A GIS Approach to Analyzing Lake Beresford for the Production and Phytoremediation Potential Using Pistia Stratiotes and Eichhornia Crassipes

A GIS APPROACH TO ANALYZING LAKE BERESFORD FOR THE PRODUCTION AND PHYTOREMEDIATION POTENTIAL USING PISTIA STRATIOTES AND EICHHORNIA CRASSIPES A PAPER SUBMITTED FOR COMPLETION OF SENIOR RESEARCH FOR THE COLLEGE OF ARTS AND SCIENCES STETSON UNIVERSITY BY ALEX CLARK IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF BACHELOR OF SCIENCE ENVIRONMENTAL SCIENCE AND GEOGRAPHY ADVISORS Dr. J. Anthony Abbott Dr. Jason Evans MAY 2016 Table of Contents Introduction 1 Literature Review & Study Area 1 Methodology & Observations 3 Discussion & Conclusion 4 Works Cited 9 List of Illustrations Figure 1 - Study area 6 Figure 2 - Various plant cover maps associated with numbers in table 2 7-8 List of Tables Table 1 - Data used in analysis, source/provider, data type and content and purpose in analysis 5 Table 2 - Plant cover/site selection and associated surface area (derived from ArcGIS), biomass and estimated removal rate for the total estimated biomass 5 Acknowledgements I’d like to thank Dr. Tony Abbott and Dr. Jason Evans for aiding in the recovery of my initial compost based project and assistance throughout the overall experience; Dr. Clay Henderson and Dr. Kristen Work for providing supplemental information; The entirety of the Theta Psi chapter of the Alpha Tau Omega fraternity for support in the process; Jake Moore, Shayne Fitzgerald, and Kenny Lane for proofing multiple drafts; And finally Stetson University for the opportunity. Abstract (the executive summary) One of the many problems our local waterways are facing is nutrient loading. Nutrient loading occurs when excessive nitrogen and phosphorus enter a given waterbody and as a result eutrophication can occur. Eutrophication is the overall degradation of water quality due to the buildup of available nutrients; algal blooms, lower dissolved oxygen, higher turbidity, changes to pH are among the changes that can occur in the ecosystem that result in a struggling aquatic community. To aid in phytoremediation efforts (using plants as a biotic agent for bioremediation) and to propose an alternative to current invasive growth management practices I used ArcGIS to model three possible scenarios in which Pistia stratiotes (water lettuce) and Eichhornia crassipes (water hyacinth) could be used in the phytoremediation of Lake Beresford and the surrounding area. ` i Introduction Nitrogen (N) and phosphorus (P) plague Earth’s water cycle (Han et al 2013; Lu et al 2010; Greening and Janicki 2006). Found in fertilizers and other industrial grade chemicals N & P end up as both point and non-point components of this water pollution (i.e. fertilizer run-off, factory effluents, sewage leaks, etc…) (Han et al 2013; Lu et al 2010). As this type of contamination builds in any waterway a process called eutrophication begins. Eutrophication, now common throughout many marine ecosystems, is when aquatic vegetation (in response to nutrient overloading) grows so ferociously that water quality and ecosystem health degrade (Gnanavelrajah and Sundaraling 2014; Lu et al. 2010; Greening and Janicki 2006). To combat the negative effects of eutrophication in the early 90s the Environmental Protection Agency (EPA) started successfully using phytoremediation practices to clean up chemical contaminates in the environment (EPA 2012). A GIS modeling project by Jonathan P Fleming et al. featured in Applied Geography sheds light at the possibilities for complexity of site selection models (Flemming et al. 2012). While this more advanced GIS model centered around the Little Bear Creek Reservoir in Alabama, both provides a wonderful starting point for the establishment and guarantees some sort longevity in the aquatic plant community establishment projects in the area, the local constraints and aquatic macrophytes used in the Alabama study compared to the ones I used for my model, have far fewer barriers to entry that the plants selected for Fleming’s project. Analyzing data like light availability, water depths, photosynthetically active radiation, fetch, slope, and much more the researchers illustrated the modeling process and created a final product that showed specific areas at which the plant communities would have little to no issues in establishment and subsequent growth. Models such as these allow people to fully visualize the concepts before further action is taken whether that’s implementation or alteration. I decided that with GIS and specific data about nutrient uptakes, growth rates, and biomass to surface area ratios, among