Spencer Rhea and Alex Smith joined us to share updates on their SWISLR projects created at the All-Hands meeting this past May. Spencer started by sharing his project “S.T.E.W.P. - Salting the Earth with Purpose” (name yet to be finalized) focused on understanding how SWISLR affects carbon emissions in coastal wetlands, with a focus on how soil properties can mediate the response. Alex Smith followed up with the “G.Y.S.T.” project focused on how we can adapt ecosystem models to SWISLR field work going on and how field work can potentially be adapted to inform models. You can watch this recording posted in the SWISLR Seminars page.
As Spencer introduces his project, he shared with us that wetlands store a lot of carbon due in part to the soils that allow wetlands to hold onto the carbon for a long time. With global climate change these carbon stocks are under threat and many of our wetlands are disappearing under sea level rise. The effects of saltwater intrusion into wetlands and soils is variable and there are many different effects that can compound on each other. To gain a better understanding of SWISLR effects on soils he collected soil samples from other SWISLR RCN member’s research sites and compared the soil and chemical properties between salt exposed sites and protected sites. This project is still in the beginning stage and Spencer will investigate wetlands soils further through a “Common garden experiment” and a “Common substrate experiment”. Through this work Spencer wants to identify what soils are most resistant to SWISLR and welcomes other soil researchers to get involved in this soil examination! Alex further discusses the complexity of SWISLR impacts to ecosystems and explains how these complex interactions compounding on each other require complex and innovative models. To arrive at these innovative models we first need to understand the popular adaptable models available. Through a synthesis of adaptable terrestrial models Alex arrived at ELMs, specifically the Energy Exascale Earth System Model (E3SM), a model that simulates coupled processes and interactions among water, energy, carbon, and nutrient cycles. However, the ELM does not include detailed hydrology, ecology, or biogeochemistry - all traits that we would need for modeling SWISLR. But, this model is adaptable and modules can be added to refine the model for SWISLR. Alex talks through the modules of FATES and PFLOTRAN and their pros and cons with modeling SWISLR. The current inclusion of coastal habitats in ELM, the adaptability of ELM through modules, and the simplicity in the structure of the model make a great case for why we should use ELM to model SWISLR. In addition, field studies could better inform these models. Therefore, Alex is completing a literature review of SWISLR field studies to identify what researchers are collecting and the traits that would be important to include in a model. Studies like G.Y.S.T. and S.T.E.W.P. are helping to inform our SWISLR RCN on how we are and how we should be collecting ecosystem disturbances due to SWISLR. As we, as researchers, continue to complete experiments and collect data, there should be a concerted effort to identify and abide by a standard set of methods.
0 Comments
Friday SWISLR webinars are happening again this school year! SWISLR Seminars will be once a month, each one focused on a project that came out of the SWISLR All-Hands meeting. Many people during the All-Hands meeting said that the SWISLR RCN can help their projects by providing connections to people or to data outside of their social network. Therefore, each project will have a chance throughout the school year to host a SWISLR Webinar and a discussion that will help further their project.
The Projects and their project leads:
For this first webinar each project got a chance to highlight what they proposed at the All-hands meeting and what has been accomplished since then. You can check out these pitches by watching the recording of this webinar on youtube. After each project lead, or project representative, introduced their project, we had breakout groups to further discuss what has been going on within the working groups. In these breakout groups people shared ideas and data sources. The S.T.E.W.P. project now has an additional soil sampling site, Stories of Change has some more people to talk to about the changes they are observing, and the SWISLR RCN has some new members joining the list serv! I am looking forward to hearing more about these projects in the coming months, and can’t wait to see what connections grow from them. A commonality that we all share is data – if you don’t collect your own data you often use publicly available data to find information through national or regional datasets. Working with big publicly available data has introduced the need for some guiding principles in order to protect sensitive data and allow access to everyone equitably. One such set of principles are the FAIR data principles. This acronym stands for findable, accessible, interoperable, and reusable and these guidelines put emphasis on enhancing the ability of machines to automatically find and use the data, in addition to supporting its reuse by individuals. Without these pieces it is hard for data reuse to happen. Reuse is important in research because many people rely on syntheses, regional generalization, and comparing methods or results. Reuse and access is also important to community members and practitioners so they can make decisions using the best data possible. Specifically with SWISLR, FAIR data practices (https://www.go-fair.org/fair-principles/) are necessary if we want to make general statements of the issues SWISLR causes. There are really hard to solve problems happening right now, and in order to come up with resilient solutions, it is more beneficial if we can use data across regions and disciplines.
