Relationship Between Socioeconomic Status and Prevalence of Mosquito-Promoting Peri-Domestic Containers Among Yards in Volusia County Mosquito Control District, Florida

RELATIONSHIP BETWEEN SOCIOECONOMIC STATUS AND PREVALENCE OF MOSQUITO-PROMOTING PERI-DOMESTIC CONTAINERS AMONG YARDS IN VOLUSIA COUNTY MOSQUITO CONTROL DISTRICT, FLORIDA. A PAPER SUBMITTED FOR COMPLETION OF SENIOR RESEARCH FOR THE COLLEGE OF ARTS AND SCIENCES STETSON UNIVERSITY BY Emma Schaefer IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF BACHELOR OF SCIENCE ENVIRONMENTAL SCIENCE ADVISORS Dr. Jason M. Evans, Ph.D. Dr. James A. Abbott, Ph.D. MAY 2017 ii TABLE OF CONTENTS Contents TABLE OF CONTENTS ...................................................................................................ii List of Figures and Tables ............................................................................................... iii ACKNOWLEDGEMENTS ...............................................................................................iv ABSTRACT ..................................................................................................................... v INTRODUCTION ............................................................................................................. 1 LITERATURE REVIEW ................................................................................................... 2 STUDY AREA ................................................................................................................. 6 METHODS ...................................................................................................................... 8 OVERVIEW .................................................................................................................. 8 CHECKING DATA SETS ........................................................................................... 10 JOINING THE DATA SETS ....................................................................................... 11 QUANTIFYING CONTAINER TYPES ........................................................................ 11 NORMALIZING CONTAINER DATA ......................................................................... 12 PRESENTING INCOME BY ZIP CODE ..................................................................... 12 PERMANENT AND NON-PERMANENT CONTAINERS ........................................... 13 SPATIAL AUTOCORRELATION ............................................................................... 13 RESULTS ...................................................................................................................... 14 CONCLUSION .............................................................................................................. 19 WORKS CITED ............................................................................................................. 21 iii List of Figures and Tables Figure 1 ........................................................................................................................... 7 Table 1 ............................................................................................................................ 8 Figure 2 ......................................................................................................................... 16 Figure 3 ......................................................................................................................... 17 Figure 4 ......................................................................................................................... 18 iv ACKNOWLEDGEMENTS I would like to thank Volusia County Mosquito Control, especially Mr. Jesse Julien, Mr. Jim McNelly, and Dr. Hong, for educating me on their agency’s mission and for providing container data for this senior research project. I would also like to thank Dr. Tony Abbott and Dr. Jason Evans for their guidance in planning and executing my senior research. A special thanks to my fellow Environmental Science and Geography seniors for their peer reviews, comments, and assistance throughout this project. v ABSTRACT Vector-borne diseases are either protozoan or viral in nature and transmitted by a few of the over 300 species of mosquito (Anonymous 2014). These diseases include malaria, yellow fever, Zika virus, dengue hemorrhagic fever, and many others (Anonymous 2016). Mosquitoes are adopting more anthropophilic behaviors that lead them to establish colonies near human settlements, bringing vectors closer to human populations and increasing the risk of an outbreak (Manrique-Salide et al. 2015). Identification of the peri-domestic containers used most often by domestic mosquitoes in East Volusia County is essential in optimizing modern mosquito control practices and preventing a vector borne