others, I located the study area to model a phytoremediation project. To further the usefulness of the model I also developed a tentative harvesting schedule with estimates on the total amount of nutrients removed and biomass produced using spatial data provided by the ArcGIS/Desktop software and values obtained in research. Literature Review & Study Area Due to excessive fertilizer misuse in the agriculture industry, the obsession over crisp green grass in first world countries, improper disposal of industrial chemicals, leaking sewage systems, and many other sources of pollution in our world today, chemicals enter the water cycle (Gnanavelrajah and Sundaraling 2014, 276). The different properties, processes and happenings of the water on our planet ensure that, unless properly used, these chemical pollutants end up contaminating surrounding waterbodies. As stated above nitrogen and phosphorus are found in excess levels in waterways all over the world (EPA 2015). Excess, in this case, is defined by the threshold amounts of eutrophication which are as follows: N > 2.5% of total and P > 500 mg kg-1 (Bukata et al. 2015, 307). Eutrophication is the overloading of nutrients, more specifically phosphorus and nitrogen, in any given body of water. The presence of excess nutrients causes the disruption of the naturally homeostatic aquatic environment by increasing the growth of the flora exposed 1 exponentially. In its early stages it isn’t a huge problem but when left unchecked and nutrient levels get above the aforementioned threshold levels, huge algal blooms occur (Phelps et al. 2006). Subsequently the overall water quality degrades – everything from dissolved oxygen, O2, turbidity, to pH levels change so much so that the contained marine life dies off until the entirety of the water source cannot support life and is then deemed a dead zone (Gnanavelrajah and Sundaralingam 2014; Han et al. 2013; Kanwal et al. 2011; Lu et al. 2010; Greening and Janicki 2006). Aside from the ecological negativities associated with eutrophic waters there are also numerous human health concerns surrounding the over consumption of nitrogen and phosphorus (Gnanavelrajah and Sundaralingam 2014; Hanwal et al. 2011). Scientists studied the effectiveness of a type of bioremediation (using biotic life to remediate a given environmental ailment) known as phytoremediation (using plants as the biotic agent). Originally studied for its practicality, inexpensive start-up, maintenance and removal costs phytoremediation can, without a doubt, be considered a successful tactic. All biological processes are enabled and fueled by the use of nutrients and various other chemicals/elements. Plants are no exception to this rule; chemicals are taken in by the plant through roots, leaves, etc… and either utilized, stored or transpired back into the environment – nitrogen and phosphorus are well known plant nutrients and both are continuously used by vegetation to support growth, reproduction, vitality and health. Numerous studies have shown phytoremediation, regardless of the plant used, to be an effective method for the removal of targeted chemicals – in this case N and P contributing to eutrophication (Lu et al. 2010; Gnanavelrajah and Sungaralingam 2014; Han et al. 2013). Pistia Stratiotes, here on after referred to as or Pistia, is one fairly controversial plant (Evans 2013) in Central Florida that has been extensively tested for its effectiveness in phytoremediation (Meier et al. 2014; Gnanavelrajah and Sungaralingam 2014; Han et al. 2013; Kanwal et al. 2011; Lu et al. 2010). These principals can be universally transferred to Eicchornia crassipe (or Eicchornia) given the uniformity of biological processes. The recently determined native status of Pistia and the prevalence of eutrophication as well as other water related issues in the Central Florida area (Bukuta et al. 2015; Kadyampakeni et al. 2015; Evans 2013; Lu 2010; Greening and Janicki 2006;), my focus for research has been on such. Everything from soil/water profiles to phytoremediation efforts has already been completed, studied, tested, and logged extremely well (Bukuta et al. 2015; Lu et al. 2010; Greening and Janicki 2006; Phelps et al. 2006) Florida as a whole is an ecologically sensitive area because of its easily percolated terrain (sands, karst rock, etc…), cultural industries/practices (golf courses, constant lawn maintance, agriculture, etc..), and it being home for countless endangered species and migratory animals add to the importance of finding better alternatives for the current inefficient and ecologically unsafe practices of not just the agricultural