PRELIMINARY COMPARISON OF GEOGRAPHIC INFORMATION SYSTEM GIS ANALYSIS

PRELIMINARY COMPARISON OF GEOGRAPHIC INFORMATION SYSTEM (GIS) ANALYSIS OF PATENT TONG AND BAR CLEANING OYSTER POPULATION ESTIMATES IN THE UPPER CHESAPEAKE BAY Allen, S, H. Lane, J. Goodwin, M. Chen, M. L. Kellogg, K. T. Paynter MEES Program, 0105 Cole Field House, University of Maryland, College Park, MD 20742

Introduction • Oysters are of vital importance to the health of the Chesapeake Bay • Oyster abundance in the bay has been on the decline since the late 1800’s • Currently, there is no standard method for population estimation

Goal: Pilot Study Explore methods of oyster population estimation

Objectives • Ground truth patent tong sub-sampling • Estimate & compare oyster bar populations using two methods (Leslie/De. Lury and GIS) • Determine the feasibility & reliability of estimates based on patent tong sampling

Patent Tonging • Modern method of oyster harvesting • Allows for sampling of a greater area than with divers

Washington, D. C.

Patent Tong Survey • Created a grid over the oyster bar that was to be Bar Cleaned • Conducted transects along grid lines with the Patent Tong boat. • All Patent Tong grab locations were marked with a GPS • All oysters collected were counted and measured

Patent Tong Survey • Created a grid over the oyster bar that was to be Bar Cleaned • Conducted transects along grid lines with the Patent Tong boat • All Patent Tong grab locations were marked with a GPS • All oysters collected were counted and measured

What is GIS? GIS is a computer system capable of assembling, storing, manipulating, and displaying geographically referenced information (spatial data).

GIS Oyster Bar Population Estimates • Raw patent tong data = number of oysters per sampling point • Oyster bar population was estimated using Inverse Distance Weighting (IDW) interpolation

Initial Population Estimates • GIS Inverse Distance Weighting Interpolation of Patent Tong data GIS IDW Estimation (bu) Spaniards Point 358 Durdin 98 Carpenter's Island 24

Bar Cleaning • Important for oyster restoration in Maryland portion of the Chesapeake • Removal of as many existing oysters as feasible from an oyster bar

Bar Cleaning • Allows for destructive sampling of an oyster bar • Three oyster bars were exhaustively dredged • Total number of bushels recovered was recorded

Initial Population Estimates • Bar Cleaning data • Leslie and De. Lury Models Bar Cleaning Removal (bu) D initial population (bu) L initial population (bu) Spaniards Point 3077 3534 3298 Durdin 580 1742 675 Carpenter's Island 544 1279 799

Results Oyster Bar Population Estimates Bar Cleaning Removal (bu) D initial population (bu) L initial population (bu) GIS initial population (bu) Spaniards Point 3077 3534 3298 358 Durdin 580 1742 675 98 Carpenter's Island 544 1279 799 24

Bivariate Fit of D initial population (bu) By GIS initial population (bu) Linear Fit D initial population (bu) = 1098. 8594 + 6. 788379 GIS initial population (bu) Summary of Fit RSquare 0. 999715 RSquare Adj 0. 999429 Root Mean Square Error 28. 45784 Mean of Response Observations (or Sum Wgts) 2185 3

Results Oyster Bar Population Estimates Bar Cleaning Removal (bu) D initial population (bu) L initial population (bu) GIS initial population (bu) Spaniards Point 3077 3534 3298 358 Durdin 580 1742 675 98 Carpenter's Island 544 1279 799 24 Correlation Matrix D initial population L initial population (bu) bu) GIS initial population (bu) D initial population (bu) 1 0. 9719 0. 9999 L initial population (bu) - 1 0. 9678 GIS initial population (bu) - - 1

Future Direction • Need more data (oyster bars) • Determine the relationship between oyster bar population estimates - bar cleaning and patent tong sampling • Any suggestions are welcome

Acknowledgments Maryland Waterman’s Assoc. Ø Larry Simns, Jeff Baker Oyster Recovery Partnership The Paynter Lab