disease outbreak if some of the population are carriers. Using data collected by Volusia County Mosquito Control (VCMC), I identified the most prevalent peri-domestic containers in Volusia County Mosquito Control District (VCMCD) and attempted to discern any trends in container prevalence based upon socioeconomic status (SES). Container prevalence and trends were obtained by conducting statistical and spatial analyses in ArcGIS and Microsoft Excel. INTRODUCTION Identifying potential breeding grounds for mosquitoes in a community is an important step in developing a defense against the most recent global, vector borne outbreak – Zika. The Zika virus was first identified by researchers in the Zika Forest of Uganda in 1947 and thought to be nonthreatening, but its 2014 emergence in Brazil initiated a global response due to its adverse effects on fetuses of infected mothers (Sampathkumar and Sanchez 2016). The modes of transmission for the virus are through sexual intercourse, blood transfusion, and mosquito bites by Aedes mosquitoes (Sifferlin 2016). In the United States, the mosquito genera Aedes, Anopheles, and Culex serve as vectors of diseases, such as chikungunya, malaria, and West Nile fever (Anonymous 2016). I would like to identify any trends in the presence of peri-domestic containers (artificial containers that are found around a residence) in Volusia County Mosquito Control District (VCMCD) that accumulate small amounts of water, usually rain, ideal for the depositing of mosquito larvae. Previous studies reveal that mosquito pupa are frequently present in buckets, tires, plastic rubbish, pet dishes, utensils, and flowerpots found around the outside of the home (Manrique-Salide et al. 2015). Containers made of plastic, specifically the color black, have been shown to be model breeding grounds for Aedes mosquitoes; metal containers heat the water to unfavorable conditions and contain metal ions, which do not provide a stable environment for larvae (Vezzani and Schweigmann 2002). A study of a residential area in Mexico identified 40% of containers sampled as harboring immatures as disposable containers (Manrique-Salide et. al. 2015). 2 Climate change is influencing the breeding behaviors of domestic mosquitoes in tropical regions by shifting their breeding grounds to shaded, and sometimes underground, areas around the home (Chadee and Martinez 2016). Identification of the most prevalent containers and any trends in their presence in VCMCD could assist the agency’s response to the next vector-borne disease crisis or aid in a prevention program by identifying topics in need of community education. I planned on identifying any trends in container prevalence based upon socioeconomic status (SES) to see if areas of lower income are of higher concern in regards to mosquito-promoting containers. Education of residents, and their resulting preventative action, will aid in controlling mosquito populations in Volusia County and possibly prevent locally transmitted cases of vector borne diseases. I hypothesized that of the sampled homes, most of the peri-domestic containers would be located around areas of low SES. I also believed that the most prevalent containers would be a disposable or preventable container. LITERATURE REVIEW Bisset et. al. (2006) claim that mosquito borne diseases spread so rampantly due to weak public health infrastructure, international trade, travel, and increasing urban populations who have limited access to health services. Aedes, Albopictus, and Culex mosquitoes are domesticated mosquitoes that coexist with humans, especially in urban areas in tropical and subtropical regions, and are primary vectors for some of the most concerning vector borne diseases (Eisen et. al. 2013). Mosquitoes present in the United States have started to exhibit anthropophillic behaviors and establish colonies near 3 human settlements (Kraemer et. al. 2015). Historically, some species of Aedes mosquitoes have preferred natural areas, such as woods (Hornby et. al. 1994). In recent years, Aedes populations have been increasing in urban areas due to the high amounts of artificial containers humans leave outside their homes (Li et. al. 2014). Ae. aegypti has been present in the United States since colonization, with the first case of yellow fever being reported in the 1680s (Juliano et. al. 2004). Ae. albopictus is a recently introduced species to the United States, having been identified in the 1980’s. Maneerat and Daude (2016) identified three main factors that affect the magnitude of mosquito populations in urban areas: local temperature and humidity, the degree of urbanization, and social utilizations of the area. All mosquito species lay their eggs differently. Ae. aegypti females lay their eggs in clusters and in several