesh cover fine enough to contain the herbivores.  Treatments receiving oil were segregated in a different section of the greenhouse from treatments not receiving oil to limit the effects of airborne compounds on the non-oiled mesocosms.

               We conducted the experiment at 20 ppt because this is the approximate salinity of saltmarshes near Grand Isle, LA that have been monitored by the Coastal Water Consortium ( since the Macondo spill in 2010.  However, because the Spartina had been growing at 0 ppt on the LSU campus in Baton Rouge, we had to acclimate our plants to salt water (made with Instant Ocean©) prior to beginning the experiment by increasing salinity by 5 ppt each week until we reached our target salinity of 20 ppt.  Following acclimation to 20 ppt, we watered the plants every two weeks with 20 ppt water.  This allowed the water levels to drop over the two weeks, simulating the very low tidal amplitude typically found in saltmarshes in this geographic area.  When we filled the mesocosms with water at the two-week interval, this simulated a flood tide event where the surface of the marsh was covered with water.  Water was lost from mesocosms through plant transpiration, as the salinity in the mesocosms remained 20ppt and did not increase as would be expected with evaporation. 

               Chemical stressors included oil (MC252 surrogate crude oil) and chemical disperant (Corexit 9500) applied in three treatments: each chemical individually, and the two chemicals together.  110ml of surrogate crude oil were added to each oil treatment, and 2.2 ml of Corexit 9500 were added to each dispersant treatment.  For treatments receiving both oil and dispersant, the two chemicals were mixed together before being applied to the mesocosms.  Chemicals were painted on the bottom half of the standing aboveground biomass to simulate oiling conditions in coastal saltmarshes during the spill (Mendelssohn 2012), and because other studies have shown impacts when half the stem biomass was oiled (Lin 2012).   To simulate the same action in no-chemical treatments, salt water (20 ppt) was painted on the bottom half of the Spartina.  Any chemicals or salt water remaining after painting were poured on the surface of the sediment in the mesocosm.  The oil dose for each experimental unit was based on the average high oil concentration found in saltmarsh sediments sampled in 2010 in Barataria Bay, Louisiana during the spill (samples from Gene Turner or CITE their oil paper maybe???).  We used a dispersant to oil ratio of 1 part dispersant for every 50 parts oil.   This ratio is in the mid-range of possible ratios at which dispersant was applied during the spill, and found in the literature and those applied at the well head (CITE Turkish guy...maybe, and other references...).   

               Herbivores included the snail Lirroraria irrorata (hereafter Littoraria), and two species of phloem-feeding insects from the genus Prokelisia (Prokelisia marginata and Prokelisia dolus).  These snails are a common and abundant saltmarsh invertebrate in Louisiana, and are strong herbivores that can decrease Spartina biomass by farming fungus on the plant tissue (Silliman 2002, Silliman 2003, Silliman 2005).  Three snails were placed in each snail treatment, and were checked once per week to make sure they were alive.  Snails that died were replaced with new live snails to maintain equal grazing pressure in all treatments.  A complex of two insect species was used because both commonly occur in Louisiana saltmarshes, and it is very difficult to differentiate individuals from the two species without dissection.  Thirty individual insects were initially added to each mesocosm receiving insects, and were allowed to increase in abundance during the experiment. 

                Sample Collection and Processing

                  We collected samples during the experiment to monitor experimental conditions and assess treatments.  Four HOBO Pendant® loggers ( spaced throughout the greenhouse monitored light levels and air temperatures during the experiment.  To further assess growing conditions in the mesocosms, we sampled porewater sulfide during the final week of the experiment (CITE METHOD??).  We collected Spartina growth data by measuring stem height (from sediment surface to top of stem including any inflorescence), stem diameter, and total number of stems in each mesocosm at two-week intervals.  At the same sampling interval, we also collected small soil cores to assess oiling treatments and stable isotopes from a randomly selected subset of mesocosms.  

                  For stable isotope analysis of carbon, nitrogen and sulfur, plant leaves were cleaned of foreign debris and rinsed with distilled water.  All samples were dried, ground to a fine powder and weighed to approximately ~4.0 mg for carbon and nitrogen and 11-13 mg for sulfur and the relative abundances of carbon (13C/12C), nitrogen (15N/14N) and sulfur (34S/32S) were determined on a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20-20 isotope ratio mass spectrometer (Sercon Ltd., Cheshire, UK) and an Elementar vario ISOTOPE cube interfaced to a SerCon 20-22 IRMS (Sercon Ltd., Cheshire, UK) at the University of California Davis Stable Isotope Facility.  Analytical accuracy was 0.1‰ for δ15N data, 0.06‰ for δ13C data and 0.3‰ for δ34S based on NIST standards.

Need further Stable Isotope Methods for the snails.

                  We used a LI-6400XT portable photosynthesis system ( to measure Spartina ecophysiological responses.  We took fluorescence measurements at night during weeks 1 - 3, and 8 of the experiment to assess plant stress through the quantum efficiency of photosystem II .  To further examine plant responses, we took carbon assimilation and quenching measurements during the day in weeks 2 - 3, and 8.  We measured three leaves per mesocosm in a random subset of mesocosms at each night or day sampling point due to the length of time required for the ecophysiology measurements.  However, at the end of the experiment three leaves from all mesocosms were measured. 

               At the end of the experiment, the mesh covers were sealed around the Spartina stems at the sediment surface to enclose all herbivores, and the stems were clipped as close to the sediment surface as possible.  Sediments were sieved and all root matter collected on a (WHAT SIZE WAS THE SIEVE???) sieve was retained.  All plant matter was frozen until it was processed.  Plant biomass was separated into live and dead stems and live and dead roots and rhizomes, dried to constant mass at 60 °C, and weighed to obtain dry mass values.  Snails were weighed to obtain final mass, and change in mass per day was calculated to examine growth of snails during the experiment.  Insects were separated from the plant stems and were counted using a dissecting microscope to obtain final abundance.      

               Statistical Analysis

              We used three-way analysis of variance (ANOVA) to examine primary effects of crude oil, dispersant, and two herbivores on S. alterniflora growth, biomass, and ecophysiology, and the effects of the chemical stressors on the herbivores.  We also calculated the size of these effects by centering and scaling each variable (subtracted from mean and divided by standard deviation) to obtain standardized regression coefficients (