Recent advances in image recognition have been catalyzed by progressions in the applications of convolutional neural networks (CNN) and deep learning (DL). In traditional artificial intelligence (AI), neural networks (NN) were represented in a “shallow” fashion; dictating only one dimensional vectors at various layers. Furthermore past networks were often confined to three main layers: input, hidden and output layers. This rigidity of the structure not only contrasted the NN’s derivation from complicated biological neural systems but also limited their capability of categorizing inputs of various sizes and orientations (like images.) CNN's sought to solve this problem by representing a NN in terms of 3D volumes in which a kernel is moved in a sliding manner over subsections of an input volume and convolved with these regions to generate a k-layer output volume. This output volume is comprised of filtered versions of the previous volume which help detect recognizable features while maintaining important spatial features. This project created a deep CNN which leverages the Java library DeepLearning4j to demonstrate these techniques and provide a simple program which attempts to categorize input images into one of 5 classes.
There are two types of hearing loss, conductive and sensorineural. The former simply reduce the sound level as it passes through the external ear canal to the middle ear, so amplifying comes close to restore hearing to normal. On the other hand, sensorineural hearing impairment results from a defect in the inner ear or the cochlea nerve. Most of the time, this condition cannot be medically or surgically corrected. This is also the most common form of hearing lost and amplification of sound alone is ineffective. However, we can combine many other techniques to manipulate sound to treat this condition. Thus, we create Distinct Sound to help patients with sensorineural hearing loss.
Author(s): Kathryn Jaslikowski Computer Science Nicholas Bomm Computer Science Phil Howell Computer Science Wills Ward Computer Science
Advisor(s): Donnell Payne Computer Science Samantha Powell Nutritional Sciences
Location: Session: 2; 3rd Floor; Table Number: 4
The purpose of this capstone project is to aid Meals On Wheels, Inc. (MOWI) of Tarrant County with in-home consultations for low-income and low-mobility clients. MOWI can only provide a client with 10 meals per week. A professor in the TCU Nutritional Sciences Department approached our group about creating a web application to provide easy-to-make, low-cost recipes in order to supplement the meals delivered. We then developed an application - easybites.org - that allows dietitians to create recipe and shopping lists based on client food preferences, allergies, and appliance/utensil restrictions. Dietitians can then print a PDF file of the recipes and shopping lists in-home for clients to keep. Foods, recipes, and stores can be dynamically added, edited, and deleted from the database by administrators and interns. We also calculate the nutritional information for each recipe using a USDA Nutrient Database API to ensure that the MOWI clients are able to see the nutrition content of the recipes.
Author(s): Rebecca Ruch Computer Science Cameron Diou Computer Science Harrison Engel Computer Science Steven Garcia Computer Science Will Taylor Computer Science
Advisor(s): Donnell Payne Computer Science Lisa Ball Computer Science
Location: Session: 2; B0; Table Number: 3
Expanding Your Horizons Network (EYHN) is a 501(c)3 nonprofit organization dedicated to providing gateway STEM (Science, Technology, Engineering, and Math) experiences to middle and high school girls that spark interest in activities and careers within these fields. EYHN accomplishes this through role-model led conferences with hands on STEM activities and workshops.
These conferences are hosted by various organizations across the country. In Fort Worth, an annual EYHN conference is hosted by Texas Wesleyan University (TxWes). Each year, this conference hosts hundreds of student participants and requires dozens of leaders, volunteers, and presenters. Handling a conference of this size requires significant organizational effort, with a bulk of pre-conference administrative work going to registering participants and creating a good schedule for the event. In previous years, organizers at TxWes used a scheduling and registration system created by TCU students in 2005. However, this program is out of date and no longer useable making a replacement necessary.
This Project, Scheduling Your Horizons (SYH), creates a replacement system for TxWes that allows TxWes organizers to register participants and generate a schedule for the conference. It does so in a modern, user-friendly manner, with an emphasis on platform independence and maintainability to extend the lifespan of the application.
Author(s): James Stewart Computer Science Michael Giba Computer Science Quang Nguyen Computer Science Son Nguyen Computer Science Thaddeus Rix Computer Science
Advisor(s): Liran Ma Computer Science Billy Farmer Computer Science Donnell Payne Computer Science
Location: Session: 1; 2nd Floor; Table Number: 8
TCU’s previous Student Research Symposium site provided an outdated submission-review system for the Michael and Sally McCracken Student Research Symposium, an event growing in popularity. The old system was mostly a front-end to a primarily manual collection of procedures to collect, review, and present research projects. There was a growing need to make a more robust system that can provide smart interfaces for various users that allows for secure submitting, balloting, and administration.