different containers filled with water; she never deposits all of her eggs into one container (Juliano et. al. 2004). Ae. albopictus females lay their eggs in one cluster and along the edge of a dry container. The eggs begin maturation once there is rainfall and the container fills up. Culex mosquitoes lay their eggs in “rafts” in pools of stagnant water. The warmer the water the eggs are maturing in, the faster the immatures develop (Kumar et. al. 2006). This is reinforced by the discovery that mosquitoes are attracted to the color black. Water kept in containers is usually cooler than the surrounding temperature, so a black container will house warmer water. Mosquitoes are more likely to lay eggs in containers exposed to direct sunlight in comparison to those located in the shade (Vezzani and Albicocco 2009). Although increased temperature promotes larval development of mosquitoes, increased humidity stunts development and decreases resulting longevity (Reiskind and Lounibos 2009). 4 In regards to the material containers are made from, a study conducted at a cemetery in Buenos Aires, Argentina, revealed that Aedes mosquitoes preferred plastic containers over ceramic, metal, and glass containers (Vezanni and Sweigmann 2002). The colors of the plastic containers preferred were usually a dark color or black. Vezzani and Schweigmenn proposed that mosquitoes utilized metal containers least due to the ions, primarily copper, present in the water contained in vessels. Another study conducted in Merida, Mexico, concluded that 40% of containers that tested positive for Aedes immatures were disposable (Manrique-Saide et. al. 2015). Ae. aegypti and Ae. albopictus are day biting species of mosquito that act as primary vectors of arboviral diseases (Bisset et. al. 2006). Historically Aedes mosquitoes, specifically Ae. aegypti and Ae. albopictus, have been the vectors of dengue hemorrhagic fever, yellow fever, and the chikungunya virus (Eisen and Moore 2013). Culex mosquitoes serve as the vectors of Japanese encephalitis, Lymphatic filariasis, and West Nile fever (Anonymous 2016). Female Aedes mosquitoes have now been shown to be the primary vectors for the Zika virus. Zika can also be transmitted through sexual intercourse, blood transfusions, and from mother to child (Sifferlin 2016). A global alarm was sounded when the incidence of microcephaly in Brazilian newborns in 2014 began to rise and was attributed to the Zika virus (Sampathkumar and Sanchez 2016). The virus was first identified on accident by a researcher in the Zika Forest of Uganda during a study on yellow fever in rhesus monkeys (Sampathkumar and Sanchez 2016). It was initially thought to only affect primates. We have now acknowledged its pathogenicity in humans and the adverse effects it imposes on 5 developing fetuses – including microcephaly and Guillian-Barre syndrome (Roodbol et. al.2014). A Zika infection is in no way fatal; not all who become infected with the virus from an Ae. aegypti or Ae. albopictus bite exhibit symptoms. Those that do experience symptoms from a Zika infection usually experience fever, rash, joint pain, and/or conjunctivitis. The infection becomes more serious if the infected individual is pregnant. The virus can affect fetal development, depending on viral introduction and the stage of pregnancy. Microcephaly itself is not fatal, but the developmental delays and brain damage that result can lead to conditions that may be life threatening (Lucchese and Kanduc 2016). A study conducted by researchers in Thailand on the biting patterns of Ae. aegypti revealed that mosquitoes more frequently feed on individuals over 25 years of age – women of child bearing age fall within this group (Harrington et. al. 2014). The World Health Organization (WHO) estimated that there would be approximately 3-4 million new Zika infections during the year 2016 (Malkki 2016). The first locally transmitted case of Zika in Florida was reported in July of 2016, so we can no longer enact primarily preventative measures. Mosquito control districts and health departments within the state of Florida need to educate communities, work to identify areas of concern, actively spray, and provide their residents with options to prevent contracting Zika and any other vecor borne disease. It is important that we minimize the risk of pregnant women contracting Zika by learning about and controlling Ae. aegypti and Ae. albopictus populations, along with other domestic mosquito species. As of May 2, 2017, only two cases of locally transmitted Zika have been reported to the Florida Department of Health (Anonymous 2017). Globalization and climate change have made the task of controlling mosquito populations more difficult. 