SPATIALLY-TARGETED PEST MANAGEMENT USING GEOSPATIAL AND AEROSPACE TECHNOLOGIES

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1001538
• Project Start Date: Nov 22, 2013
• Project End Date: Sep 30, 2018
• Program Code: [(N/A)]- (N/A)

Recipient Organization
WEST VIRGINIA UNIVERSITY
886 CHESTNUT RIDGE RD RM 202
MORGANTOWN,WV 26505-2742

Performing Department
Plant & Soil Sciences

Non Technical Summary
United States agriculture and forestry faces a major threat by many pests (e.g. insects, weeds, and plant diseases) distributed over large geographic areas. Specifically, noxious invasive species have caused considerable economic loss and environmental damage to agriculture and forestry in the United States. The result of damage and loss caused by such pests is sometimes irreversible without accurate management of the threat in a timely manner. The goal of this project is to develop spatially-targeted pest management systems using geospatial and/or aerospace technologies for efficient detection of pests and for precise delivery of control measures against major pests in agriculture and forestry. To achieve this goal, there are three objectives for the proposed project: development of unmanned aerial vehicle (UAV) systems for aerial detection of major pests (Objective 1), development of precision aerial delivery systems for biological and chemical control measures using UAV systems (Objective 2), and investigation of spatial distribution patterns of hemlock woolly adelgid and invasive bees using geospatial technology (Objective 3). For objective 1, UAV systems will be developed for aerial detection of major pests. Modular image sensors will be equipped and image analysis modules will be developed. For objective 2, precision aerial delivery systems for biological control measures will be developed. Mass rearing of natural enemies including spined soldier bugs will be established and aerial delivery modules including "bug cannon" and "bud pods" will be invented. Flgiht tests and efficacy of delivery systems will be conducted in various agroecosystems. For objective 3, spatial distribution patterns of major insects pest will be investigated using geospatial technologies. Spatial distribution patterns of hemlock woolly adelgid will be elucidated with geostatistical analysis and community-level spatial dynamics of invasive bees will be determined with diversity indices and spatial analyses.

Animal Health Component

40%

Research Effort Categories

Basic

10%

Applied

40%

Developmental

50%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
211	0699	1130	70%
125	0613	1070	10%
213	0799	1140	10%
215	1199	3100	10%

Knowledge Area
211 - Insects, Mites, and Other Arthropods Affecting Plants; 215 - Biological Control of Pests Affecting Plants; 125 - Agroforestry; 213 - Weeds Affecting Plants;

Subject Of Investigation
0799 - Rangelands and grasslands, general; 1199 - Deciduous and small fruits, general/other; 0699 - Trees, forests, and forest products, general; 0613 - Mixed conifer-broadleaf forests;

Field Of Science
1130 - Entomology and acarology; 1070 - Ecology; 1140 - Weed science; 3100 - Management;

Keywords

biological control

invasive speices

pest management

spatial analysis

pest survey

Goals / Objectives
The goal of this project is to develop spatially-targeted pest management systems using geospatial and aerospace technologies for efficient detection of pests and precise delivery of control measures against major pests in agriculture and forestry. To achieve this goal, there are three objectives: objective 1, development of unmanned aerial vehicle (UAV) systems for aerial detection of major pests; objective 2, development of precision aerial delivery systems for control measures using UAV systems; objective 3, investigation of spatial distribution patterns of hemlock woolly adelgid and invasive bees using geospatial technology.

