Fuzzy Logic Enhancement to Decision Tree Machine Learning
Dr. Juan Li, Department of Computer Science
Peter Krivoshik, Andrey Suvorov, Claudia Fortes(Students)
Machine learning is an important subject in Artificial Intelligence (AI) focusing on algorithms that take sampling data to create prediction models for making future decisions. Decision Trees are an important type of algorithms for such predictive modeling of data features in machine learning. The conventional decision tree algorithms were invented decades ago and modern enhancements like random forest are among the most powerful machine learning techniques available. The extension to decision trees creates multiple decision trees and uses their mean or mode to make the final decision. The trees and forest are the models learned from sampling data that can be used to make predictions on unseen data.
In this project, Dr. Li’s team will first investigate decision tree algorithms, as well as their extensions such as bagged decision trees, random forest and boosted decision trees. All existing extensions to decision trees focus on the uncertainty and accuracy of the final decision. The representation for the decision tree model is a binary tree representing binary logic for each feature. Each root node represents a single input variable (x) and a split point on that variable (assuming the variable is numeric). The leaf nodes of the tree contain an output variable (y) which is used to make a prediction.
Given a dataset with two features of height (x1) in centimeters and weight (x2) in kilograms, the output of health (y) as healthy or unhealthy, a binary decision tree can be represented as a graph or a set of rules. For example, a decision tree can be represented as a set of rules as follows: 1) if Height > 140 cm Then healthy; 2) If Height 80 kg Then healthy; 3) If Height <= 140 cm AND Weight <= 80 kg Then unhealthy. With the binary logic representation of the decision trees, making predictions is relatively straightforward. Given a new input, the tree is traversed by evaluating the specific input started at the root node of the tree.
A learned binary tree is actually a partitioning or classification of the input space along the features: each input feature variable as a dimension on mdimensional space. The decision tree split this space up into rectangles (when m=2 input variables) or some kind of hyperrectangles with more inputs. New data is filtered through the tree and lands in one of the rectangles and the output value for that rectangle is the prediction made by the model. This indicates the type of decisions that a decision tree model is capable of making, e.g. boxy decision boundaries with straight lines. Even though the extensions to decision tree algorithms such as random forest take into consideration the uncertainty in the decision. It does not accommodate the situations when the feature boundaries are not linear.
Digital Hand Skills for Enhanced Productivity Part ll
Craig Konyk, Michael Graves College of Architecture
James Victorio, Lillian Agutu and Ana Torres (Students)
This Students partnering with Faculty Grant proposal for summer 2017 builds upon the results of the just completed SpF Summer 2016 Research into the productivity potential of large format digital interfaces. Last summer explored the productivity curve of Architectural Software utilization on laptops versus newly acquired Ideum Digital Touchscreen Tables and documented these results.
This proposal for SpF Grant for Summer 2017 would continue this productivity exploration but this year measuring its impact upon the Professional Architectural Office Workplace. Two students will be “embedded” in an Architect’s Office in New York, where they will perform as a kind of “externship” by working on certain tasks within that office side by side employees of that office, once again demonstrating the potential productivity gains of touchscreen surface technologies in their real time use in an actual office environment.
