FATE AND TRANSPORT MODELING OF URBAN RADIOLOGICAL CONTAMINATION Jonathan Shireman (Jonathan.Shireman@Aptim.com) APTIM Federal Services, Knoxville Tennessee, USA Katherine Ratliff (ratliff.katherine@epa.gov), Anne Mikelonis (Mikelonis.Anne@epa.gov) United States Environmental Protection Agency, Durham, North Carolina, USA Parameter Building Land Use Types Road Urban q, in/hr Case Study Description 2D Overland Flow Boundary Layer Construction SWMM Model Storm and Washoff Inputs and Run Results APTIM. Expect the Extraordinary. APTIM &EPA Objective: Develop data driven strategies and methods for containing, monitoring, and remediating radioisotopes following detonation of a dirty bomb. Data Requirements and Sources: Ground deposition plume High resolution elevation data Surface permeability map Meteorological data Building footprint Transportation features Storm Sewer Features Drainage Control Measured by survey Lidar NLCD2016 NOAAdata tools City/Metro Planning City/Metro Planning City/Metro Planning City/Metro Planning Methods: Use hypothetical event in a known setting and historical meteorological conditions to simulate overland transport into a stormwater management runoff control system. ~ Historical prevailing wind direction and speed ~ Develop storm intensity and return frequency ~ Compare to design storm characteristics ~ Use existing outfall locations and hydrologic controls ~ Create 2D model, surface gradient, obstructions, grid layout, and land use types ~ Model a range of contaminant transport properties 2-Year 1-Hour Peak Rain Fall events ~ Range 0.996 - 2.08 inches total precipitation ~ 8 storms between 2004 -2017 meet criteria ~ Median Rainfall Storm July 27, 2014 ~ Duration 1 hr 3 min ~ Sustained Winds from SE Storm Tvoe Real Storm Julv 27. 2014 Duration 1.03 hr. Total Precipitation 0.96 in Max Rate 3.86 in/hr Storm Tvoe MSE Tvoe3 Desian Duration 24 hr. Total Precipitation 2.35 in Max Rate 3.86 in/hr MSE Type 3 Storm July 27, 2014 Storm 09:36 12:00 14:24 16:47 19:11 Measurement Time (hh:mm) Buildings Rapidly flushing; value near high end of parameter range of values applied Roads Roads and pavement flush more slowly due to greater roughness and lower slope; geometric mean of parameter values used Urban Mixture of open and high-density development; low end of parameter range of values used SWMM Subcatchments and Digital Elevation Raster for Radiological Contamination Transport Case Study Model Case Study Model Boundary Layer Created in ArcMAP Features within and downgradient of the radiologic hazard area: ~ Select subcatchments intersecting the plume and down gradient to the nearest outfalls ~ Clip roads to subcatchments and create polygons using ArcGIS buffer tool ~ Clip buildings to subcatchments ~ Merge SWMM subcatchment and road layers into 2D boundary layer Import 2D Boundary Layer into PCSWMM® ~ 2D Boundary Layer enforces node spacing, cell geometry ~ Tag attribute used to differentiate land use types ~ SWMM subcatchment sublayer set to hexagonal geometry with 80 foot resolution (center to center spacings) ~ Road Sublayer set to rectangle geometry with 40 resolution Generate 2D Mesh PCSWMM® ~ 2D mesh generates 2D cells, and SWMM conduits and junctions Import 2D cells and Buildings as SWMM Subcatchments and Build Runoff Model Estimate of Washoff Parameters for Modeling Radiological Contaminant Behavior Washoff Estimates for Three Land Uses Stormwater subcatchments potentially affected by air Sustained Winds 1990 - 2017 dispersion of radiological contamination Dominant sustained Winds from SW at 20 to 30 mph Storm Return Frequency Analysis 2D Cells by % 2D Cells by Impervious Slope 2D Cells by Elevation Under actual incident conditions the ground plume geometry would be obtained by empirical measurements, For the case study model (CSM) a hypothetical ground deposition plume was modeled using data from NOAA National Center for Environmental Information > Data Access > Land Based Stations > State and County, 20 years of climate observations for stations near the hypothetical incident. Wind Speed and Directional Analysis o Wind Speeds ~ ~ ~ ~ ~ ~ Import 2D Cells into subcatchment layer Merge buildings into subcatchment layer and assign outlets using PCSWMM® tools SWMM IMPERV, SLOPE and ELEV Attributes populated using PCSWWM® Area Weighting Tool and NLCD or DEM raster files Assign outlet nodes for buildings using PCSWMM® tools Create outlets and connecting conduits Create Rainfall Time series from historical precipitation data Components Needed for 2D Mesh Literature Review of Contamination Transport ~ Primary fallout contaminant is Cs-137, also most likely isotope to be used in a dirty bomb ~ Most studies of Cs-137 soil contamination focus on rural or agricultural areas, findings indicated about 0.1 % of watershed inventory lost per year ~ Studies of other pollutants (e.g. suspended solids or total nitrogen) focus on urban settings ~ Cesium adsorbs strongly to clay and organic matter; primary components of suspended sediment (TSS). Managing Uncertainty: Range of TSS washoff parameters is wide and location specific. Washoff assessments assuming standard exponential function illustrates behavior used for the case study model. Peak runoff depth (q) used for sensitivity analysis. Conclusion: Cesium in the deposition plume is expected to adsorb to sediment, therefore TSS can be used as a surrogate. However, the low loss of cesium inventory of studied watersheds indicates lower end of washoff factors apply. Model Washoff Results 2-year Return Frequency Storms ~ Building first flush primary contribution to initial runoff; ~ With increasing duration greater fractions of initial inventory removed from roads and eventually from urban areas ~ Primary transport is along roadways ~ Contamination does not reach outfalls during short duration storm, contamination discharges during 24-hour Storm Conclusion: A 2D overland flow model based on available urban planning data and a hypothetical ground deposition plume is an effective method to identify likely overland transport pathways and to visualize mass flux of contaminants in response to high return frequency rain events. This method allows rapid identification of infrastructure features that have a high potential for impact, provides insight for planning response actions and supports development of long-term monitoring programs. Cumulative Washoff Assessment Identify Potentially Affected Subcatchments ~ Prevailing wind direction indicates potential contamination dispersal to north-north east ~ Initially 215 subcatchments were identified as potentially directly impacted or receiving contaminated runoff ~ Air dispersal modeling showed the footprint of the ground deposition plume representing a potential health hazard was restricted to an area near the detonation point ~ Mesh generation using all 215 subcatchments and building obstructions required over 24 hours of computational time on standard desktop processor. ~ By focusing on the area where high level hazards (orange cross hatch overlay) are anticipated and removing small buildings (< 200 ft2 ) mesh generation could be accomplished in abut 1 hour. The areas selected for inclusion in the 2D overland flow model are outlined in white above. EPA's Storm Water Management Model (SWMM) was used and implemented through PCSWMM® to model 2D overland flow. ~ 0.1 1 10 Runoff Duration (hours) y = 0.5672ln(x) + 0.6537 R2 = 0.9612 Ł = 0.1 Prevailing Wind Direction Detonation point ------- |