Project Methods
Development of UAV systems: The design concept of the unmanned aerial vehicle (UAV) for this project is small, versatile, low-cost, and transportable so that flight operations can be handled by a two person crew with minimal operational cost. An existing UAV, the WVU Sharpshooter will be used as the airborne platform which provides substantial on-board computational power and hardware redundancy for increased operational safety, including aircraft sensors, power systems, flight control system. The video camera and sensors equipped on the UAV allows for real-time and continuous monitoring of a location. To validate the system, plants for experiments will be systematically laid out in the field for aerial imaging. Once aerial images are taken from the UAV, a composite image will be generated using readily available software packages. The composite image will be georeferenced to confer spatial attributes to the map coordinate system, i.e. ortho-mosaicking. From the images, normalized difference vegetation index (NDVI) will be calcuated to detect pests. Design and development of aerial delivery module: To take advantage of the flight capability of the spined soldier bug, bug cannon will be used. The delivery device is externally mounted on the wings of the airplane. Because mile-a-minute weevil does not have a high flight capability compared to spined soldier bugs, it is critical that they are delivered precisely where the mile-a-minute weed infestation is located. To accomplish this goal, a low cost delivery mechanism will be designed, constructed, and field test validated. A candidate design for the proposed delivery mechanism would be delivery pods, which are un-powered and designed to descend in free fall. Prior to each flight, a number of such delivery pods will be loaded with the insect natural enemies and a supply of moistened cotton, and stored in the aircraft. A field test will be conducted for the delivery approaches for both insect natural enemies at Jackson's Mill Flight Test Facility. Each natural enemy will be deployed at two different altitudes and three different flight speeds to investigate the effects of flight condition to precision of delivery. A total of 60 flights (i.e. 10 replications per altitude/speed combination) will be made and 20 natural enemies for each of the spined soldier bug and mile-a-minute weevil will be deployed from the module per flight. Once the natural enemies are deployed, we will recover and rear them in the laboratory to compare with the control (i.e. insects from lab colony) in order to investigate the effect of aerial deployment on insect body survival, development, and reproduction. If the deployment affects the insects negatively, we will adjust flight altitude and speed to minimize the negative effect of aerial deployment on the natural enemy. Spatial and temporal distribution of hemlock woolly adelgid (HWA): Currently available HWA sampling programs, including binomial sequential sampling, are based on the assumption that each sample is independent. In such case, it is important to know the minimum distance at which each datum can be expected to be independent by analyzing spatial dependence. Spatial dependence means that sample locations are so close that sample values tend to be more similar than sample values that are far away. To quantify spatial dependence, the semivariogram and spatial analysis by distance indices (SADIE) will be used. Each of three research sites in Preston, Grant, and Monongalia counties in West Virginia will be divided into 100 sections (10 m by 10 m), and a tree located in the approximately middle part in each section will be selected. The selected tree will be marked and the egg masses per sample unit in each section will be counted and recorded every two weeks. Environmental factors will be sampled at each sample location. Elevation, aspect, and slope will be measured using GPS, site map data and ArcGIS® 9 (ESRI, Redland, CA). Community-level spatial dynamics of invasive bees: Community ecology is concerned with the distribution, abundance, and diversity, and interactions of multiple species. One important aspect of community level studies is the classic inquiry of biogeography, "what forces across the landscape drive species richness and abundance?" This study will elucidate the role of invasive bees on bee community structure in West Virginia. Emphasis will be placed on the spatiotemporal relationship between these factors, how bee community composition changes, and what impact the landscape has on invasive bees. This study will be conducted within the boundaries of the Canaan Valley National Wildlife Refuge. A 6.25-km2 area will be broken into a 100 × 100 grid cells. Sampling will be conducted within each grid cell at the end of each month from April - September. Sampling for bees will be conducted with a pan trap array (ten traps per array) placed in an open location within each grid cell for one week and with a standardized sweeping protocol where every flowering plant within a height of 4 m and diameter of 20 m will be sampled from using a sweep net. For each grid cell, a Shannon-Weiner Index value and bee abundance value will be calculated (Magurran, 2004). Floral variables will be measured within each grid cell using varying protocols. Floral diversity will be measured by identification of all flowering plants within each sweep sampling point in a grid cell, and the Shannon-Wiener Index will be used as the measurement of floral diversity in each grid cell. Floral abundance will be determined by estimating the number of flowering plants of each species at each sweeping site. The centroid of each grid cell will be used as a point for geostatistical analysis. Averages of bee diversity and abundance and flower diversity and abundance will be attributed to the centroid of each grid cell. Data will be analyzed using spatial exploratory techniques (kriging and SADIE), time series analysis, and Sørensen's Index (for community similarity). Kriging will be used to explore patterns between measured variables and bee diversity and abundance. Multivariate analyses will be used to quantify the impact of different variables on bee communities. Time series analysis will be used to determine how seasonality affects bee diversity and abundance. Sørensen's Index will be used to compare bee community structure at different locations and during different times. Capture of invasive bee species will be analyzed separately and in tandem with capture of native bees in all analyses.