6 By identifying trends in mosquito colonization and the environments provided to them, specifically peri-domestic containers, VCMC can create an individually tailored plan to minimize vector borne disease transmission and control mosquito populations in Volusia County. STUDY AREA This study was conducted in VCMCD, seen in Figure 1. VCMC established the Domestic Inspection Program in 2014. Only data from 2016 was utilized in this study. This program was designed to have inspections of 100 homes per zip code conducted, with investigators specifically inventorying mosquito-promoting containers located around a residence. These investigations were conducted by parcel number. Investigators were instructed to inventory the type and number of containers observed at each parcel, and to enter it into their monitoring system. Containers identified by VCMC during inspections were classified as barrels, bird baths, boats, bromeliads, buckets, cans, cinder blocks, fountains, garbage can, garbage can and lid, garbage can lid, gutters, other, plant containers, pools, tarps, tires, tree holes, urn, vehicle, or yard ornaments. Beyond that, containers were classified by materials: cement, ceramic, metal, other, plant/tree, plastic, or rubber. I did not find it necessary to reclassify the container data from what was provided to me by VCMC, based upon previous research and my absence during the inspections. A study conducted in Merida, Mexico, utilized a container classification system like that of VCMC (Garcia-Rejon 2011). Table 1 indicates how many sites per zip code were inspected as part of VCMCD’s Domestic Inspection Program. The number of sites inspected per zip code ranged from 1 parcel to 7 190 parcels. To account for this variability, zip codes with less than 15 sites sampled were excluded from the spatial autocorrelation (Moran’s I) analysis. Figure 1. Volusia County Mosquito Control District’s territory by zip code median income. 8 Zip Code # of Homes Sampled 32114 4* 32117 35 32118 92 32119 12* 32124 1* 32127 190 32128 5* 32129 23 32130 1* 32132 103 32141 34 32168 26 32169 28 32174 91 32176 29 32720 37 32725 74 32738 41 32744 1* 32759 11* 32163 33 32764 1* Table 1. Number of parcels sampled by VCMC investigators per zip code in 2016. *Zip codes with less than 15 sites sampled were excluded from spatial autocorrelation and correlation analyses. They were represented in the dot-density maps. METHODS OVERVIEW The data for my study had already been collected by the investigators of VCMC during all of 2016. VCMC investigators enter their observations into an internal GIS interface that keeps record of site, inspection, and container data collected in the field. 9 VCMC provided me with three Microsoft Excel spreadsheets with data pulled from this internal GIS system. The spreadsheets provided and their attributes included:  “Container_Table”: Link to Inspection Table, Container Type, Pool Detail, # of Containers this Type, Container Material, Container Avg Size, Date, Avg Water Volume, # of Containers with Larva, # of Containers with Pupa  “Inspections_Table”: Link to Container Data Table, Link to Sites Table, Inspection Date, Condition, Status, Action, Larval Sample Taken, Adult Sample Taken, Total Count, ID by Genus, Instar, Average larva per dip, # of Containers, # of Container with Larvae, # of Container with Pupa, # of Larval Samples, # of Adult Samples, Premises Inspected, Follow up, Adult on Wing  “Sites_Table”: Link to Inspection Table, Site ID, SS_Postal, Site Creation Date, Site Acres I investigated the prevalence of peri-domestic containers around homes in VCMCD with the data provided by VCMC using ArcGIS and Microsoft Excel. VCMC had collected data from each zip code throughout VCMCD, so it should be representative of all homes in the study area. A potential issue could be that several zip codes (32114, 32119, 32124, 32128, 32744, 32759, 32764) include less than 14 parcels sampled. To counter this issue, only zip codes with at least 15 sites sampled were included in the spatial autocorrelation analyses. Containers were then classified by myself as permanent or non-permanent containers, based upon the ability of the homeowner to inhibit water from being retained. The amounts of permanent and non-permanent containers per zip code were then represented in a dot-density map over a gradient map of zip code median income. Geospatial autocorrelation (Moran’s I) analysis was 10 conducted to identify any statistically significant clustering of containers, followed by a series of correlations between zip code median income and the number of each container type per zip code. A regression analysis was intended to trace back any statistically significant clustering to a specific zip code or group of zip codes. CHECKING DATA SETS Before joining the data sets, consistency of the data was tried through a series