Visualization of Geotechnical Data for Hazard Mitigation and Disaster Response (2015)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

21 The following sections generally describe the state of the practice with respect to visualization of geotechnical data rather than the state of the art. It is evident from the study sur- vey and interviews that the state of the practice encompasses a wide range of experience and tools. DATA MANAGEMENT Data management refers to the collection, storage, and retrieval of data. Results of the survey of state geotechnical leaders indicate that most DOTs store their geotechnical data in some combination of paper and electronic files. Among all of the geotechnical data identified by the DOTs, approximately 28% are kept as paper only, 27% are kept in electronic form only, and 45% are kept in both formats. On average, the DOTs keep seven different types of geotechnical data. The survey also indicates that only about 10% of the DOTs are using a centralized database capable of storing multiple types of geotechnical data. This suggests that many of the DOTs are storing data in isolated, and likely incompatible, for- mats, making it difficult to visualize related data. For example, laboratory data may be stored separately from boring log or instrumentation data. If these data were stored in a common database, retrieval and visualization of related data would likely be easier to accomplish. Interviews also indicated that re-use of geotechnical information is limited by the lack of a common data format and storage location. Inefficient loca- tion and retrieval of historical data often leads to unnecessary replication. The ability to rapidly access and visualize geotechnical data may not be critical to the success of a geotechnical hazard miti- gation project or long-term recovery from a geotechnical disas- ter; however, during the immediate and short-term response to a geotechnical disaster, speed is essential to organizing rescue efforts, maintaining public safety, and minimizing transporta- tion system impacts. A common database could expedite the retrieval and visualization of critical geotechnical data. Adopting a standard data interchange format could greatly improve data management for the DOTs, facilitating storage and retrieval of data within the organization and simplify- ing data delivery for third-party data providers (e.g., drillers, laboratories, consultants). Having a standard data interchange format for geotechnical data also would encourage visualiza- tion software developers to incorporate the standard in their products, making GDV simpler to achieve throughout a DOT. Because a number of the state DOTs and the ASCE’s Geo- Institute have participated in the development of the DIGGS standard (described later in this chapter), it may be reasonable to assume that DIGGS will eventually become the de facto standard for all state DOTs. GEOTECHNICAL ANALYSIS Only a few geotechnical analysis software packages, including slope stability, settlement, retaining wall, and terrain modeling software packages, were identified in the survey of the state DOTs. Because geotechnical analysis software was not the focus of the survey, respondents may not have identified this software as having a significant role in visualization of geo- technical data; however, it can be assumed that the geo technical engineering staff at the state DOTs use more analytical and numerical software packages than were identified in the survey. The visualization capabilities of geotechnical analysis soft- ware packages are continually improving. This is true with respect to visualization of input parameters and analytical results, but also with respect to parameter entry. Command-line and form-driven parameter entry processes are being replaced by more effective and intuitive graphical interfaces. INSTRUMENTATION Geotechnical instruments are a significant source of data for the state DOT geotechnical engineers (see Figure 15). About 95% of the DOTs report using inclinometers; 88% are using piezometers; 83% are using settlement gages; and 73% are using open stand pipe wells on their projects. Other less com- mon instruments include optical and automated surveys, load and displacement gages, and seismometers. Nearly all (93%) of the state DOT geotechnical leaders report that expert opinion and engineering judgment is a pri- mary driver of the decision to use instruments on a geo technical project. About one-third of the DOTs use risk analysis meth- ods to support the opinions and judgments. (The study survey did not ask what type of risk methods were used, but presum- ably they are typical qualitative and quantitative risk assess- ment and risk management methods.) Only 17% of the DOTs reported that geotechnical instrumentation is required by the chapter six CURRENT PRACTICE AND EXPERIENCE