The new system provides a host of automated processes that facilitate in the management of the SRS event from year to year, including such things as automatic archiving of previous year’s information. This is possible due to a myriad of free technologies such as Django. To complement the many processes we have automated, we have created tools for administrators to change information in the website without entering the codebase. Among the automated processes and features that help with administration, we have embedded advanced algorithms which reduce the need for human involvement, such as cost-analysis table assignments, a procedure that once required hours of laborious calculations.
In this experiment, we examine the non-linear dynamics of a mechanical system consisting of an inverted pendulum with one free-turning rotational degree-of-freedom attached to a computer-controlled cart with one linear degree-of-freedom. Using a Quanser Linear Servo Base Unit with Inverted Pendulum and paired software package, we used first principles to develop the non-linear control system needed to move the pendulum from stable equilibrium to unstable equilibrium and maintain unstable equilibrium. This combines the self-erecting inverted pendulum experiment and the classic pendulum experiment. Through the paired software package, we were able to derive the dynamic equations to develop the transfer function and proportional-velocity feedback system that describe the linear motion of the cart, successfully creating the non-linear control system for both phases of the experiment.
This report examines the function, accuracy, and ease of use of an XBOX Kinect™ as a 3D surface scanner. The purpose of this experiment is to demonstrate the utility of a Kinect™ for XBOX 360 (Microsoft®) paired with Skanect (Occipital) and MeshLab software packages as a low cost solution to surface scanning and processing. My conclusion is that the Kinect™ is able to accurately model the recorded point cloud as a continuous 3D surface that matches the contour and scale of the test subject surface. Both Skanect and MeshLab effectively interpolated the smoothing of the 3D surfaces and provided higher resolution imaging than an unaltered image. The resultant resolution of the contoured surface is higher than the resolution of the 3D printers used in this experiment, demonstrating an effective digital duplication of a physical surface.
Author(s): Lauren Getz Engineering Robin Livesay Engineering Nathan Loewen Engineering Karla Lopez Engineering
Advisor(s): Robert Bittle Engineering
Location: Session: 2; 3rd Floor; Table Number: 4
For this project, a digital grip gauge was designed for Lockheed Martin to measure the grip length of the aircraft skin of the F-35. The objective of the electrical group is to ensure that the gauge will be capable of recognizing when the measurement has stabilized. When stabilized, a light will turn on, which allows the operator to know the measurement is ready for reading. We developed three prototypes to complete this objective. The three prototypes utilize Arduino, comparators, or push buttons. While each of these prototypes satisfies the objective, the third prototype was ultimately selected due to size constraints of the gauge design.
Compressive line sensing is a process of acquiring data and reconstructing images. The objective of this study is to explore the impact of the two parameters that are used in the image reconstruction algorithm on the quality of the reconstructed image. These two parameters are the compression ratio and the line group. The compression ratio is the ratio of the number of measurements taken at each line vs. the resolution of each line. The line group is the number of lines that are grouped together and solved jointly when reconstructing the image. A higher compression ratio results in degraded image quality because less measurement data is used to reconstruct the image. The larger the line group, the better the quality of the image at a cost of longer computation time. The key is to find a balance between the compression ratio and line group choices so that the image is reconstructed with as little data as possible while still maintaining a high image quality. We will present images reconstructed with different compression ratio and line group based on the data obtained in air and in water.
The highest electric power demand in the lower temperate and equatorial zones is encountered during summer months, when air-conditioning systems are invariably and almost continuously used in commercial and residential buildings. During the summer months the electric power corporations utilize most of their generating capacity, including smaller units that are older and significantly less efficient. Because high ambient temperatures, the need for air conditioning, and the available solar energy are highly correlated, solar energy is a suitable energy source to supplement and, eventually, replace the fossil fuel power plants to satisfy the burgeoning air-conditioning demand in the summer months. Since solar energy is periodically variable, a zero-energy building must also include a reliable system for energy storage. This paper presents a detailed hourly analysis of the power needs, the seasonal energy usage and the seasonal energy storage requirements of a zero-energy building in the South-West part of the USA, where air-conditioning demand is significant. The input data for the calculations include the hourly electric power demand during a typical year. The hot water and space heating needs of the building are satisfied by a heat pump that uses the electric power produced by photovoltaics. Two energy storage systems are considered, hydrogen storage and solid-state batteries. The paper reports the actual hourly electric power and energy demand of the building throughout the year; the hourly energy production by a system of photovoltaics; the hourly energy storage needs throughout the year; the overall photovoltaics area requirements; the overall capacity and seasonal use of the energy storage system needed; and the effects of the various component and systems performance/efficiencies on the overall area of photovoltaics needed and energy storage requirements for the building to become grid independent.