of sorting tests. The point of this was to verify all links from the sites table were unique and there was a repetition of links in the inspections and container tables. To do this, order data in the “Sites_Table” from smallest to largest value for the “Link to Inspection Table” attribute. Create a new column to the right of the “Link to Inspection Table” attribute. Title the new column “Test”. In the first cell in the “Test” column, enter “= (click on first ‘Link to Inspection Table’ feature) – (click on ‘Link to Inspection Table’ feature immediately below the first)”. Drag the formula down so it can be applied to all “Link to Inspection” features. Highlight the whole “Test” column and copy all cells with answers to the previously mentioned formula. Paste the answers back into the “Test” column as values so that the numbers do not change once the table is resorted. Highlight the attribute table and sort by “Test” in ascending order. If the values for “Test” are anything but zero, all links are unique and there is no repetition of sites. Repeat this test for the “Inspections_Table” by testing the values in “Link to Container Table “ and “Link to Sites Table”. To test the consistency of the data in the “Container_Table” perform the same test on values in the ”Link to Inspection Table”. Any zeroes appearing during the “Inspections_Table” test indicate more than one 11 inspection performed at a single site. Any zeroes appearing during the “Container_Table” test indicate more than one type of container was found during a single inspection. JOINING THE DATA SETS To join the “Sites_Table”, “Container_Table”, and “Inspections_Table”, two joins must be made. In ArcMap, the “Inspections_Table” is made into an attribute table and through a many-to-one join, the “Link to Inspection Table”, “SS_Postal”, and “Site Acres” attributes based upon the tables’ “Link to Inspection Table” and “Link to Sites Table” attributes. Any sites not joined to inspection data is are discarded from the attribute table. The second join will be a summarized joint of “# of Containers this Type” to the data set using the “Link to Inspection Table” and “Link to Container Table” attributes. This series of joins will indicate the amountnumber of containers per investigation for every site. The new table is called the “linked table”.Cumulatively, this could be used to quantify the amount of containers found in each zip code through VCMC inspections. An index can be generated by dividing the amount of containers found. QUANTIFYING CONTAINER TYPES By right-clicking on the “ZCODE” (zip code) column and selecting “Summarize,” the sum of each container type can be quantified per zip code. The “Summarize” function yields an output table for each container. Each output table can be exported as 12 an Excel Workbook and condensed onto a single spreadsheet of sums. Each column title was styled like “sum_(container name)”. NORMALIZING CONTAINER DATA Utilizing the “Summarize” function, total acres sampled for each zip code can be obtained. This results table can then be exported and added into the Excel Spreadsheet of sums of each container type. Using the areas sampled (acres) per zip code, containers counted per zip code can be normalized by the total acres sampled. This normalization is done for each container type by zip code. PRESENTING INCOME BY ZIP CODE Before making a map, ensure that the projection is appropriate for the data. This geodatabase was projected into NAD 1983 GDL FL Albers. A map of Volusia County by zip code was adopted as our base map. Using the United States Census Bureau spreadsheet of national zip codes (2010) and their relative median incomes, mean incomes, and population, pull the zip codes from Volusia County and save them as a CSV file. Join the data to the joined table by zip code. Under Symbology, create a choropleth map by selecting “Quartile”. This creates a four-step color gradient for median income. Use the same color but different shades, with the color darkening as income increases, as the indicators for median income by zip code. The median incomes will be used as indicators of SES. 13 PERMANENT AND NON-PERMANENT CONTAINERS Use the choropleth map of median incomes of Volusia County as a base to the dot density representation. To visually explain container distribution, containers are presented in dot density maps based upon the number of permanent or non-permanent containers observed per total acres sampled in each zip code. Permanent containers were those that cannot have their primary function altered to not include the accumulation of water. Permanent containers of those sampled were decided to be birdbaths, boats, bromeliads, fountains, gutters, pools, tree holes, and vehicles. Non-permanent containers were determined to be