3-D Data Visualization Taking the Next Step in Data Presentation

(1)

 www.trainex.org/osc2012 

15^th Annual OSC Readiness Training Program

3-D Data Visualization

Taking the Next Step in Data Presentation

Stephen Dyment USEPA

Office of Superfund Remediation and Technology Innovation [email protected]

(2)

Two Types of Software for Environmental Data Reconstruction / Visualization

 Geographic Information Systems (GIS)

 Examples – Google Earth Pro, ArcGIS, RockWorks™

 Map (2-D) view of information

 Useful in looking at data distributions and details of some data sets

 Doesn’t allow analysis of data with depth or elevation changes

 Prerequisite to running of most 3-D programs

 3-D & 4-D data reconstruction/visualization programs

 Examples – EarthVision®, EVS/MVS, GMS, RockWorks™, ArcGIS 3D analyst

 Allows analysis of environmental data as a function of space (3-D) / time (4-D)

• e.g., hydrogeology, bedrock, vadose/saturated zone distributions, sampling protocols – discrete intervals versus lengthy well screens, source to plume linkages

 Important differentiation in types of data analysis produced by different programs

• Geostatistical versus subjective correlations

• Flexible (accepts all site data) versus fixed program structure

(3)

 15^thAnnual OSC Readiness Training Program  www.trainex.org/osc2012 2

Why 3-D, Why Now?

 Rapid acceleration of benefit and utility of

visualization platforms in the environmental industry

 Use of conceptual site models (CSMs) to support decision making

 Moving beyond conceptual “cartoons”, PRN diagram-based CSMs

 Geo-referenced geologic, hydrogeologic, and analytical data facilitate resolution of technical challenges

 Reconstruction limits data “interpretation bias”

 For information value data must be interpreted, but interpretations can be incorrect or incomplete

 EPA renewed emphasis and new focus areas

 Renewed emphasis on high quality characterization in support of remedy selection, design and optimization

 New focus on more meaningful and effective community engagement

(4)

We Live in a 3-D World

 In many cases 2D “map” views may provide sufficient detail to convey information

 Typically provides the what, limited where

 3D and 4D provide the why, how

 Depth, hydrogeologic context, time

 It’s dark down there…

 3D data visualization provides a platform to convey multiple independent data sets in simple form

 Communication tool or technical analysis?

(5)

How Effectively Can Stakeholders Understand Contaminant

Distribution and Relevance With This 2-D Visualization?

(6)

3-D Visualization of TCE Plume Escaping Groundwater Extraction System. Provides easy understanding

of threat to public well.

Here is the Same Data Set Integrated with Hydrogeology

TCE Plume Configuration and Extent is Controlled by Geology.

Control must be addressed in management strategies.

12A Plume and GETS 12A Geology and

Plume Morphology

(7)

6 Mass that

“moves” and what

monitoring wells see

Back diffusion causes challenges like rebound and long cleanup times

Courtesy of Fred Payne- Arcadis

Matrix, Contaminant, and Temporal Complexities

(8)

Technical Disagreements Among Stakeholders

 Often attributable to competing CSMs



3-D data reconstruction / visualization helps us understand the

“Rumsfeld Principle”

 Changing PMs, contractors, property owners



Variability often compromises quality of data and conclusions



3-D data reconstruction / visualization treats all data equally

 Data type and density versus resources and SOPs



Analytical and direct sensing quality vs. spatial and temporal measurement density

“As we know, there are known knowns. There are things we know we know. We also know there are known unknowns. That is to say we know there are some things we do not know. But there are also unknown unknowns, the ones we don't know we don't know.”

Donald Rumsfeld, Feb. 12, 2002 Department of Defense

(9)

Summary of Best Practices for 3-D Data Reconstruction / Visualization / Analysis



Step 1 - Identify basic questions to answer with existing data



Step 2 - Identify the types of hard data needed to answer questions



Step 3 - Determine what component reconstructions are needed



Step 4 - Sort and document hard data from interpretations



Step 5 - Import hard data into database format for building reconstruction components



Step 6 - Use GIS and 3-D data analysis to evaluate sample distributions in map format



Step 7 - Evaluate and ensure adequate distribution of geologic log data



Step 8 - Use actual (measured) data to ensure objective 3-D reconstructions

* Note - Be aware of the principal of significant figures; not only for

contaminant data; but also geology and hydrogeology

(10)

How Do I Get Started?