Organizations such as NSF, NASA, and journals (e.g. AGU) are creating policies where researchers are required to share their data and make the data publicly available in data repositories. There is a large landscape of data repositories available on the web that can make it harder to find the source of data you are looking for. Although they are helpful for housing data and making them publicly available, the number of repositories available can make it harder to know where to look for the data you are trying to find. There are efforts to combine these repositories as data atlases that constrain the geography or google data that houses the links to these data sources. However, they do not cover the full depth of data available on the web. Given this, Dr. Anna Braswell has asked – “How can we make data more discoverable?” One way is that we can create a centralized data repository for all data like NOAA’s National centers for environmental information (https://www.ncei.noaa.gov/) or the European Marine Observation and Data Network (https://emodnet.ec.europa.eu/en). However, this requires a lot of time, space, and money that is not readily available. Another option is to require improvements to the information included with the open data so it can be more findable for systems like google data that scrape the web for you when you search. But this takes a lot of buy in from all the repositories housing data and researchers who publish their data. Although this is doable, it will take time. Finally, one option forward is to create a community around data. To achieve this, Dr. Braswell is creating a site where people can create posts about data they use and post questions they have about data availability and useability: https://copecomet.github.io/Coastal-Data/. The goal of this data curation service is to create a community around data and help improve the ways in which we use and find data. The most recent SWISLR webinar discussed ecological system shifts throughout the NACP. Justin Wright and Aeran Coughlin introduced us to their work on plant and microbial community change in the Alligator River National Wildlife Refuge. The RCN hopes to synthesize the ecological impact of SWISLR. Existing studies are typically site-specific and species-based, limiting the possibilities for generalizations. However, a syntheses of existing studies will help determine what to expect and help develop best-practices for future coordinated research. Therefore the second half of the webinar was a discussion about what people are studying and how traits are being linked to SWISLR induced changes.
Justin Wright starts off by talking through how SWISLR is inducing changes not only along the coast but affects inland plant communities due to human alterations. Specifically in the Albemarle-Pamlico peninsula, the drainage networks allow for salt to move inland. The wright lab has looked at how plant communities have changed because of this salt incursion into the peninsula. They found that species composition has shifted to varying degrees, but were not quite sure what the causes of these changes were. There was some hint that salt and elevation was correlated to the composition shifts, however the relationships were stronger for plant-traits such as basal area (Anderson et al. 2022). These findings indicate that SWISLR is responsible for the drastic ecosystem shifts into the transient state of ghost forests that are seen throughout the NACP. However, the mechanisms of these changes is not entirely known since correlations do not prove that salt is responsible for composition changes. So, Aeran is diving deeper into these interactions to identify why plants are reacting to salt coming into the soil in the Alligator River National Wildlife Refuge. They are focusing on plant-soil interactions and have already found that the microbial soils are important for tree and shrub growth, with forest soils in the Alligator River being primed for shrub growth. Being at the start of their dissertation, Aeran has many more questions to answer and has primmed the RCN to think about how to study plant interactions moving forward. The RCN contains more than just plant ecologists, and people are thinking about many aspects of SWISLR from human land use to coastal hydrology. Most plant ecologists who were in the room said they mostly study community composition and measurements of function aren’t as prevalent. Many use indicator species (like Justin and Aeran) to identify ecosystem changes due to SWISLR but it is hard to connect species to salinity specific stress since a lot of plant trait changes are the same regardless of whether a tree is stressed by inundation or salinization. A paper by Keryn Gedan and Eduardo Fernandez-Pascual was highlighted as an example of linking plant traits to salinization. Another example of linking plant traits to salinization is Serina Wittyngham’s work at VIMS. They are curious about what constrains the lower limit of Phragmites in the salt marsh since the upper limit is better understood. They are measuring plant traits and rhizome starch content of Phragmites to see how treatments such as salt and herbicide influence the energy stores and ability for spread. You can watch the Wright lab talk and the summary of the discussion rooms on youtube. The discussions so far have been 15-30 minutes long, often cutting some connections and thoughts short. We hope you all will continue these discussion at the upcoming all-hands meeting! We took a break from the regular format of our webinars to host a Show-and-Tell! This format was to help us start talking about the 3rd theme: How are water management and climate change interacting to determine the magnitude, extent, and duration of saltwater intrusion within and across the NACP? Although we know that humans have altered the hydrology across the coastal plain, the extent to which human activities lead to variation in SWISLR impacts is not well known. Regional differences in water management practices create areas of “weird” water movement since human infrastructure can potentially constrain the natural hydrological processes through flow control technologies. Variabilities within the water infrastructure makes the interactions between water management and water harder to understand regionally. When predicting the timing, magnitude and duration of saltwater intrusion and sea level rise “bathtub” models are typically used that assume topography as the dominant driver of vulnerability. However, the coastal plain ecosystem is dependent on human decisions of water management and infrastructure. Therefore, to better understand the full impact of SWISLR we require models that integrate sea level rise, rainfall-runoff relationships, evapotranspiration, natural geomorphology, water infrastructure (canals systems, dikes, levees) and water management (pumps, check dams, irrigation, groundwater extraction). In the SWISLR Show-and-Tell we invited RCN members to tell us about the weird water where they work and live: Some alternatives to photos were Ashley Helton's coastal viewer, Xi Yang's drone video, and Jean-Christophe Domec's site info. You can watch everyone's description of their photos here.