Optimization of CCR1 Antagonists for Glioblastoma
Dr. James Merritt, NJ Center for Science, Technology & Mathematics
Fathima Faizal and Ethan Shepherd (Students)
CCR1 is a cell surface receptor involved in the recruitment of white blood cells. Over the past two years, in collaboration with Dr. Coniglio’s group at NJCSTM, we have discovered that CCR1 is also involved in the communication and trafficking of cells involved in glioblastoma, brain cancer. Current treatments for glioblastoma involve surgery to remove the tumor and aggressive chemotherapy with low patient survival rates, typically only two years. My group has constructed novel, small molecules that block the CCR1 receptor. Several of these molecules have demonstrated functional activity in Dr. Coniglio’s biological assays and effectively shut down cells involved in glioblastoma. With the support of recent SpF grants, we have constructed additional molecules that we expect to be more potent and have a higher probability of crossing the blood brain barrier. These are improvements that should bring us closer to a drug that could be used as a treatment for glioblastoma. The cellular assays that we have been using at Kean are low throughput, meaning that only a few compounds can be tested at once. Because of the variable nature of these assays and low numbers of compounds run each time, it is difficult to determine exactly how potent our compounds are for comparative purposes. This has prevented us from understanding structure activity relationships that would inform our decisions about how to improve the structures and build more potent compounds. High throughput CCR1 binding assays are available but involve a radiolabeled protein and would be extremely difficult and expensive to implement at Kean. For this SpF, we plan to utilize an outside vendor with a working CCR1 binding assay to obtain potency data for some of our existing compounds. Students will work with me to select a representative set of compounds and prepare and ship the samples to the vendor. With a turnaround time of only one month, we will obtain and process the data early in the summer enabling us to make informed decisions about what compounds to make next and begin making them in the summer and fall of 2017. With this dataset, we will also determine if CCR1 binding potency in the high throughput assay correlates with the activity that we see in Dr. Coniglio’s cellular assays. If there is correlation, as we expect, the high throughput assay would provide us with a way to triage our entire collection of ~200 compounds and test new compounds as they are generated. This is a critical dataset that we need in order to patent and eventually publish and present these novel compounds. Correlated high throughput binding and functional cellular assays are vital first steps in early stage drug discovery and enable progression of drug candidates. We will also be using this data to support research being done as part of the RFI (Research First Initiative with undergraduates) and the GSSRP (Group Summer Scholars Research Program with high school students).
“Scaling up” the Very Young Audience: Bringing Theatre and Mindfulness to PreSchool Children
Rachel Evans, Theatre
Christina Foti, James Fogerty, Darus Bridges (Students)
The use of contemplative practices, such as mindfulness, is proving to be effective in the education of young children, with research supporting claims that mindfulness improves selfregulation, selfesteem, and interpersonal interactions. As reported in Developmental Psychology, the Center for Investigating Healthy Minds at Waisman Center, UWMadison conducted a 12week study using a mindfulnessbased Kindness Curriculum (KC) on executive function, selfregulation, and prosocial behavior in preschool children (Flook, Goldberg, Pinger, and Davidson, 51.1: 4451). Those who received the KC intervention saw greater improvements in social regulation and earned higher grades in categories related to learning, health, and Social Emotional Learning (SEL); the control group exhibited more selfish behavior over time.
This SpF17 project seeks to perform Breathe In, Speak Out: Everyone Counts! For as many local area preschool students as possible. A result of SpF13 and SpF16, Breathe In is a live theatre experience with mindfulness exercises used to create uniquely meaningful performances for very young audiences, created by Professor Evans and her six research assistants from the two summers. The script for this original theatrical experience embeds both themes of empathy and interactive mindfulness practices in the traditional structure of theatregoing. The awardwinning children’s picture book, One, by Kathryn Otoshi, is dramatized using hand puppets with meditative breathing, mindful movement, and visual art projects to engage small audiences of less than 25 children, ages 4 7, to maximize engagement.
Prof. Evans will use the script completed last summer and, for SpF17, rehearse three new Kean students to be the cast of the play. Because the play was developed and performed previously, the time/effort needed to put the production together with new actors is minimal. The scenery, props, puppets and costumes are completed and materials collected. As soon as the new students learn the script, the show is able to be performed regularly. Prof. Evans and the student researchers will travel to approximately 12 different local preschools, probably with at least 2 performances per site, reaching an estimated 600 children. The team will use observation, discussion, and reflection to further refine the dramatic structure, theatrical elements, and performative procedures, as needed to improve the production’s effectiveness.