barrels, buckets, cans, cinderblocks garbage bins, garbage bin lids, garbage bins with lids, plant containers, tarps, tires, urns, and yard ornaments. Under the properties of the joined table in ArcGIS and on the Symbology tab, select all the permanent containers normalized by acres sampled to transfer to the attribute field. Make all the field symbols uniform (one color) and set the dot value to 1. For zip codes that had no data collected during VCMC sampling, create a new layer by selecting zip codes from the attribute table. These areas are presented as gray and identified as “No Data” in the legend. Repeat this process with non-permanent containers. SPATIAL AUTOCORRELATION Export the joined table as a .dbf file and create a column in Excel that shows the sums of total, permanent, and non-permanent containers per zip code. Normalize the sums by the total acres sampled per zip code and save these as a CSV file. Join the 14 two new columns to the existing joined table in ArcGIS based on zip code. In the table, select all zip codes with greater than 15 sites sampled and create a new layer file from the selected zip codes. Once the data is joined and the new layer file created, open toolbox  spatial statistics tools  analyzing patterns  spatial autocorrelation (Moran’s I). Select the layer file of zip codes with greater than 15 sites sampled as your input feature class. Select the column indicating the total containers observed normalized by the acres sampled as the input field. Check the box that says “Generate Report (optional)” to save a copy of the results. Select inverse distance for the conceptualization of spatial relationships, Euclidian distance for distance method, and none for standardization. Run the tool. Repeat test for permanent containers per acres sampled and non-permanent containers per acres sampled. Moran’s I Geospatial Autocorrelation is used to identify statistically significant clustering or dispersal of points of interest in a study area. RESULTS Figure 2 and Figure 3 provide visual representations of container prevalence in VCMCD, displayed over a gradient of the median income of each zip code. Looking at Figure 2, mosquito-promoting containers are present throughout the sampled zip codes in Volusia County, but there are greater densities of non-permanent peri-domestic containers in the northeastern and southwestern parts of the county. It also appears as if the higher density zones are in or near areas of low median income. This interpretation can also be made when looking at Figure 3. While the number of mosquito-promoting containers per acre sampled appear to be fewer, based upon the 15 overall density of dots, there are more dots centered around the northeastern and southwestern zip codes of the county. 16 Figure 2. Dot density distribution of non-permanent peri-domestic containers inventoried by Volusia County Mosquito Control District as a part of the 2016 Domestic Inspection Program. Zip code median income for Volusia County was provided by the United States Census Bureau (2010). 17 Figure 3. Dot density distribution of permanent peri-domestic containers inventoried by Volusia County Mosquito Control District as a part of the 2016 Domestic Inspection Program. Zip code median income for Volusia County was provided by the United States Census Bureau (2010). Moran’s I was run for total containers observed per acres sampled, non-permanent containers per acres sampled, and permanent containers per acres sampled. No significant clustering of observations was found in the analysis of total containers observed or permanent containers observed. The Moran’s I analysis did identify a clustering of observed non-permanent containers per acres sampled within VCMCD. The p-value (significance level) that the test yielded was 0.040705, which is less than 0.05, and indicative of a significant clustering of observations in the subject area. As seen in Figure 4, the p-value does indicate a relationship, but not one as strong as a relationship with a significance value of 0.01. There is a less than 5% chance that this clustering of observations is due to random chance, as expressed by the z-score of 2.0465. 18 Figure 4. Output generated by ArcMap for the Moran’s I function run on the dispersal of non-permanent peri-domestic containers in Volusia County Mosquito Control District. Clustering of containers was indicated by a p-value of 0.0407. A regression analysis was not performed on this dataset, but could be performed on a dataset that is not clustered in regions as large as zip codes. This analysis is better suited for a data set that is georeferenced to census tracts, census blocks, or coordinate data. Additionally, correlations were run between the sums of each container type per zip code and zip code median income. Some relationships came back as statistically significant, but these were for container types with low counts, so it could be due to sampling error. Also, many correlations were run, so a chance correlation was possible. 