The SWISLR RCN has now hosted three webinars covering different themes of research related to saltwater intrusion and/or sea level rise. This webinar covered the question of: Where are ecosystems vulnerable to SWISLR induced transitions along the NACP and how can we predict vulnerability? You can find a recording of this seminar here.
Dr. Elliott White starts us off by talking about what we can do, and what we can’t do with remote sensing. Modeling has shown us that we are losing coastal wetlands overall (Osland et al. 2022). There are multiple threats to coastal wetlands that lead to increasing salinization of surface groundwater which in turn yields declines in coastal wetlands. There is also spatiotemporal variation in the drivers of saltwater intrusion and wetland migration/habitat loss (White and Kaplan 2017). Some wetland types are much more vulnerable, freshwater forested wetlands are the most vulnerable when talking about wetland migration. The habitat of freshwater forested wetlands is decreasing with marsh migration and saltwater intrusion which is increasing the amount of ghost forests we see along the coast. This increase in ghost forests has been an interest for scientists for a while now but is growing in public attention due to the clear visual they provide of climate change. These changes are clearly seen through remote sensing practices, and Dr. Elliott White talks about his paper showing the net loss of 13,682 km2 of loss in coastal forested wetlands along the coastal plain (White et. al. 2021). He shares his experience in learning and engaging in remote sensing methods. Remote sensing can take less time, less money, less permitting, and less maintenance of field equipment than field science. Additionally, remote sensing provides us the ability to scale up and collect information at larger spatial scales. After Dr. Elliott White’s talk, the participants had a discussion on what information we would like to have to better understand vulnerabilities to SWISLR impacts along the NACP. Participants were invited to list their data wishes if no collection limitations were applied. About 27 unique ideas of data to include when accounting for NACP vulnerability were brought up. Participants were then asked to look at their data wishes and tell us of current data or useful indices that can be used to show the information participants want. About 16 of the 27 ideas currently had data sets, remote sensing abilities, or proxies available. The information collected is a growing database of data needs for the SWISLR RCN. You can see the board of ideas that were brough up here. We know that while many groups are having the conversation about how to carry out SWISLR mitigation efforts, not all groups have a seat at the table or are even consulted. This excludes valuable voices and perspectives while also creating the conditions for already vulnerable groups to shoulder the bulk of the consequences of SWISLR without being recognized. In addition to the regional differences of SWISLR, the climate impacts are not evenly distributed and often those who have the most to lose often have the least capacity to deal with the impacts of climate change. The goal of this webinar was to discuss this issue of exclusion.