Progress 11/22/13 to 09/30/18

Outputs
Target Audience:Findings from the research were presented at the departmental/institutional seminars and seven entomological conferences. At the conferences which over 3,000 people attended, two presentations were given associated with the project. In addition, > 3,000 people including K-12 students learned about the project and its outputs through the outreach program at the WVU Insect Zoo and Museum. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two graduate students, three undergraduate students, and one entomology staff were involved in this project and trained for conducting a multidisciplinary approach to solve the pest problem in the forest ecosystem. How have the results been disseminated to communities of interest?The results of this project have been disseminated to the science community through various entomology conferences and institutional/departmental seminars. Also, the results of the research were disseminated to the public (e.g. growers, master gardeners, and K-12 students) via outreach with "WVU Insect Zoo and Museum" that the PI is operating at West Virginia University. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? For objectives 1 and 2, a new multi-rotor platform, named "WVU Octo", was developed and configured to carry approximately 15-16 lbs. of payload with estimated flight time (hover) of about 15 minutes. In addition to this unmanned aerial system (UAS), we developed an aerial dispensing system for precision release of mile-a-minute weevil (a biological control agent against mile-a-minute weeds), Rhinoncomimus latipes. Rhinoncomimus latipes (Coleoptera: Curculionidae) is a major biological control agent against the invasive plant Persicaria perfoliata. A low cost, releasemechanism was designed, constructed and tested in the field. After the aerial release of the weevil, we recovered the weevils and conducted a feeding assay. We found that aerial release of the weevil did not affect their feeding capability. Our study demonstratedthe potential of UAV for the precision aerial release of biological control agents to control invasive plants. The aerial deployment systems we have developed, including both pods and a dispenser, are low cost, logistically practical, and effective with no negative effects on aerially released R. latipes. In addition, we conducted aerial detection of pests including insects, plant diseases, and weeds in various ecosystems. Fire blight was surveyed in a block of pear trees located in the WVU research farms. The UAV with a camera could detect the decline of individual branches. A follow-up laboratory test confirmed the presence of Erwinia amylovora, the causal bacterium for fire blight. For the detection of tent caterpillar nests, all individual tents made by tent caterpillars of five fruit trees were located and counted by a ground survey. By using the rotary-wing UAV, we found that tent caterpillar nests were very distinguishable on the aerial imagery. We also detected other pests including an insect (saddleback moth), plant diseases (fire blight and apple scab), and a weed (morning glory) using UAS equipped with a camera. We demonstrated that saddleback moth pupae could be detected during the winter with rotary-wing UAS and morning glory flowers in vineyards can be detected from aerial imagery taken from fixed-wing UAS. Image analysis with various techniques showed the potential of removing shadows of grapevine to better detect morning glory flowers in vineyards. For objective 3, we analyzed a 3-yr data collected to characterize spatial distribution patterns of the hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae) on eastern hemlock and its spatial association with the surrounding environment. We found from spatial analyses that A. tsugae ovisacs were spatially aggregated regardless of sites and generations and 10 out of 12 cases of spatial analyses showed spatial structure. The spatial associations of aspect, diameter at breast height, and hemlock tree height with A. tsugae population showed some significant spatial associations although elevation, plant species diversity and basal area showed an inconsistent spatial relationship.