19 CONCLUSION Summary s My hypothesis that containers would be clustered around areas of low SES was not supported by this analysis. Since container data was already clustered by zip code, running a regression could yield a false result, indicating a clustering of containers around an already clustered data set. Additionally, zip codes contain a variety of incomes within their vast area, so the median incomes provided by the United States Census Bureau (2010) may not have been representative. Because of the container data’s complexity, a full analysis was not possible and a central point of clustering could not be identified. Also, the distribution of median income of zip codes was not normal, indicating that this data was not ideal for regression analysis. Despite the inability to trace back a clustering of peri-domestic containers, a clustering of non-permanent containers in VCMCD was identified using the Moran’s I function in ArcGIS. Of all sites sampled, 13,366 non-permanent containers and 3,836 permanent containers were observed. Around 78% of containers inventoried by VCMC as part of the Domestic Inspection Program were non-permanent containers; these containers could be either thrown away or manipulated to inhibit their retention of stagnant water. This corresponds with the pattern of peri-domestic containers found by Manrique-Saide et al. (2015), who found that over 40% of mosquito promoting containers in their study area were were disposable. With a more geographically specific data set, these methods could indicate a potential relationship between socioeconomic status and the prevalence of mosquito promoting containers in an observed area. A study utilizing these methods could be conducted using container data georeferenced to census tracts, census blocks, or 20 coordinates. To yield more specific results, I would advise VCMC to record the location where the container was found (under shade or in direct sunlight), if the container was pupae or larvae positive, and the color of the container. An analysis of container prevalence in areas of varying SES in Volusia County, coupled data indicating the areas where containers are most productive, could lead to more informed actions executed by VCMC and personalized preventative measures.IDENTIFYING MOST PREVALENT CONTAINERS Using the “Container_Table” excel spreadsheet provided by VCMC, container types and quantities found at each inspection will be organized by type and the amount of each container quantified for the whole VCMC district. Containers types will be ranked by in descending order by the amount of containers found in the year 2016. For each container type, another summarized join will be executed in order to create contribute to a sum of containers of a specific type for each county. An index will be generated by dividing the number of each type of container per zip code by the total number of sites in the zip code. The median household income for each zip code is available in the VC zip code layer file provided by VCMC and can be used to differentiate zip codes of high and low SES. Zip code areas of high SES will be represented by one color and low SES areas by another in a choropleth representation of the county. On top of this SES map, a dot density can be generated for total containers found in each zip code divided by the amount of sites, and other dot density maps can be generated for each container type divided by the number of sites in each zip code. 21 IDENTIFYING THE MOST PRODUCTIVE AREAS A binary of the “Status” column in the “Inspections_Table” can be used to identify container productivity. The “Status” column contains either “Larvae present” or “Larvae NOT present”. “Larvae present” is indicated by “1” and “Larvae NOT present” is indicated by “0”. The binary system can be joined to the existing dataset in ArcMap and a visual representation of container productivity can be represented by the amount of larvae positive containers, divided by sites, in a choropleth map. The three classes of productivity are distributed by natural breaks into “low larval productivity”, “normal larval productivity”, and “high larval productivity”. It has been observed that some domestic species do not venture far from where they originate and their biting patterns indicate that they feed on more than one host at a time (Harrington et. al. 2014). During a study conducted by Harrington et. al. (2014) it was estimated that 43-46% of mosquitoes feed on more than one host, with transients being bitten along with residents of the area. Identifying zip code areas of high productivity could identify potential public health concerns if a vector borne disease were to break out. WORKS CITED Anonymous. 