22 department, and that they use expert opinion, engineering judg- ment, and analysis to determine which instruments to use. A critical step in using geotechnical instruments is establish- ing warning and action levels for each instrument. Approxi- mately 90% of the state DOTs reported that expert opinion and engineering judgment are used to set warning and action levels. Nearly 70% use some form of analysis to support the opinions and judgments, and about 40% also use field or laboratory tests. Only about 16% rely solely on opinions and expert recommen- dations. One respondent noted the potential for adverse con- sequences arising from instrumentation false alarms. While a low frequency of false alarms is a desirable goal for geotechni- cal instruments, some understanding of the significance of the instrument readings must be incorporated in any instrumenta- tion system to realize the full value of the data collected. Geotechnical instrumentation data are collected and stored in a variety of formats, including manual readings recorded on paper, portable recording devices, and highly automated, database-driven systems accessible by the web. Portable recording devices are hand-held, usually battery-powered, and generally require field personnel to connect them to an installed sensor to record a current reading. Data from por- table devices are typically uploaded to another system for processing and interpretation. Automated systems are either permanently connected to an array of sensors or can contact the sensors at specified time intervals. In some cases, auto- mated systems can also serve as the data processing and interpretation system or transmit readings directly to another system for processing and interpretation. Nearly 90% of the DOTs report using manual methods to record instrument data; 70% are using portable electronic devices; and 60% are using automated data acquisition sys- tems. Approximately 40% of the DOTs use two of these col- lection methods; and 36% use some combination of all three methods. About 90% of the state DOTs use spreadsheets to man- age their instrument data; 53% use vendor software; 22% use department developed software; and about 10% use web- based instrumentation software. Nearly 90% of the DOTs use two or more methods to manage their instrument data. The variety of formats used to store and retrieve data is very likely a significant obstacle to the efficient and effective visu- alization of instrumentation data in the DOTs. REMOTE SENSING DATA Nearly 80% of the state DOTs use air photography, and about 60% use LiDAR, topographic data, and satellite images (see Figure 16). About 25% use some form of radar data; for exam- ple, synthetic aperture radar (SAR), interferometric synthetic aperture radar (inSAR), or GPR. Less than 5% report using thematic data. The typical DOT uses three to four different forms of remotely sensed data and images. The primary sources the DOTs use for remotely sensed data are free websites, department generated data, and the USGS. About 60% of the DOTs use these sources. About 25% of the DOTs use commercial and U.S. Department of Agriculture data. HAZARD MITIGATION The state DOT geotechnical leaders report that data visual- ization is used in nearly all aspects of geotechnical hazard mitigation; 60% to 70% report using GDV in identifying, assessing, monitoring, analyzing, designing, and construct- ing hazard mitigation measures. The importance of GDV in the development and implementation of these measures is also widely noted: About 60% of the DOTs reported it has contributed to better identification, assessment, monitor- ing, analysis, and design of geotechnical hazard mitigation measures. FIGURE 15 Geotechnical instruments used. FIGURE 16 Remotely sensed data usage.