Publications

Progress 10/01/17 to 09/30/18

Outputs
Target Audience:The results of this project were presented to more than 300 scientists via seminar and conference presentations and > 600 people including K-12 students learned about the project and its outputs through the outreach program at the WVU Insect Zoo and Museum. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One staff entomologist and one graduate students were involved in this project. There were trained for the potential use of UAS in solving various pest problems in agricultural ecosystems. How have the results been disseminated to communities of interest?Research results were disseminated to science and agricultural communities via presentations. Also, an outreach to public including K-12 students was conducted with the WVU Insect Zoo and Museum which the PI is in charge of at West Virginia University. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? During this project period, we used unmanned aerial system (UAS) to detect various pests including an insect (saddleback moth), plant diseases (fire blight and apple scab), and a weed (morning glory). We used both fixed-wing UAS and rotary-wing UAS flown at various altitudes. The results of our study showed that saddleback moth pupae could be detected during the winter from the aerial imagery taken with rotary-wing UAS 1 m above the canopy of poplar trees. Morning glory in vineyards can be detected from aerial imagery taken from fixed-wing UAS flown at the speed at 60 km per hourand 20 m above the ground. Image analysis with various techniques showed the potential of removing shadows of grapevine to better detect morning glory flowers in vineyards. Fire blight and apple scab detections were performed, but the results did not show any promising detectability due to a very low incidence of the plant diseases in the research sites.

Publications

• Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Park, Y.-L. 2018. Pest detection using unmanned aerial systems. Department of Plant Protection, National Institute of Agricultural Sciences, Wanju, South Korea.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Park, Y.-L. 2018. Detecting Insects from the Sky: Satellites to Drones. Annual Meeting of Entomological Society of America. Vancouver, Canada.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2018 Citation: Park, Y.-L. 2018. Bug bombs for biological weed control using natural enemy insects. Korean Applied Entomological Society annual conference. Mokpo, South Korea.

Progress 10/01/16 to 09/30/17

Outputs
Target Audience:The resultsof this project werepresented to more than >100 scientists via seminar and entomological conference presentations. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One staff entomologist, one graduate students, and three undergraduate students wereinvolved in this project. There weretrained for a multidisciplinary approach to address pest issues in various agricultural ecosystems. How have the results been disseminated to communities of interest?Research results were disseminated to science and agricultural communities via presentations. Also, an outreach to public including K-12 students was conducted with the WVU Insect Zoo and Museum which the PI is in charge of at West Virginia University. What do you plan to do during the next reporting period to accomplish the goals?Objective 3 will be the main focus of the research during the next reporting period. We will investigate spatialdistribution patterns of hemlock woolly adelgids using geostatistics and spatial modeling.

Impacts
What was accomplished under these goals? Objective 2 (development of precision aerial delivery systems for control measures using UAV systems) was accomplished during the project period. UAV and aerial delivery systems weredeveloped and tested for precision release of Rhinoncomimus latipes.Rhinoncomimus latipes (Coleoptera: Curculionidae) is a major biological control agent against the invasive plantPersicariaperfoliata. The release of R.latipesis challenging with the current visit-and-hand release approach because P.perfoliata shows a high degree of patchiness in the landscape,possesses recurved barbs on its stems, and often spreads into hard-to-accessareas. This study developed and evaluated UAVfor the precise aerial release of R.latipes to control P.perfoliata.We have developed twoUAVs (i.e. quad-rotor and tri-rotor) and an aerial release system to disseminate R.latipes. These include pods containing R.latipes and a dispenser to accommodate eight pods. Results of field tests to evaluate the systems showed no significant (P>0.05) effects on survivorship and feeding ability of R.latipes after aerial release. Our study demonstrates the potential of UAVfor the precision aerial release of biological control agents to control invasive plants. The aerial deploymentsystems we have developed, includingboth pods and a dispenser, arelow cost, logistically practical, and effective with no negative effects on aerially released R.latipes.