The play script to be used is the one developed as part of Professor Evans’ SpF13 and SpF17 grants, respectively: Creating Theatre for the Very Young (TVY): Taking Children Ages 2 5 on ‘an Emotional Journey of Delight’ and Theatre and Mindfulness for PreSchoolers: Using Performance as Action Research for Interactive Contemplative Practices Onstage. The 2013 student researchers (Dagnall, Gower, and Vergara) collaborated with Evans to conduct a literature review of children’s fiction that explores the themes of mindfulness for young readers, and created a bibliography of nonfiction covering strategies to teach mindful behaviors to children. Multiple improvisatory acting prompts and creative writing exercises led the SpF13 team to outline a theatrical experience for young people with a narrative structure, performance elements, and specific contemplative learning objectives. The team from SpF16 (Booth, Borowski, and Gonzalez) applied what was learned 3 years earlier, and presented formal performances at educational facilities. The drive to bring live theatre and the message of kindness to as many youngsters as possible has lead to this proposal attempting to “scale up” the size of the audience reached. Evans will use the student researchers as actors and collaborators to schedule performance sites, coordinate logistics, write a teachers’guide, and to take the show “on the road.” The opportunity for Kean students to be paid for their acting abilities would be rare and valued. Because the Kean calendar and course schedules don’t allow students to perform off campus during the school year, the SpF summer format is ideal in granting time and opportunity to bring three Theatre Education majors into schools and daycare facilities.
Online Medical and Biometric Imaging Management System
Dr. Ching-Yu Huang, Computer Science
Xiaoding Li, Shean Ballester and Reuben Hernandez (Students)
In today’s society, web based applications play an important role in international collaborations on research because of their centralized features and accessibility. It would be helpful in the medical and forensics community if imaging applications could be also managed and run online through a centralized image processing and management system. Medical imaging can range from the typical X-Ray to Computed Tomography (CT), Magnetic Resonance Imaging (MRI), or Ultrasound, while biometric imaging typically involves processing facial features, fingerprints, or irises. However, the sizes of these images are usually extremely large. Since most online applications are 3tier based architecture: browser, web server, and database server, the images have to be transferred and processed between several computers. This often takes a toll on the overall performance, making it a point of concern. The key to making these online systems successful is figuring out how to effectively and securely retrieve, manage, and analyze the images and personal information through a shared online environment. The goal of this project is to research and develop a framework for online medical and biometric imaging management systems that has advanced webbased Graphic User Interface that can perform functions in real time to load and process images, extract important features, and save the information into the back end database server. In addition, people will be able to view, edit, and save the personal information while working on the images. This project will provide advanced online tools and algorithms for people who are working at different locations and working in different fields to collaborate with one another. The project will integrate my expertise in image processing, pattern recognition, bioinformatics, web development, and databases, all of which are areas that I have been working in for more than 20 years. During these years, I have published many peer reviewed publications in these fields. I hope to take advantage of the fact that Kean University has strong programs in Physical Therapy, Nursing, and Computer Science.
The Livingston Jay Wedding at Liberty Hall
Dr. Jonathan Mercantini & Dr. Elizabeth Hyde, History
Matthew Griffi, Sarah Roberts and Megan Tobias (Students)
On April 28, 1774 Sarah Livingston was married to John Jay in the Great Hall at Liberty Hall. Sarah was eighteen years old. Jay, an accomplished lawyer, was twenty-nine. Unlike many weddings in the 18thcentury, theirs was described as a “love match” by family members. As Sarah’s brother wrote later, “Mrand Mrs. Jay can be unhappy now where, they love each other too well.”While little is known about the ceremony, the event not only brought together two people, but also united two politically powerful and wealthy families. Weddings, and the marriage of Sarah Livingston to John Jay in particular, were therefore not only social occasions, but also political affairs. And hosted at Liberty Hall, the home built by William Livingston, the wedding was one of the first of many politically significant events to take place on the estate.As Liberty Hall survives intact and, of course, is now part of the Kean University campus, historians have a unique opportunity to imagine what such a social cum political event would have been like on the eve of the American Revolution. For weddings in the eighteenth century, much like today, were occasions through which families could make a public display of consumption–consumption of food, drink,clothing, flowers, decor, and music–by which they laid claim to social, cultural, or even political rank.The Jay-Livingston union represents an opportunity to explore how this was or might have been accomplished among eighteenth-century American elites.And importantly, while the Jays and Livingstons both represented elite America, the hosting of a wedding in one of their households necessarily involved Americans from diverse social strata–from the merchants who provided consumables needed for the wedding party to the servants and slaves who would have carried out the labor to bring it about.Working with Professor Ed Johnston of the Robert Busch School of Design and his students, students in the Department of History Honors program will be conducting original research about the Jay-Livingston wedding, the history of Liberty Hall during this period, and about 18th century elite marriages more generally. History honors students will be providing the factual background that will enable Professor Johnston and his students to create a reimagining of this wedding using state-of-the-art Virtual Reality360 technology. Very little is known about the wedding in the existing secondary sources. Students will therefore conduct research in manuscript primary sources to try to discover as much as possible about the wedding, including: number of guests and attendees (one question we have: was Alexander Hamilton there?); the food; flowers, decorations, and other aspects of the ceremony. Additional details will be filled in through an investigation of elite wedding customs during this era. At the same time, this research represents an opportunity to build a greater understanding of Liberty Hall during the Livingston years, the years that spanned the American Revolution, the creation of the new American republic, and the establishment of the state of New Jersey.The wedding, then, offers Kean history honors students, the larger Kean community, and the eventual public with whom the video will be shared ways of understanding the construction of history as well as ways of using the past to better understand the present.By examining this wedding, Kean honors history students will gain greater insights into the social and cultural history of the 18th century. In particular, aspects of women’s history, but also the politics of American elites on the eve of the American Revolution will be explored.