Author(s): Evan Schmitzberger Engineering Steven Culver Engineering Avery McGrath Engineering Robbie Schwarz Engineering
Advisor(s): Robert Bittle Engineering
Location: Session: 1; 2nd Floor; Table Number: 4
The objective of our work is to design and build a depth gauge that efficiently and accurately measures the depth of a narrow hole, and give feedback via an electronic screen on the device. This design is being made for Lockheed Martin and will allow their employees to measure a large amount of rivet holes both quicker and more accurately than their current solution. Speeding up the measuring process while retaining accuracy will cut down on production time significantly. Our design was founded on the idea of a small hole gauge, we modified the gauge to be set up as a probe and anchor onto the back side of the hole. The probe has been coined as a “split-ball” due to its inner shaft splitting the outer shaft that contains a ball type end effecter. Our prototype has been through many iterations utilizing the on campus Fab Lab to 3D print most of our parts. Our mechanical team has been in close work with our electrical team to ensure that the mechanics and electronics function together seamlessly.
Author(s): Logan Smith Engineering Rachel Frank Engineering Braden Frullo Engineering Marissa Hayes Engineering
Advisor(s): Robert Bittle Engineering
Location: Session: 1; B0; Table Number: 7
The objective of our work is to design and build a depth gage that efficiently and accurately measures the depth of a narrow hole, and give feedback via an electronic screen on the device. This design is being made for Lockheed Martin and will allow their employees to measure a large amount of rivet holes both quicker and more accurately than their current solution. Speeding up the measuring process while retaining accuracy will cut down on production time significantly. Our design is small enough to be held in one hand and contains a wire probe that is plunged into the hole and latches onto the other side. The probe is “Tweezer-like” in design, with two wires that collapse and expand with the use of a button. Many parts of our design are made using a 3D printer for convenience and repeatability. Our design is able to communicate with electronics stored within the gage that measures the depth and displays to an LCD screen.
Klein Tools is a major hand tools manufacturer in US focused on electrical and utility applications for professionals. One of Klein Tools products is called a fish rod that is used by professional electricians to pull wiring through walls, conduit, and plenums to route wire from one place to another. The current fish rod assembly process at Klein Tools involves manual dispensing of glue into the metal connectors before affixing them to fiberglass rods. The objective of this Klein Tools-sponsored project is to improve the throughput of assembly system and increase the accuracy and the consistency of the amount of glue dispensed to reduce product failures and adhesive waste.
The overall system in development consists of an automated metal connector orientation system, conveyor belt assembly, a glue dispensing system and a control system. Through the application of vibratory hopper feeders, pneumatic rotary tables and grippers, sensing cameras, break line sensors, and a conveyor belt, the system will orient the metal connectors glue side up, and present the connectors with adhesive to the operator for final assembly of the fish rods.
The goal of this project is to develop a small electric vehicle that can operate autonomously or from user commands. The vehicle would be useful in cases where it is dangerous or impossible for a human to complete a task, such as risk of infectious diseases, cave-ins, or explosions. The desired functions include facilitating communication with quarantined patients, scouting dangerous buildings and caves, and developing a virtual map of its surroundings.
In the future, we will upgrade from the current proof of concept model to more a powerful and reliable base vehicle to build the desired functionality onto. We will write programs allowing remote control of the vehicle, and affix cameras and sensors to improve performance for both manual and autonomous operation. To create a virtual map of the vehicle’s surroundings, we will develop software to observe locations of walls and obstacles and transfer that data back to a computer to create a virtual map.