Publications

• Type: Journal Articles Status: Accepted Year Published: 2017 Citation: Park, Y.-L., S. Gururajan, H. Thistle, R. Chandran, and R. Reardon. 2017. Aerial release of Rhinoncomimus latipes (Coleoptera: Curculionidae) to control Persicaria perfoliata (Polygonaceae) using an unmanned aerial system. Pest Management Science. DOI: 10.1002/ps.4670.

Progress 10/01/15 to 09/30/16

Outputs
Target Audience:The product of this project was presented to more than >100 scientists via seminar and conference presentations. Also, ca. 200 people including K-12 students learned about the project and its outputs through the outreach program at the WVU Insect Zoo and Museum. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One staff entomologist wasinvolved in this project and she was trained for a multidisciplinary approach to address pest issues in variousagricultural ecosystems. How have the results been disseminated to communities of interest?Research results were disseminated to science and agricultural communities via a conference and a seminar. Also, an outreach to public including K-12 students was conducted with the WVU Insect Zoo and Museum which the PI is in charge of at West Virginia University. What do you plan to do during the next reporting period to accomplish the goals?For objectives 1 and 2, the UAV will be further developed to finalize operation protocols for the detection and delivery of control measures aerially. Field tests will be performed in various agricultural and forest ecosystems. For objective 3, we will conduct further data analyses and publish the results in entomological and agricultural journals

Impacts
What was accomplished under these goals? For objectives1 and 2, avariety of pests were surveyed aerially with a rotary-wing UAV by detecting feeding damage of insect pests, and signs and symptoms of diseases.Fire blight was surveyed in a block of pear trees. The UAV was flown at 20 m above the canopy of the pear block first, and then flown at 5 m to obtain aerial images with a higher resolution. The aerial image could detect the decline of individual branches. A follow-up laboratory test confirmed the presence ofErwiniaamylovora,the causal bacterium for fire blight. Because fire blight starts from the upper part of fruit-tree canopy, UAVs can be an ideal tool to be used for early detection of fire blight in orchards. For the detection of tent caterpillar nests, all individual tents made by tent caterpillars of five fruit trees were located and counted by a ground survey. By using the rotary-wing UAV, aerial images were taken at 20 m and 5 m above the canopy to locate individual tents. Tent caterpillar nests were very distinguishable on the aerial imagery. The result of statistical analysis showed that there were no differences (P > 0.05) in numbers of tents made by tent caterpillars between two survey methods (i.e. ground survey and aerial survey) (t = 1.87, d.f. = 4 with the UAV flown at 5 m above the canopy; t = 2.37, d.f. = 4 with the UAV flown at 20 m above the canopy). This indicates that tents made by caterpillars can be readily detectable by aerial survey with UAVs.

Publications

• Type: Journal Articles Status: Awaiting Publication Year Published: 2017 Citation: Turcotte, R. M., A. Lagalante, J. Jones, F. Cook, T. Elliott, A. A. Billings, and Y.-L. Park. 2017. Spatial and temporal distribution of imidacloprid within the crown of eastern hemlock. Journal of Insect Science, doi:10.1093/jisesa/iew120.
• Type: Theses/Dissertations Status: Published Year Published: 2016 Citation: Turcotte, R. 2016. Temporal and Spatial Distribution of Imidacloprid and the Arthropod Fauna Associated with Eastern Hemlock, Tsuga canadensis (L.) Carr. Ph.D. Dissertation, West Virginia University, Morgantown, WV.

Progress 10/01/14 to 09/30/15

Outputs
Target Audience:Findings from research were presented at the department seminar at Kansas State University and two annual conferences of the Entomological Society of America. More than 60 faculty, staff, students, and USDA-ARS researchers were attendedat the departmental seminar. At the conferences which over 3,000 people attended, two presentations were given associated with the project. Also, the results were featured in the media for entomology community over the world. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?One graduate student and one staff entomologist were involved in this project and there were trained for conducting a multidisciplinary approach to solve pest problem in the forest ecosystem. How have the results been disseminated to communities of interest?Research findings from this project were disseminated to science communities via four different conference presentations and two institutional seminars. Also, the results of research were disseminated topublic (e.g. growers, master gardeners, and K-12 students)via outreach with "WVU Insect Zoo and Museum" that the PI is operating at West Virginia University. What do you plan to do during the next reporting period to accomplish the goals?For objectives 1 and 2, we will further develop UAV with field tests to optimize operation protocols. In addition, we will upgrade the delivery system for spined soldier bugs (Podisus maculiventris) which is a key predator of caterpillars such as many agricultural and forestinsect pests. For objective 3, we will conduct further data analyses andpublish the results to journals.