Dr. Ryan Emanuel leads us through some though provoking questions and valuable resources surrounding this historical issue of exclusion. He began by asking “Who is at the preverbal table where decisions are being made about SWISLR?” He later questioned the need for this table at all, and if there are better ways to communicate SWISLR mitigation efforts. In an ideal world, decisions being made reflect consensus among people who participate in these decisions and those who contribute information. If the people who participate, or contributions being made, don’t capture the diversity or perspectives on the issue at hand, then decisions made might not meet the needs of communities that have much at stake. Additionally, many decisions do not end up reflecting the consensus. An example of vulnerable communities being left out of the management plans is the Jean Charles Choctaw Nation in Louisiana. The relocation strategy that the Jean Charles Choctaw Nation originally came up with has since been changed without their engagement by the state of Louisiana who secured the funds for the relocation plan. This lack of engagement resulted in plans that would have harmed the Choctaw Nation community and ultimately that relocation has not actually happened. Addressing issues in environmental planning in meaningful and culturally appropriate ways is more than just insuring raw participation. It is also about how we do the science that accompanies mitigation planning. The research we do is likely to become incorporated into the management and planning along the coast, therefore we need to do our due diligence to include the voices and methods of the vulnerable communities. A global review of how indigenous voices are included in our research found that the vast majority of climate studies practice an extractive model in which we use Indigenous knowledge systems with minimal participation. If we, as researchers, hope to make meaningful contributions to society than an effort needs to be made to include the communities of our study locations in deciding where to conduct the research, what questions to ask, and the methods we use. This idea of inclusion should also be extended to the institutions that serve marginalized individuals. A great resource to better our science is AGU’s Thriving Earth Exchange. The zoom room was primarily filled with individuals affiliated with universities, working with other universities or state agencies, and researching environmental science. Through our discussions we concluded that coastal residents do not want to know what is causing the flooding they are currently dealing with, instead they want to know how to fix it. However, our skill sets don’t mesh well because we don’t have the answers on how to fix it. Our roles in as scientists or researchers makes it difficult to provide solutions for people who just want to know what we can do to stop the SWI impacting my crop field. Instead, our information is becoming inconvenient and upsetting. Additionally, people on the ground are experiencing frustrations due to seeing some things work and some things not with seemingly no reason. A focus on community informed research is needed if we are to create inclusive solutions for climate change impacts. What happens when a bunch of scientists all join a zoom room together? An interactive discussion on what is needed to better understand SWISLR! The quotes throughout this blog are taken from the seminar at the beginning of the webinar and from the discussion board created by the participants at the end of the webinar.
Saltwater Intrusion and Sea Level Rise (SWISLR) is occurring globally, causing regional impacts and outcomes due to the variability of the coast. Our goal as a Research Coordination Network (RCN) is to not only collect and synthesize the current scientific understanding of SWISLR impacts, but to ensure that future research is providing the knowledge and insights that coastal community leaders need to protect the most vulnerable communities. To learn more, you can watch the recording of the seminar found here. To start, the RCN is hosting monthly seminars to introduce the issues of SWISLR. Emily Bernhardt starts off the first meeting with an introduction to the RCN and to SWISLR. She states that “the goal of these monthly webinars is to build out a community to hear from folks about how we can best coordinate and work together.” Coordination is increasingly important due to the increasing vulnerability and the regional differences seen throughout the eastern coastal plain. The causes of SWISLR are relatively well known, however the study and knowledge of impacts and areas of heightened vulnerability are localized and disconnected. Working with others on an interdisciplinary team can facilitate “wholistic systems thinking regarding the causes and consequences [of SWISLR]” (Ellen Herbert) which can then lead to “richer research and more relevant solutions” (Ryan Emanuel). When posed with the questions of; “What would you bring to the RCN?”, “How would it help your work to connect with others who work on SWISLR across the NACP?”, “What would you need from the RCN?”, and “What should the RCN prioritize within SWISLR research?”, the participants had similar answers to each other. Many people were bringing their own perspectives on SWISLR to the RCN. People who have been working in the field their whole career have unique knowledge of study sites, William Conner says that he has “almost 50 years of working in forested wetlands”. Additionally, people are bringing specific methods to the RCN, like “remote sensing skills” (Xi Yang), “physics-based modeling approaches of storm surges and subsurface salt transport” (Holly Michael) and “predictive mapping” (Becky Epanchin-Niell). People wanted more collaboration and a synthesis of current research. Chris Elphick specifically wanted “a better sense of what is being done by others so as to not duplicate effort.” Some wanted better solutions and strategies for agricultural impacts, ecosystem transitions/loss, and shifts in demography (Matt Kirwan). Looking at what people would bring to the RCN and what people need from the RCN led to a discussion of the final questions we asked: What should we prioritize? Moving forward, the RCN needs to prioritize two main ideas. First, to facilitate diverse partnerships that not only benefit but also engage people currently facing SWISLR-related changes. And second, to synthesize SWISLR knowledge, research, data, and other SWISLR-related information. A synthesis would allow us to create connections, provide a broader spatial scale of SWISLR, and identify gaps in the data. |
AuthorsKiera O'Donnell:kiera.odonnell@duke.edu Archives
December 2023
Categories |