industry (Das et al. 2010) but the rest of the state. Providing scientific evidence to support the use of Pistia and Eicchornia in phytoremediation practices in combination with estimated outcomes could aid in more directed and numerous environmental restoration projects. The federal government deemed Pistia as well as Eicchornia a “noxious weed” meaning it has multiple negative environmental impacts in its given habitat – for example Pistia became so much of a problem in the Rio Grande River that, with the help of a few other noxious weeds, various infrastructures like irrigation channels and storm water drains were effectively over ran (Meier et al. 2014). To combat the aforementioned issue in Central Florida, current growth 2 management practices involve spraying herbicides and allowing the plant to decompose from then on in the waterbody it grew in. This inherently contributes to nutrient loading and is yet another source of pollution in the managed waterways. Pistia and Eicchornia’s high growth rate, native status, overwhelming presence in my study area, effectiveness at phytoremediation and virtually no physiological barriers for removal like deep in-ground rooting systems make it the ideal candidate for the continued use in phytoremediation efforts as well as the originally suggested compost medium already proven to generate a viable product elsewhere in the world (Gnanavelrajah and Sundaralingam 2014; Meier et al. 2014; Evans 2013; Han et al. 2013; Kanwal et al. 2011; Lu et al. 2010;). Using all of the Central Florida area would complicate the overall coherence of my GIS model, require ridiculous amounts of data and hardware capability for data processing/management and overall set the scale and scope a bit too large to be comprehensible and practical. To set a more realistic scale and bring down the data and hardware requirements I focused my study on the local waterbody Lake Beresford (Figure 1). Stetson University, the host of my research, owns a parcel of land next to the lake that would be used as an experimental base of operations. Using GIS data provided by Volusia County, I was able to locate more than 500 septic tanks within a mile of the lake; recent water profiles of Gemini Springs, a nearby spring shed, suggest that septic tanks pose a threat to local waterbodies. It is imperative that we consider alternatives to current plant management techniques, starting up new environmental restoration projects such as my proposal in this three-part model and focus on preserving our natural ecosystems for the generations to come. Methodology & Observations Overall analysis was based on the depth of the lake, averaged nutrient uptake and growth values of both Pistia and Hyacinth, as well as overall surface area of the selected sites. As a bench mark of current estimated nutrient removal, values were determined using current aquatic vegetation location information. Data (Table 1) were collected from the Volusia County Department of Financial and Administrative Services online geodatabases, the Florida Department of Transportation, Stetson University’s Dr. Kirsten Work, and University of South Florida and were geoprocessed using various ArcMap 10.2.2 operations. These data include a hydrologic shapefile, point data associated with septic tank locations, a wetland vegetation shapefile, marina location point data, a 2015 aerial photograph base map, LakeWatch data from 1991 to 2006 for Lake Beresford, and a digitized 1979 depth map of Lake Beresford. For the purpose of this analysis I assumed that both Pistia and Hyacinth face no barriers to high growth rates or establishment issues – especially in the Central Florida/Deland, Florida area around Lake Beresford. The average temperature for the selected “growing/harvest season” (growing/harvest season defined as March-July) range from 25 degrees Celsius to around 32 degrees Celsius (usclimatedata.com) which fall between the ideal 25 to 35 degrees Celsius growth temperatures of the two plants. In combination with the high nutrient levels present in the Lake Beresford area and the availability of sunlight in Florida during these months allow for Pistia and Hyacinth to double their population size in roughly two weeks (Akinbile and Yusoff 2012). It’s also shown that the nutrient uptake capability of both Pistia and Hyacinth decrease after 25 days of growth (Radu et al. 2015). To determine estimated N and P removal I used averaged third week nutrient uptake values observed during research on Pistia and Eichhornia’s effectiveness in waste water 3 treatment; nitrogen removal rate 0.15375 grams per litre per kilogram, phosphorus removal rate 0.32878 