Impacts
What was accomplished under these goals? For objectives 1 and 2, during this project period we have tested rotor UAVs for aerial delivery of the mile-a-minute weevil, a biological control agent against mile-a-minute weeds.Although delivery pods were developed and tested successfullybased on initial tests, it was determined that small helicopters were not ideal platforms to carry additional payloads as they were unstable and hard to control. Therefore, we conducted field tests using quad-rotor UAVs for aerial delivery of mile-a-minute weevils in bug bombs; rotor UAVsaccommodated more stable hovering and dropping of bug bombs housing the weevils.After aerial delivery with bug bombs, we recovered the weevils and conducted a feeding assay. The amount of foliage feeding by weevils collected from the field (control) was compared with that by weevils recovered after aerial delivery with bug bombs. This experiment was replicated with 30 weevils and we did not find any significant (P < 0.05) difference in feeding amount, indicating that aerial delivery did not affect weevil's feeding capability. For objective 3, we analyzed a3-yr data collected to characterize spatial distribution patterns of the hemlock woolly adelgid, Adelges tsugae Annand (Hemiptera: Adelgidae) on eastern hemlock and its spatial association with the surrounding environment. In each of three hemlock stands, a 100-m2 area was selected and divided into 100 grids (i.e. 10 m by 10 m per grid) in eastern hemlock stand. Two 50-cm branches from a tree located in the middle of each grid were sampled and the number of ovisacs were counted. This sampling was done at peak ovisac occurrence for each of A. tsugae generations (i.e. progrediens and sistens). In addition, environmental and biological factors including elevation, aspect, tree size, basal area, and plant species diversity, were sampled to investigate their spatial associations with A. tsugae. Spatial distribution patterns of A. tsugae and its spatial associations with the surrounding environments were analyzed with semivariogram and spatial analysis by distance indices (SADIE). The results from the spatial analyses showed that A. tsugae ovisacs were spatially aggregated regardless of sites and generations and 10 out of 12 cases of spatial analyses showed spatial structure. The spatial associations of aspect, diameter at breast height, and hemlock tree height with A. tsugae population showed some significant spatial associations although elevation, plant species diversity and basal area showed inconsistent spatial relationship.

Publications

• Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: Development of systems for predicting spatial distribution of stink bugs using geospatial analysis, sensing technology, and population genetics. Department of Insect Pest Management, National Academy of Agricultural Science, Rural Development Administration, Wanju, South Korea.
• Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: Park, Y.-L. 2015. Aerial monitoring and management of insects and weeds using unmanned vehicles. Appalachian Grazing Conference, West Virginia University, Morgantown, WV.
• Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: Park, Y.-L. 2015. Robotics and sensors - Examples from the field. 8th International Integrated Pest Management Symposium, Salt Lake City, UT.
• Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: Park, Y.-L. 2015. Precision detection and control of pests using unmanned aerial vehicles. Annual Entomological Society of America Meeting, Minneapolis, MN.
• Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: Park, Y.-L. 2015. Aerial detection of invasive pests using drones. West Virginia Invasive Species Working Group Meeting. Wesleyan College, Buchannan, WV.
• Type: Conference Papers and Presentations Status: Published Year Published: 2015 Citation: Park, Y.-L. 2015. Bug Recon and Bug Bomb: Drones for IPM. Annual Entomological Society of America Meeting, Rehoboth Beach, DE.