Author(s): Morgan Bailie Environmental Science
Advisor(s): Michael Slattery Environmental Science Tamie Morgan Environmental Science
Location: Session: 1; 3rd Floor; Table Number: 6
Despite wide-spread reliance on land and water, many developing countries face food insecurity, malnutrition, and water insecurity as a result of increasing environmental degradation. Nuclear technology presents a valid solution for these issues, despite public misperception regarding nuclear-related activities. This research examines the basis for international nuclear regulations and the effects on developing countries. The main focus of this research is on the application of gamma ray irradiation, isotope hydrology, nitrogen tracking, and receptor binding assay in Bangladesh, the Iullemeden basin of Sub-Sahara Africa, Sudan, and El Salvador respectively. Moreover, application of Geographic Information Systems (GIS) is used to quantify the successes and failures of these nuclear technologies relating to agricultural significance, land use, and water use on the effected populations. The variety of countries was chosen based on the major threats to their land and populations, the regional diversity of each country, and the use of nuclear applications in each country. Data was compiled on the cultural and economic relevance of agriculture and water-use in each country in order to accurately depict the significance of continued environmental degradation to the population.
Exploring the effectiveness of a textured surface at reducing bat activity near wind turbine towers
Bats are dying at wind farms in large numbers globally. The most effective way to mitigate these fatalities would be to address why bats come into contact with wind turbines. Texas Christian University has been working on the hypothesis that wind turbines provide or appear to provide a resource to bats, in particular foraging and/or water source. Studies have shown that the smooth metal surfaces of wind turbine towers have similar acoustic properties to those of water, which may be attracting bats to the turbine tower surface. If this is the case then removing the acoustic similarity between tower surfaces and water may reduce bat activity at wind turbines. Thus, we conducted a feasibility study in which we monitored bat activity with night vision technology, thermal cameras, and acoustic bat detectors at wind turbine tower surfaces at Wolf Ridge Wind, LLC. The goal of our study was to refine our equipment set-up to maximize bat detection and determine if a paired experimental study to test a textured turbine surface treatment was viable. During this study we confirmed the most effective equipment set-up and that the timing of our nightly survey effort was appropriate. Over 21 nights, we observed 171 bats near the turbine tower surfaces and recorded 181 acoustic echolocation calls from bats representing 5 of the 6 local bat species. Among the observations we detected 7 of 8 defined behaviors: passing, reversing, looping, foraging, skimming, sweeping, and colliding; the exception was gleaning. We further classified these behaviors into three categories based on proximity to the turbine tower surface: contact/close contact, close-range, and far-range. Of the behaviors observed, 9% were classified as contact/close contact, 42% as close-range, and 49% as far range. We then explored bat activity at specific turbine sites and found that it did not vary within turbine pairs, providing strong support for a paired study design. Altogether, our findings validate a paired study design to test a texture treatment bat deterrent and emphasize the importance of separating bat activity and behavior into distance categories, as we would expect that the application of a texture treatment to wind turbine tower surfaces would most likely only reduce contact/close contact and close-range bat activity while far-range bat activity would remain unchanged.
Author(s): Jimmy Greene Environmental Science Tory Bennett Environmental Science Tamie Morgan Interdisciplinary Michael Slattery Environmental Science
Advisor(s): Tory Bennett Environmental Science
Location: Session: 1; 1st Floor; Table Number: 1
South Africa is unique in that the majority of its wildlife is managed in privately owned game reserves. One major challenge for reserves is maintaining healthy stable populations, particularly with large species, such as the big five (white rhinoceros (Ceratotherium simum), African elephant (Loxodonta africana), Cape buffalo (Syncerus caffer), African leopard (Panthera pardus), and lion (Panthera leo)). Nevertheless, there has been very little research on management of these charismatic species in such size restricted reserves. To address this need, we are studying the impacts of white rhinoceros on the structure, composition, and spatial distribution of vegetation in Amakhala Game Reserve. The reserve was created in 1999 from 7,500 ha of agricultural land. Since the formation of the reserve, succession of vegetation to a more natural state has been encouraged. However, the introduction of large herbivores, such as the rhino in 2006, may have altered or slowed down this succession. To explore this hypothesis, we conducted a GIS analysis study. Through the use of Landsat imagery, we classified the vegetation type and analyzed historical changes due to rhinos using the Normalized Difference Vegetation Index (NDVI). We hope that these findings will facilitate game reserve management and provide a better understanding of rhino carrying capacity based on the size of a game reserve.