grams per litre per kilogram (Akinbile and Yusoff 2012). For comparative purposes I used three varying levels of plant surface coverage and converted the surface areas of each condition to an estimated biomass (87.35m2 = 1 kg) to which the respective estimated removal rates were applied and nutrient removal rates were generated (Table 2) (Armstrong et al 2003). Various plant cover maps can be seen in Figure 2. The workflow for preparing the data for analysis is explained below. I collected data from the various sources and reprojected them from their respective coordinate systems to NAD 1983 Albers to preserve the total area. Using the select by location function I selected only data within 1 mile of the Lake Beresford polygon which was extracted from the overall hydrology data file provided by Volusia County. The depth map was created using a 1979 raster file that I spatially referenced based on the boundaries provided by the hydrologic shape file and the outlines of the same hydrologic features depicted in the raster file. Points were then assigned to the specific depth markers and subsequently interpolated to produce a raster file with estimated depth gradients along the lake bed floor. A shape file was generated after a reclass function was completed to bring values slightly closer to the original raster file using the raster to polygon tool. Using the LakeWatch data provided by Kirsten Work I averaged out the most recent (01-06) five years of data to determine an approximate amount of total nitrogen and total phosphorus currently affecting the lake; respectively 1.178x10-3g/l and 8.127x10-5g/l. A current vegetation shapefile provided by Volusia County was used to calculate current nutrient uptake values as a bench mark for comparison of the alternative coverages suggested. To aid in cartographic appeal and to assist in further graphic representation aerial photographs and marinas were added to the model for illustrative purposes. Overall, the key in lake surface area selection was depth; I wanted to make sure the “suggested” model left a majority of the lake open for its intended purposes. Many local residents find the lake to be a relaxing and enjoyable park/public recreation area. Unless needed total use of the lake for phytoremediation and harvesting practices would take away from the natural beauty of the lake and also disrupt the overall ecosystem in ways we haven’t quite imagined. For the purposes of preserving the use and beauty of the lake only depths between 1 and 2 feet were selected for proposed values in the suggested plant cover model. Discussion and Conclusion I suggest that a cyclical three-week 50% harvest of the biomass occur to not only remove the maximum amount of nutrients possible but also allow for the population to grow to a significant size before harvested again as suggested by the 25-day nutrient efficiency rating and the roughly 2 week doubling time that Pistia and Eicchornia would possess given the local conditions. The nutrient uptake rates of Pistia and Eicchornia are much higher than the mean total nitrogen and phosphorus rates of Lake Beresford observed in the lake watch data, however, this does not invalidate the importance of my study nor does it suggest that Lake Beresford shouldn’t be host to a project such as what my model suggests. The hydrology of Florida is all interconnected and much of the surrounding ecology depends on a stable and healthy water body. Using Lake Beresford as a host site for various scales of phytoremediation efforts would not only clean the host Lake but also add cleaner, less nutrient rich water to the St. Johns River, the Floridan Aquifer, and other surrounding hydrologic features. 4 Tables & Figures Table 1. Data used in analysis, source/provider, data type and content and purpose in analysis Table 2. Plant cover/site selection and associated surface area (derived from ArcGIS), biomass and estimated removal rate for the total estimated biomass Source Feature Type Content Use Vol. Co. Dept. FAS Aquatic Plant Vegetation Vector - Polygon Various plant communities found within the county Locating existing Pistia and Hyacinth communities Vol. Co. Dept. FAS Septic Tanks Vector - Point Locations of septic tanks in the area Supplemental information Vol. Co. Dept. FAS Marinas Vector – Point Location of Marinas Supplemental information Vol. Co. Dept. FAS Hydrology Vector – Polygon Local hydrology with polygons representing local waterways Basemap Florida DOT 2015 Aerial Photos Raster Aerial Depiction of the area Basemap Dr. Kirsten Work – LakeWatch data provider Non-Georeferenced nutrient profile of the Lake Beresford area Excel file Ave total nitrogen & phosphorus levels (1991-2006) Estimated nutrient values University of South Florida/NOAA NOAA map of Lake Beresford area Raster Depths of lake Creation of georeferenced depth map Condition/Plant Cover Selection Surface Area (m2) Estimated biomass (kg) Phosphorus Removal (g/l) Nitrogen Removal (g/l) 1. Existing plant community (Current) 15,131.02 173.22 56.95 26.63 2. 