World of Data: A Multi User Virtual Reality Distributed Data Environment
Dr. David Joiner, NJCSTM
Bryce Gomes and Xavier White (Students)
World of Data is software designed to help students, teachers, and researchers view data in a virtual reality environment. The goal of the project is to develop a 3D world that users can log into, communicate in, and upload and view data sets in 3 dimensions. Users in World of Data will be able to visualize and analyze multidimensional data in a unique interface that makes use of modern virtual reality hardware such as the Oculus Rift and Gear VR headsets, as well as Kean’s CAVE environment, but that also runs on standard PCs, Macs, tablets, and smartphones. World of Data is built using the Unity3D Game Engine, one of the leading development environments for virtual reality software. Scientific Visualization is a rich field with many tools to choose from, however the focus of the tools currently existing is on high resolution camera ready still images for publication. Realtime visualization has largely taken a back seat, particularly with regards to screen refresh rate. This has slowed the adoption of VR hardware in science visualization, as there is a disconnect between the high refresh rate required of modern VR hardware and the high resolution expected of professional science visualization tools. VR has largely been a poorly supported add on in the scientific visualization community, and the recently released lowcost headset based solutions have gone unadopted by most science visualization software projects. VR tools, on the other hand, have largely focused on game development and other entertainment projects, and there is a lack of readymade science and math software components to plug into existing VR projects. World of Data bridges this gap by taking a VR first approach to scientific visualization, creating the code to visualize scientific data from the ground up within the Unity3D environment, along with any efficiency improving algorithms, custom shaders, or data constructs required to not only render visualizations of data, but do so at frame rates required for an interactive VR experience. Development for World of Data began as part of a Research First Initiative project in Fall 2016, during which time the software tools needed to create standard 2D and 3D science visuals as well as data constructs for storing and manipulating data were created. World of Data supports unstructured data sets as well as arbitrary dimensional gridded data sets, and can display scatter and line plots in 2 and 3D, surface plots in 3D, vector glyph renderings in 3D, isocontours of functions of 3 variables, and chemical structures. We have created a proof of concept software application for PC and Oculus Rift, the World of Data Gallery, which incorporates our key visualization algorithms and showcases what World of Data users will be able to do. Work scheduled as part of the Research First Initiative in Spring 2017 will focus on the user interface to World of Data, during which time the multiuser capabilities will be added, as well as the ability to upload data and choose visualization types during runtime. In Summer 2017 as part of this SpF project, we will continue to refine and test the user interface, and expand the platforms on which World of Data runs to include the Gear VR headset, the Kean CAVE, and Android based tablets. We will focus first on Android tablets as the Gear VR is also an Android system, but will port the software to the iPad during the 2017-2018 year.