Author(s): Nicholas Haber Environmental Science
Advisor(s): Mike Slattery Environmental Science Becky Johnson Environmental Science Tamie Morgan Environmental Science
Location: Session: 1; B0; Table Number: 3
The Mill Branch Mitigation Bank (project site) is located in northwest Denton County, Texas approximately 17 miles northwest of the City of Denton and stores stream credits that are used to offset the degradation of our rivers. Mitigation banks also helps the U.S. Army Corps of Engineers achieve the goal of “no overall net loss” of the nations wetland aquatic resource functions that will be lost or impaired by authorized activity. The Mill Branch Mitigation Bank restored 22,876 linear feet of channel and approximately 83 acres of riparian areas within the Mill Branch and Cannon watersheds in Denton County. Before being restored, the Mill Branch stream was used by grazing animals for drinking water. Decades of intensive grazing have caused the channel to degrade, erode and widen; in addition the pasture management and operations have displaced native vegetation. The goal of this project was to define the watershed that drains into the Mill Branch Mitigation Bank, including the historical conditions of the stream and also soils impact on drainage. I used LiDAR data and produced a triangulated irregular network (TIN) that represents the surface of the watershed and then aerial photo were overlaid on the TIN making the surface more realistic. Historical aerial photos were used to examine to the impact of the changes before and after the mitigation occurred. GIS soils data will be used to look at infiltration rates and the impact it has on the volume of water entering into the stream.
This research was designed to address the drainage basin infrastructure in Arlington Heights, Fort Worth, Texas. The Central Arlington Heights watershed is a residential subdivision of roughly 454 acres, dating back to the late 1800’s. Over this extended period, this area has experienced a high rate of urbanization with both commercial and residential development with an ever increasing percentage of the area covered in impervious layers. Subsequently, the original storm water drainage system is outdated and inadequate and several residential properties have experienced significant reoccurring damage from floods.
The previous process of modeling and quantifying the input of impervious cover for storm water drainage design is outdated. A GIS analysis of the impervious cover layer in present conditions is critical in determining the percentage of cover per land parcel and comparing it to the zoning and model parameters set by the city. Using image segmentation, a remote sensing analysis was used to analyze color infrared aerial photography at a resolution of 0.5. Once segmented, a supervised classification was performed to map impervious cover. The percentage of impervious cover per land parcel and land owner was calculated and compared to present storm water design standards and City zoning requirements.
Author(s): Devon Kassler Environmental Science
Advisor(s): Tamie Morgan Environmental Science Robert Denkhaus Environmental Science
Location: Session: 2; 3rd Floor; Table Number: 7
The Fort Worth Nature Center is a nature reserve located northwest of Fort Worth, Texas. The 3621-acre reserve serves as a huge habitat for an abundant number of terrestrial and aquatic animal species. The nature center uses environmental management practices to keep the refuge as healthy and native as possible. To get a better understanding, game cameras are strategically placed by employees to take “inventor” of what creatures are on the refuge with minimal human interruption.
GIS analysis were conducted to determine species diversity and population based on locations and times. A map was created to show where the game cameras were set up in relation to each other and the boundaries on the property. The game cameras not only provide the photos but the time and date as well. The data from the game cameras were then analyzed individually and sorted to create a population density map. The results were then presented to the management which allowed them to make any adjustment they saw would benefit the refuge. In addition, the results were taken and shared in the form of an ESRI story map on the Fort Worth Nature Center’s website for public educational purposes.
A GIS project was conducted working with a client dataset of a company who offer swimming pool repair and service company in the region. The client locations were geocode using the addresses and a spatial statistical cluster analysis was performed. Given the scale of their business (the greater DFW metro), an area of high clustering and run geoprocessing tools from there identified. False-color infrared composite aerial photograph was used to identify swimming pools in the area. The location of current clients from the dataset were mapped and new clients in the area were identified. A GIS network analysis was performed to provide the business with optimal travel routes through the area with current clients and potential clients.
Opal’s Farm Project
Katrina Klawiter and Elizabeth Weber
Texas Christian University
Faculty advisor: Omar Harvey
There are more than 5,200 homeless individuals located within Fort Worth, Texas. Most rely on food received from shelters, church groups, etc. Much of this food lacks adequate amounts of essential vitamins and minerals, and is, overall, unhealthy for the people consuming these products. There is a need for access to free fruits and vegetables within the Fort Worth homeless community. In 2016, Opal Lee, a Fort Worth community leader, was given access to 7 acres of land located along the west fork of the Trinity River. Named Opal’s Farm, after Ms. Lee, her goal is to convert the area into a non-profit urban farm to help the homeless and those in need find both work and healthy food. To begin this process, Opal’s Farm needed information about the land to better determine if the west fork of the Trinity River is a good area to grow a community garden. Looking for ways to help the community, TCU’s spring Soils in the Environment class, instructed by Dr. Omar Harvey, offered to complete that analysis. During the Spring semester, soil samples were collected on site and analyzed for their essential qualities to determine whether this plot of land will be a successful and productive area for Opal's Farm.