1-2ft Depth (Suggested) 1,375,452.45 15,746.45 5,177.12 2,421.02 3. Entire Lake (Maximum) 3,340,141.59 38,238.60 12,572.09 5,879.18 5 Legend and north arrow apply to Figure 1 and 2 Figure 1. Study area 6 7 Figure 2. Various plant cover maps associated with numbers in table 2 Figure 2.1 Existing plant community (Current) Figure 2.2 1-2ft Depth (Suggested) 8 Figure 2.3 Entire Lake (Maximum) Works Cited Akinbile, C.O., and Yusoff, M. 2012. Assessing Water Hyacinth (Eichhornia crassopes) and lettuce (Pistia Stratiotes) effectiveness in aquaculture waterwater treatment. Internation Journal of Phytoremediation 14:201-211. Armstrong, N., Planas, D., Prepas, E. 2003. Portential for estimating macrophyte surface area from biomass. Aquatic Botany 74:173-179. Bukata, B., Osborne, T., Szafraniee, M. 2015. Soil nutrient assessment and characterization in a degraded central Florida swamp. Water Air Soil Pollut 226(307):1-11. Chattopadhyay, S., Fimmen, R., Yates, B., Lal, V., Randall, P. 2012. Phytoremediation of Mercury- and Methyl Mercury-Contaminated Sediments by Water Hyacinth. International Journal of Phytoremediation 14(2):141-161 Das, A., Baiswar, P., Patel, D.P., Munda, G.C., Ghosh, P.K., Ngachan, S.V., Panwar, A.S., Chandra, S. 2010. Compost quality prepared from locally available plant biomass and their effect on rice productivity under organic production system. Journal of Sustainable Agriculture 34(5):466-482. EPA. 2012 A citizen’s guide to phytoremediation primer. Office of Solid Waste and Emergency Response. EPA 542-F-12-016. EPA. Nitrogen and Phosphorus Pollution Data Access Tool. In United States Environmental Protection Agency [database online]. 2015 Available from http://www2.epa.gov/nutrient-policy-data/nitrogen-and-phosphorus-pollution-data-access-tool (last accessed 12 Nov 2015). EPA. Using Phytoremediation to Clean Up Sites. In United States Environmental Protection Agency [database online]. 2012 Available from http://www.epa.gov/superfund/accomp/news/phyto.htm (last accessed 15 Sept 2015). Evans, J. 2013. Pistia stratiotes L. in the Florida Peninsula: Biogeographic evidence and conservation implications of native tenure for an 'invasive' aquatic plant. Conservation and Society 11 (3): 233-246. Flemming, J., Madsen, J., and Dibble, E. 2012. Development of a GIS model to enhance macrophyte re-establishment projects. Applied Geography 32(2):629-635. Gnanavelrajah, N., and Sundaralingam, T. 2014. Phytoremediation Potential of Selected Plants for Nitrate and Phosphorus from Ground Water. International Journal of Phytoremediation 16:275-284. Greening, H., and A. Janicki. 2006. Toward reversal of eutrophic conditions in a subtropical estuary: water quality and seagrass response to nitrogen loading reductions in Tampa Bay, Florida, USA. Environmental Management 38:163-178. 9 Han, P., K. Vijayaraghavan, S. Reuben, E. S. Estrada, and U. M. Joshi. 2013. Reduction of nutrient contaminants into shallow eutrophic waters through vegetated treatment beds. Water Science & Technology 68:1280-1287. Kadyampakeni, D., Morgan, K., Nkedi-Kizza, P. & Kasozi, G. 2015. Nutrient management options for Florida citrus: A review of NPK application and analytical methods. Journal of Plant Nutrition 38(4):568-583. Kanwal, S., Iram, S., Khan, M., and Ahman, I. 2011. Aerobic composting of water lettuce for preparation of phosphorus enriched organic manure. African Journal of Microbiology Research 5(14):1784-1793. Meier, E., Waliczek, T., Abbott, M. 2014. Composting invasive plants in the Rio Grande River. Invasive Plant Science and Management 7:473-482. Lu, Q., He. Z., D. A. Graetz., P. J. Stoffella., and X. E. Yang. 2010. Phytoremediation to remove nutrients and improve eutrophic stormwaters using water lettuce (Pistia stratiotes L.). Environmental Science and Pollution Research 17:84-96. Phelps, G., Walsh, S., Gerwig, R, Tate, W., 2006, Characterization of the Hydrology, Water Chemistry, and Aquatic Communities of Selected Springs in the St. Johns River Water Management District, Florida, 2004: USGS Open-File Report 2006-1107, 51. Radu, V.M., Ionescu, P., Deak, G.Y. 2015. Decreasing nutrients concentration from aquatic environment using aquatic plants. Journal of Environmental Protection and Ecology 16(2):610-619. 10