Taekwondo as a Therapeutic Recreation Intervention: Effect of a Taekwondo intervention for Fall Prevention in older adults
Dr. Kyoung Kim, Physical Education, Recreation & Health
Marie Khalil and Natasha Benosa (Students)
Falling is one of the most devastating events that an individual can experience late in life. The study found that fall related injuries and death are one of the major public health problems in developed countries (Kannuset al., 1999).Taekwondo is an official IOC (International Olympic Committee) sanctioned Olympic sport and is widely practiced in many countries. Taekwondo has many similarities to Tai Chi in that it promotes physical movement, is practiced in a safe environment, and with some modification, can be a fun and enjoyable experience.A plethora of research has provided evidence that a challenging and safe exercise program can improve balance and mobility, hence decreasing the risk of falling among older adults. Tai Chi and other forms of martial arts have received extensive attention in fall prevention studies. There are many research studies involving Tai Chi as a therapeutic intervention to prevent falls. However, other forms of martial arts have appeared much less frequently than Tai Chi due to a lack of awareness regarding their ability to prevent falls.One form of martial arts that is commonly practiced and easily available in the U.S. is Taekwondo.Developing a protocol on Taekwondo for the elderly population could be a very strong tool to produce effective and consistent outcomes. Therefore, the main purpose of the current study is to examine the effects of a Taekwondo intervention program for fall prevention among older adults: Specifically, (a) to examine the differences in balance ability among pretest, midterm test, and post test conditions, and (b) to examine the differences in mobility ability among pretest, midterm test, and post test conditions.
Developing a database on organic waste recycling and reuse as soil conditioners and nutrient sources
Dr. Dongyan Mu, SESS
Jason Wang, Andrew Diaz and Emmanuel Henry (Students)
In this project, I propose to conduct a project to develop a holistic database of information related to recycle organic wastes collected from yards, homes, restaurants, wastewater treatment, and farms, and reuse them as soil conditioners or nutrients to grow various foods for human and animals. The database is supposed to provide information on 1) the replacement of synthetic fertilizers by various recycled wastes, 2) the improvement in soil conditions in gardens/farms, 3) types and yields of harvest foods, and 4) the environmental impacts in the food growing stage. I am going to collect data by reviewing existing literatures and/or by growing vegetables in a farm at Kean University. The database will be published online for open access, which facilitates researchers and industries from different areas to understand environmental impacts and benefits in organic wastes recycling and reuse. Disposal of organic wastes in landfills has raised a significant environmental concern over methane (CH4) emission to atmosphere because breaking down organic substances in landfills releases methane, a much more potent greenhouse gas (GHG) than carbon dioxide (CO2). Especially, as landfills have become the third largest source of CH4 emissions in the United States, accounting for 18% of total U.S. methane emissions in 19902012 [USEPA Website], there are increasing interests in diverting organic wastes from landfills. Currently, encouraging organic wastes recycling and reuse has been proposed as a major solution to reduce wastes disposal in landfills. Therefore, many studies and efforts have been devoted to this area where organic wastes from homes, yards, restaurants, and farms have been collected and converted to various products through different technologies. Focused on food wastes recycling, Kean University (KU) established a composter in 2009 with which food wastes collected on campus were converted to organic fertilizers that had been applied to soil in the Liberty Hall Farm to grow organic vegetables. The harvests then were sent to student cafeterias and also the high end restaurant, Ursino for use. Since then the composter system has offered both research and educational opportunities for faculty and students in studying waste to nutrients technologies that promotes sustainability on campuses. Many research projects have been conducted on composting at Kean. For example, I conducted a life cycle assessment (LCA) and an economic analysis (TEA) for Kean’s composting system and published a paper in Waste Management (Mu et. al. 2017). In my study, the composting system has been proved to be environmental beneficial and economic feasible.