2014. Mosquito-Borne Diseases. American Mosquito Control Association. Accessed at: http://www.mosquito.org/mosquito-borne-diseases Anonymous. 2016. Vector-borne diseases. World Health Organization. Accessed at: http://www.who.int/mediacentre/factsheets/fs387/en/ Anonymous. 2017. Zika Virus. Florida Department of Health. Accessed at: http://www.floridahealth.gov/diseases-and-conditions/zika-virus/ 22 Bisset, J. A., Marquetti, M. C. Marquetti, S. Suarez, M. M. Rodriguez, H. Padmanabha. 2006. Application of the pupal/demographic-survey methodology in the area of Havana, Cuba, with low densities of Aedes aegypti (L.). Annals of Tropical Medicine and Parasitology 100:S45-S51. Chadee, D. D. and R. Martinez. 2016. Aedes aegypti (L.) in Latin America and the Caribbean region: With growing evidence for vector adaptation to climate change? Acta Tropica 156:137-143. Eisen, L. and C. G. Moore. 2013. Aedes (Stegomyia) aegypti in the Continental United States: A Vector at the Cool Margin of Its Geographical Range. Entomological Society of America 50:468-478. Garcia-Rejon, J. E., M. P. Lopez-Uribe, M. A. Lorono-Pino, J. A. Farfan-Ale, M. R. Najera-Vazquez, S. Lozano-Fuentes, B. J. Beaty, and L. Eisen. 2011. Productive Container Types for Aedes aegypti Immatures in Merida, Mexico. Entomological Society of America 48.3:644-650. Harrington, L. C., A. Gleisher, D. Ruiz-Moreno, F. Vermeylen, C. V. Wa, R. L. Pouson, J. D. Edman, J. M. Clark, J. W. Jones, S. Kitthawee, T. W. Scott. 2014. Heterogeneous Feeding Patterns of the Dengue Vector, Aedes aegypti, on Individual Human Hosts in Rural Thailand. FLOS Neglected Tropical Diseases 8:1-16. Hornby, J. A., D. E. Moore, and T. W. Miller. 1994. Aedes albopictus distribution, abundance, and colonization in Lee County, Florida, and its effect on Aedes aegypti. Journal of the American Mosquito Control Association 10:397-402. Juliano, S. A., L. P. Lounibos, and G. F. O’Meara. 2004. A field test for competitive effects of Aedes albopictus on A. aegypti in South Florida: differences between sites of coexistence and exclusion? Oecologia 139:583-593. Kraemer, M. U. G., M. E. Sinka, A. K. Duda, A. Q. N. Myline, F. M. Shearer, C. M. Barker, C. G. Moore, R. G. Carvalho, G. E. Coelho, W. V. Bortel, G. Hendrickx, F. Schaffner, I. R. F. Elyazar, H. Teng, O. J. Beady, J. P. Messina, D. M. Pigott, T. W. Scott, D. L. Smith, G. R. W. Wint, N. Golding, S. I. Hay. 2015. The global distribution of the arbovirus vectors Aedes aegypti and Ae. albopictus. eLife 1-18. Kumar, G., R. Singh, V. Pande, R. Dhiman. 2006. Impact of container material on the development of Aedes aegypti larvae at different temperatures. Journal of Vector Borne Diseases 53:144. Li, Y., F. Kamara, and Zhou, G., Puthiyakunnon, S., Li, C., Liu, Y.,Zhou, Y., Yao, L., Yan, G., Chen, X. 2014. Urbanization Increases Aedes albopictus Larval 23 Habitats and Accelerates Mosquito Development and Survivorship. PLOS Neglected Tropical Diseases 8:1-12. Lucchese, G. and D. Kanduc. 2016. Zika virus and autoimmunity: From microcephaly to Guillain-Barre syndrome, and beyond. Autoimmunity Reviews 15:801-808. Malkki, H. 2016. Zika virus infection could trigger Guillain-Barre syndrome. Nature Reviews Neurology. Maneerat, S. and E. Daude. 2016. A spatial agent-based simulation model of the dengue vector Aedes aegypti to explore its population dynamics in urban areas. Ecological Modeling 333:66-78. Manrique-Saide, P., C. R. Davies, P. G. Coleman, E. Rebollar-Tellez, A. Che-Medoza, F. Dzul-Manzanilla, and A. Zapata-Peniche. 2015. Pupal Surveys for Aedes aegypti Surveillance and Potential Targeted Control in Residential Areas of Merida, Mexico. Journal of the American Mosquito Control Association 24:289-298. Reiskind, M. H. and L. P. Lounibos. 2013. Spatial and temporal patterns of abundance of Aedes aegypti and Aedes albopictus in southern Florida. Medical and Veterinary Entomology 27:421-429. Roodbol, J., M. C. Y. de Wit, F. K. Aarsen, C. E. Catsman-Verrevoets, B. C. Jacobs. 2014. Long-term outcome of Guillain-Barre syndrome in children. Journal of the Peripheral Nervous System 19:21-126. Sampathkumar, P. and J. L. Sanchez. 2016. Zika Virus in the Americas: A Review for Clinicians. Concise Review for Clinicians:514-521. Sifferlin, A. 2016. What You Need to Know About Zika. Time 187:32-38. Elaborates on modes of transmission of Zika. Vezzani, D. and P. Albicocco. 2009. The effect of shade on the container index and pupal productivity of the mosquitoes Aedes aegypti and Culex pipiens breeding in artificial containers. Medical and Veterinary Entomology 23:78-84. ***Will be adding the zip code layer file to works cited once provided by VCMC