In 2016, the Fort Worth Independent School District (FWISD) conducted voluntary lead testing of drinking/drinkable water at 127 schools and administrative locations across the district. The goal of this testing was to assess whether students were being exposed to high lead levels in drinking water while at school. There is no level of lead that is considered harmless or acceptable. Because children are still in the developmental phase of growth, they are particularly vulnerable to the negative health effects associated with exposure to lead contamination. Based on current understanding and best guess estimates, the United States Environmental Protection Agency has set a lead action level in drinking water of 15 parts per billion. Results from the FWISD study showed that 60 of 127 locations had lead levels that exceeded the action level at one or more sample points. As a corrective measure, steps were taken to remove and replace over 500 non-compliant drinking fountains and other plumbing components. This infrastructure was older and was suspected to be the primary cause of the lead contamination due to the leaching of lead from lead-containing components. However, the issue of lead contamination and its potential link to old infrastructure transcends the school system and necessitates a comprehensive assessment on a citywide level. Given that schools are developed around communities of similar age, the FWISD lead data may be useful as a proxy for assessing wider citywide potential lead-in-water and infrastructure replacement issues. In this project, lead data from the FWISD study was combined with infrastructure-related data, spatial analysis, and spatial statistics techniques to identify potential high-priority areas for the city’s lead pipe replacement project. Infrastructure-related factors included in the analysis are parcel age, pipe material, and pipe age. Greater priority will be given to vulnerable populations, including children, low-income communities, and minorities. The results are likely to indicate which areas of Fort Worth have the greatest potential for lead contamination of drinking water and which areas should receive high priority for addressing lead-in-water potential.
The Amakhala Game Reserve, located in the Eastern Cape region of South Africa, was established in 1999 as a result of a joint conservation effort between six different landowners. Prior to the establishment of the game reserve, these properties were converted to, and primarily used for farming. Since then, Amakhala has collectively committed to letting the land revert back to its natural state with great success. The reserve has grown to include 11 different properties that cover an area of roughly 75,000 hectares, and host all manner of native vegetation and wildlife, including the Big Five. The latter were considered to be the most dangerous game animals in Africa; the Leopard, Lion, Elephant, Rhino, and Cape buffalo, but currently the ability to see these animals up close and in the wild, is a major selling point for the reserve. Unfortunately, the majority of these species are under severe threat of poaching, with rhino horn and elephant ivory being the most sought after items. This poaching is often driven by the demand for traditional medicine and status symbols in Asian cultures, particularly Vietnam and China, and the perpetrators of these crimes are either affiliated with larger criminal organizations, or are driven to this action as a result of a lack of economic opportunity.
The safety and security of Amakhala’s Big Five is one of the reserves biggest priorities. In particular, the reserve places the most emphasis on the protection of the rhinos and elephants they have by using GPS collars to keep constant supervision on location and habitat use. For this project, I used digital elevation data, aerial photos, and the movement data from elephants to analyze elephant herd movement patterns and activity hotspots, and determine if their movements are influenced by season. Through a better understanding of how this herd moves, and where they move, I can inform Amakhala Game Reserve managers on not only how to protect these animals, but also identify areas where viewing opportunities would be best for their tourists.
Habitat loss is the primary cause of declining bat populations globally and urbanization represents a major contributing factor. Yet, studies have indicated that urban parks, green belts and tree-lined suburbia can support a diverse and abundant bat community. These areas must, therefore, be providing suitable and readily available resources, such as water. However, in hotter environments many natural water sources are ephemeral. Thus, we propose that bats may be relying on residential swimming pools as a water source. To explore this hypothesis, we used camcorders with night vision capabilities and bat acoustic detectors to record bat activity at a pond in a local park along with four nearby swimming pools in Fort Worth, Texas. Video and acoustic data was then analyzed to determine if bats were using swimming pools and how frequently. Our study revealed that bats drank at all four swimming pools. One pool was used by bats every survey night with multiple drinking events being recorded, while drinking frequency increased at the other three pools as the local pond began to dry up. These results indicate that swimming pools represent an important water resource for bats. The next stage would be to determine what characteristics of swimming pools are bat friendly and use this information to promote healthy, stable bat populations in urban habitats.