When conducting LCA project, I found out the results of existing studies on fertilizer replacement, soil conditioning, and plant growth with compost varied significantly in waste types, plant types, and soil conditions. However, the study focused on applying compost in growing vegetable was very rare. This increased uncertainties of results in my study, because using compost in the farm were counted in the LCA modeling as a benefit to replace synthetic fertilizers, and the harvest vegetables were a major income source in economic analysis. Although I conducted a primary lab of vegetables growing in soil with different compost treatments in the summer 2015, the accuracy and reliability of the lab results were still questionable, because I could only grew one vegetable with 5 trials in each soil treatment and no trial has been conducted in natural settings (in a farm). Same as the compost, results varied significantly on beneficial reuse of organic wastes by other technologies such as anaerobic digestion or pyrolysis. However, there was no comprehensive review or holistic database developed that allow the researchers and industries to easily check with information on how much effective contents are in recycled wastes to replace synthetic fertilizers, how the soil quality could be improved, how the yields would be with different foods, and the environmental impacts and benefits by using recycled wastes. The information is important to many researchers and industries to develop and evaluate waste reduction strategies. For example, when developing algal biodiesel technologies, process residues are organic materials that are full of nutrients. Researchers would be interested in applying those residues in farmlands if they have full knowledge of total benefits they could obtain. The research project proposed here is to develop a database that can provide holistic information related to beneficial reuse of various wastes as soil conditioners or nutrients. More important, the database will present previous results on a common basis. Major tasks will include literature review, vegetable growth with compost, and database establishing. The vegetable growing lab will fill the knowledge gap in using compost to grow vegetables in farms and supplement to existing literatures in this area. The project will conduct with more trails in a natural condition (a farm) to obtain more reliable results. The database is expected to publish online and can be downloaded for free. With the database, the users are able to directly access to the results of related studies, understand all aspects of organic waste reuse, compare different technologies, and evaluate different waste treatment options.
Reference: 1. USEPA Website. Basic Information about Food Waste. http://www.epa.gov/osw/conserve/materials/organics/food/fdbasic.htm.
2. Dongyan Mu, Naomi Horowitz, Maeve Casey, Kimmera Jones. Environmental and economic analysis of an invessel food waste composting system at Kean University in the U.S. Waste Management. 59 (2017): 476 486
The Allen Tavern Reclamation Project
Dr. Brian Regal and Dr. Frank Esposito, History
Matthew Waplehorst, Kelly Seader and Lillian Agutu (Students)
Few people realize that there is an important architectural and historical building site on the Kean University campus. The Allen Tavern, located just off Morris Avenue directly across from Liberty Hall, first opened during the Colonial era. In fact, it predates Liberty Hall. It continued in operation, we think, into the twentieth century. The building has since disappeared, but the original foundation stills exists. There is a long term project, supported by the university President, to resurrect the building and promote it as an historical site. Before that can take place, research must be done into the site’s history, including gathering up relevant primary documents and artifacts to tell the story of the tavern. This SPF application is intended to seek funding to do that preliminary research. The PIs will engage two history majors and one design major. Our work will become the foundation upon which to apply for further outside grant money to bring the wider long term project to fruition.
The Effect of Various Agricultural and Restoration Strategies on Vegetation Succession in the New Jersey Pine Barrens
Dr. Daniela Shebitz and Dr. Bianca Wentzell, SESS
Marionela Gavrilliuc, Bethania Rocha and Bruce Galdos (Students)
Wetlands are incredible and important ecosystems that have been described as “kidneys of the landscape” because of their ability to filter out pollutants and absorb floodwaters (Mitsch and Gosselink 2015). They have also been described as “biological supermarkets” because of their remarkable ability to support the nutrition and habitat for a wide variety of organisms (Mitsch and Gosselink 2015). Unfortunately, these ecosystems have been taken for granted throughout much of human history and have been drained, burned, or used as convenient waste disposal sites (Mitsch and Gosselink 2015). Indeed, over 50% of the world’s wetlands have been lost since the 1970s (Mitsch and Gosselink 2015) and over 47 million hectares have been lost in the last 200 years (Kent 2001). Recently, however, there has been increasing awareness about the ecological and economic value of wetlands, and more public and political calls for their protection. A provision for this concept was included in Section 404 of the Clean Water Act of 1972, which was intended to prevent future loss of wetlands in the United States, as well as to mitigate wetland loss through both the creation and restoration of wetland ecosystems. The passing of this act inspired a government mandated wave of wetland restoration activity across the country, as well as associated scientific efforts to understand and improve these restoration processes for ecological and economic benefits (Mitsch and Wilson 1996, Zedler 1996, Zedler 2000, Spieles 2005, Mitsch et al. 2012, Weinstein 2013). Wetland restoration practices are commonly employed in the state of New Jersey, which was historically dominated by unique wetland ecosystems of which only fragments remain today. Some of the most unique wetland ecosystems in the state exist throughout the New Jersey Pine Barrens (NJPB), which spans 1.1 million acres in southern New Jersey. It encompasses a wide variety of wetland types including bogs, forested swamps, and freshwater marshes and has often been described as desolate, mysterious, and beautiful, with many of its observers often underestimating its ability to support various unique organisms (Ruset 2012). Through this proposed project, we aim to determine to what extent wetland restoration has been effective on the plants, soil and water on retired cranberry bogs of the NJPB.