Newmark Orientation

 Start with a spatially correct platform

 Add elements/data sets to explore CSM issues

 Flexible, scalable, upgradable, timely

 Remote operation

(11)

10

Example #1- Newmark GW Site R9

Ongoing Remedial Actions (EPA 5-YR Review, 2008)

Modified from Stantec, 2009

(12)

 Optimize interim P&T

 ‘Information-based’ RI

 3D used as platform for preliminary conceptual site model (PCSM)

 Administrative data review

 Identify additional locations with potentially applicable information

 Perform 3-D visualization and analysis

 Review, organize, summarize historical info

 Plan and perform RI/FS

Project Goals and Approach

Newmark Newmark

(13)

 Hydrogeologic plume control

 Primary and secondary sources

 Remedy evaluation- North plant treatment

Preliminary Findings

North Treatment 1997 PCE

Newmark- CJ 10

Newmark GW levels and geologic plume controls

Integrated

Muscoy/Newmark

North Treatment 2003PCE

(14)

How Has 3D Been Used at Newmark?

 Communication and planning tool

 Presented PCSM and findings to stakeholders September 2011

 Technical analysis tool

 ICs in place to protect remedies

 Requires new permit for any new well or change of existing pumping conditions

 MODFLOW

Pathlines 1997 PCE

(15)

14

Example #2- Modern Electroplating R1

Brownfields Technical Support Center

Modern Electroplating Children’s Services Prop.

(16)

 Prior efforts focused on tactical activities and basic site description

 Evaluate historical removal/remedial actions

 Assess future data needs- particularly VI

 Build PCSM, cost effectively for Brownfields applications

 Recommendations

Project Goals and Approach

Chemical GW-2 (ug/L) GW-3 (ug/L)

PCE 50 30,000

TCE 30 5,000

Groundwater GW-2/GW-3 Criteria

(17)

 Hydrogeologic plume control

 Primary and secondary sources

 Vapor Intrusion

Preliminary Findings

TCE 5000 ppb 2001-2011

Water Levels Geology

Vertical Gradient Summary (Positive values indicate

downward direction)

PCE 100 ppb 2001-2011

Well Nest Unit

Vertical Head Difference

(ft)

Vertical Hydraulic

Gradient (ft/ft) MW101S Overburden

0.98 0.02

MW101R2 Deep Bedrock MW205S Overburden

-0.08 -0.01 MW205R Shallow Bedrock

MW109 Shallow Bedrock

0.22 0.02

MW108 Deep Bedrock MW9 Overburden

1.4 0.10

MW107 Shallow Bedrock

Average Vertical Hydraulic

Gradient 0.03

PCE 2001-2011 GW-2 Compliance

TCE 2001-2011 GW-2 Compliance

(18)

How Has 3D Been Used at Modern Electroplating?

 Communication and planning tool

 Presented PCSM and findings to stakeholders December 2011

 In conjunction with nearby sites, supports corridor redevelopment

 Technical analysis tool

 Better understanding of hydrogeologic plume controls

 Help to identify future data needs

•

Bedrock

•

Optimize VI sampling priorities

•

Limit uncertainty, expedite decision making

(19)

Color Coding

A Final Word of Caution

Significant Heterogeneity Exists

 Many EPA projects have historically used color coding as a means to convey information

 These do not convey other elements of risk management

 Concentration

•

Exposure scenarios

•

Geologic, hydrogeologic context

•

Role of heterogeneity

(20)

Ignore Heterogeneity at Your Peril

(21)

Depth-integrated, flow weighted averaging

1 10 100 1,000 10,000 100,000

176 178 180 182 184 186

Elevation (m)

PCE (ug/L)

10^-3 10^-2 10^-1

176 178 180 182 184 186

Hydraulic Conductivity (cm/sec)

(22)

Location #1 #2 #3 #4 #5 #6 #7

Vario-plot layout Determines short- scale heterogeneity

(SS) (on order of feet). SS

heterogeneity causes SS data

variability.

Initially start with a 12 X 12 ft area.

Using in situ XRF, analyze each location

in the designated pattern

(see QC for in situ XRF) 2 ft spacing between locations on one side

#8

12 ft between locations #1 and #9 12 ft between locations #7 and #10

6 ft between #4 and #8

(23)

2215^thAnnual OSC Readiness Training Program  www.trainex.org/osc2012 22

VarioPlot Example – RR Lot 3

(24)

 Lot 3 after

Lead Avg Concentration Surface Soil

Field in-situ= 40 ppm

Lab ex-situ/IS sample= 40 ppm Lead Avg Concentration

Surface Soil Field in-situ= 169 ppm

Lab ex-situ/IS sample= 171 ppm

Lead Avg Concentration

Lab ex-situ/IS sample= 1367 ppm Lead Avg

Concentration Surface Soil Field in-situ= 693 ppm

Lab ex-situ/IS sample= 860 ppm Lead Avg Concentration Lead Avg

Concentration Surface Soil Field in-situ= 204 ppm

(25)

Answers in the Field

Color Coded Decision Units

-How many Increments?

-What can we expect the mean of the ICS to be? (In-situ

surface results) -Where should be take a depth sample (highest in-situ

surface reading).

-Anything unusual?

shoot with the XRF

(26)