Progress 11/22/13 to 09/30/14

Outputs
Target Audience: Findings from research were presented at the department seminar at Kansas State University and two annual conferences of the Entomological Society of America. More than 60 faculty, staff, students, and USDA-ARS researchers were attended at the departmental seminar. At the conferences which over 3,000 people attended, two presentations were givenassociated with the project. Also, the results were featured in the media for entomology community over the world. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? One graduate student, one undergraduate student, and one staff were involved in this project and trained for conducting a multidisciplinary approach to solve pest problem in the forest ecosystem. How have the results been disseminated to communities of interest? The results of this project have been disseminated to science community through two entomology conferences and one departmental seminar. What do you plan to do during the next reporting period to accomplish the goals? For objectives 1 and 2, we will conduct field tests of WVU Octo and the natural-enemy-delivery systems in WVU Flight Test Facility. A sub-system for aerial delivery (e.g. ejecting or releasing natural enemies) from the model WVU Octo will also be tested and the delivery mechanisms as well as the housing for the insect natural enemies will be refined. For objective 3, we willconduct a field study to determine the distribution pattern of HWAand its spatial associations with surrounding environment. A total 100 grids (each grid being 10 m by 10 m) will be laid out inside eastern hemlock stands in Blackwater Fall State Park (WV), Cathedral State Park (WV), and Buchanan State Forest (PA). In each grid, two 50-cm branches from a selected tree will be cut and surrounding trees (i.e. species, diameter at breast height, and height) and geographic factors (i.e. altitude and slope) will be sampled. Spatial distribution patterns of A. tsugae and its spatial associations with the surrounding environments will be analyzed by using Spatial Analysis by Distance Indices (SADIE).

Impacts
What was accomplished under these goals? For objective 1, anew platform, multi-rotor named WVU Octo was developed and configured to carry approximately 15-16 lbs. of payload with an estimated flight time (hover) of about 15 minutes. For objective 2, An aerial dispensing system for precision delivery of mile-a-minute weevil (a biological control agent against mile-a-minute weeds) was developed. As the weevil does not have a high flight capability, it is critical that they are delivered precisely where the mile-a-minute weed infestation is located. Towards that goal, a low cost, delivery mechanism was designed, constructed and tested in the field. They were constructed out of medium strength cardboard, featuring a ridge like interior structure that would house the weevils. For objective 3, we conducted an experiment to characterize the distribution and to decide a suitable sample units of hemlock woolly adelgid (HWA) ovisacs in Blackwater Fall State Park (WV), Cathedral State Park (WV), and Buchanan State Forest (PA). A total 744 branches with 100cm in 31 trees were sampled. In each tree, 24 branches were measured its vertical height and then, cut with regard to six vertical heights and four cardinal directions. The cut branches were brought to the lab to count the number, measure twig lengths at the smallest node level, and map the location of ovisacs within a branch. The effect of branch height and direction on the number of HWA ovisacs was determined and HWA ovisacs within a branch were mapped with ArcGIS and analyzed to estimate aggregation index with SADIE. This study revealed that 50cm branches in approximately 3m height could be an efficient sample unit for HWA sampling. By using a standard and smaller sample unit, the results related with HWA samplings could be more consistent, reliable results might be acquired with less efforts, and the tree damage by sampling could be decreased.

Publications

• Type: Conference Papers and Presentations Status: Accepted Year Published: 2014 Citation: Park, Y.-L. 2014. Spatially-targeted pest management using geospatial and aerospace technology. Departmental Seminar Series. Kansas State University, Manhattan, KS.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2014 Citation: Baek, S., and Y.-L. Park. 2014. Spatial and temporal associations between hemlock woolly adelgids (Hemiptera: Adelgidae) and surrounding environment). Annual Entomological Society of America Meeting, Portland, OR.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2014 Citation: Baek, S., and Y.-L. Park. 2014. Spatial distribution and population dynamics of hemlock woolly adelgids (Hemiptera: Adelgidae). Entomological Society of America Branch Meeting, Williamsburg, VA.