The NJPB have often been utilized for cranberry farming, since cranberries thrive in the characteristic sandy soil of this ecosystem, and the harvesting practice of flooding the area is easy to employ in existing wetland areas (Wen 2010). As a result of cranberry agricultural practices, the hydrology of the wetland area is often significantly changed since combinations of canals and dikes are added to control the water table, the diversity of plants in bordering areas decreases, the soil changes in composition and becomes severely compacted, water resources can become eutrophic from the application of phosphorusbased fertilizers and can even become contaminated with substantial quantities of pesticides and herbicides, which take longer to break down in the naturally acidic, sandy soil. These changes to the once natural wetland ecosystem can pose significant challenges for restoration efforts (Eck 1990, Wen 2010, New Jersey Conservation Foundation 2011, Mitsch and Gosselink 2015).
In December of 2003, one such cranberry farm of the NJPB became the focus of wetland restoration efforts after its purchase by the New Jersey Conservation Foundation (NJCF) (New Jersey Conservation Foundation 2011). The purchased cranberry farm and associated property encompassed 9,400 acres of land that now are known as the Franklin Parker Preserve (FPP). Over 1,100 acres of that land, which had been previously used as a series of cranberry bogs, then became the focus of a multiyear, multistrategy wetland restoration project conducted by the NJCF, the U.S. Department of Agriculture’s Natural Resources Conservation Service, and Ducks Unlimited (New Jersey Conservation Foundation 2011). During what is now considered the largest wetland restoration project in the northeast United States, these groups employed several different restoration strategies in order to encourage restoration success and encourage the return of Atlantic whitecedar trees, the characteristic dominant species in that area of the NJPB (New Jersey Conservation Foundation 2011, Shebitz et al. 2014).
The restoration plan was designed to address two very different agricultural strategies that had been used throughout the cranberry farm. The first agricultural strategy, which we will refer to as “traditional,” involved minimal soil and hydrological modification since this strategy was mainly employed on existing floodplains with thick hydric soil deposition. The second agricultural strategy, which we will refer to as “modernized,” involved intensive management with digging of ditches, additions of drains and pipes, and the use of diesel pumps to maintain the water table. Modernized bogs also experienced intermittent sand addition in order to promote even drainage, allow for transportation of machinery, and promote crop turnover (Eck 1990, Wen 2010, Shebitz et al. 2014). The NJCF restored both traditional and modernized bogs using any of the following three strategies, all of which included returning to natural hydrological cycles but had various modifications of the topography and plant diversity: 1. Microtopographic modification: broke up soil, created hummocks and ponds 2. Microtopographic modification with additional planting of Atlantic white cedar 3. No direct modification (self design): no topography changes, but abandoned water control structures This restoration approach resulted in 6 different combinations of agricultural and restoration strategies.
The overall purpose of this study is to examine the effects of these various agricultural and restoration strategies on vegetation succession with the ultimate goal of determining which combination of strategies is the most effective. Since the majority of the restoration activity occurred in 2004 and 2005, this project provides a unique opportunity to observe succession over a decade later in 2017. Our well rounded team of faculty and undergraduate students at Kean University will thoroughly examine the vegetation communities, soil properties, hydrology, and water chemistry at these sites in order to answer two consequential questions. First, which combination of strategies will result in a restored wetland site most similar to that of unfarmed reference sites? Second, how does each combination of strategies contribute to the successional story of each site? The answers to these questions will not only inform local restoration efforts in the New Jersey Pine Barrens, but they will also provide invaluable information for the growing field of wetland restoration. Field and laboratory methods used to answer the questions will include intensive plant diversity assessments as well as soil and water analyses, further described in the following section.
The results from this study will help to inform the world about wetland restoration efforts in previously agricultural areas, especially those with unique ecologies. For example, Dr. Shebitz is going to be linking the agricultural restoration of the NJPB to similar projects in Wuhan, China through the Central China Normal University.