23 More than 70% of the DOTs report having used GDV in the successful mitigation of unstable embankments and land- slides; 50% to 55% of the DOTs have used it in the success- ful mitigation of settlement or heave hazards and rockfall hazards; and about 40% of the DOTs have used visualiza- tion in the successful mitigation of sinkhole hazards. Fewer DOTs have had to mitigate other geotechnical hazards, such as slope creep, debris flows, avalanches, and frost heave; but GDV has contributed to such efforts. However, visualization of geotechnical data is not a guar- antee of successful hazard mitigation. About 20% of the state DOT geotechnical leaders report that they have been unsuc- cessful in mitigating one or two types of geotechnical haz- ards; and the unsuccessfully mitigated hazard types are often the same as the successfully mitigated ones. Approximately 70% of the DOTs report that GDV has helped in implementing measures to improve public safety; 35% to 40% believe that visualization has helped improve worker safety and traffic mobility; and 20% believe that visualization has helped improve the speed of implementing hazard mitigation measures. DISASTER AND EXTREME EVENT RESPONSE Visualization of geotechnical data during disaster or extreme event response is limited at many of the state DOT geotech- nical divisions. Only about 30% of the DOTs report having disaster-ready visualization access to such basic geotechnical data as boring logs, geologic maps, and geotechnical reports. In spite of these apparent limitations, about 60% of the DOTs use visualization of geotechnical data in damage assessment, safety analysis, and temporary repair design when respond- ing to geotechnical disasters or extreme events. Between 40% and 50% have used GDV during construction of tem- porary repairs and to maintain public and worker safety; and 20% to 30% have used GDV to facilitate communication, coordination, and traffic control. The relatively widespread use of GDV following disasters or extreme events indicates the value placed on it in such cir- cumstances. This is supported by the large percentage of DOT geotechnical leaders who would find almost all geotechnical data useful during disaster response; about 80% reported that they would find boring logs, geotechnical reports, ground- water data, as-built drawings, geologic maps, and pre-event photographs helpful. LONG-TERM DISASTER RECOVERY Long-term disaster recovery refers to activities undertaken well after the emergency response period, including the planning, design, and construction necessary to remediate the damages and to at least restore the transportation system to its pre- disaster safety and functionality. Long-term recovery often affords a transportation agency the opportunity to build a more resilient system than existed previously. About 87% of the state DOT geotechnical leaders report that they use GDV during the design phase of long-term disaster recovery; 60% to 70% report that they use visualization during assessment, planning, analysis, and construction of long-term remediation measures. The state DOT geotechnical leaders’ responses to this sur- vey question were generally similar to their responses to the same question in the context of geotechnical hazard mitigation. The exception to this generalization is in the design phase; 75% of the DOTs report using GDV in the design of hazard mitigation measures, whereas 87% report using visualization in the design of long-term disaster recovery measures. The responses may imply that geotechnical hazard mitigation and long-term disaster recovery are similar processes that do not have the urgency associated with disaster response, and that consequently, the DOTs have more time to retrieve and visual- ize geotechnical data. The DOTs were asked to indicate how GDV affects their ability to achieve a more efficient and effective recovery from a transportation disaster. About 70% responded that using GDV during long-term recovery from a disaster led to a more economic design and construction process; about 65% of the DOTs reported that visualization contributed to improved public safety; and approximately 60% said that visualization led to a more rapid recovery. Just over 40% of the DOTs noted that visualization led to improved worker safety and about 25% said it improved public communication and recovery of traffic mobility. VISUALIZATION USAGE AND EXPERIENCE The final section of the study survey asked the state DOT geo- technical leaders to do a self-evaluation of their use of and experience with GDV tools in order to gauge the current level of their expertise in applying these tools to solving the hazard mitigation, disaster response, and disaster recovery challenges they face. About 29% of the state DOT geotechnical leaders reported that their organization is a frequent user of GDV tools; approx- imately 37% are occasional users; and about 29% of the DOTs use these tools only rarely. One DOT (2%) never uses these tools, and another did not know its organization’s level of use. The total of about 66% who are frequent or occasional users is generally consistent with the overall level of usage that can be inferred from the responses to other questions in the survey. Among the DOTs that do use GDV tools, just over half consider their organization to be at an entry level of expertise, about one-third are at an intermediate level, and 12% consider their organization to be at an expert level. A previous study of DOT usage of advanced geospatial tools (Olsen et al. 2013) concluded that a higher level of expertise exists in the DOTs

24 than the current study would imply; however, that study sur- vey targeted geospatial and “other relevant contacts” within the DOTs, whereas the current study survey targeted the DOT geotechnical leaders. It is reasonable to assume that a DOT’s geotechnical staff may rely on other staff within the organiza- tion to provide a high level of visualization expertise. The state DOT geotechnical leaders were asked to rank their agreement or disagreement with the following three statements: • GDV improves our ability to mitigate geotechnical hazards. • GDV improves our ability to respond to geotechnical disasters. • GDV improves our ability to achieve long-term recov- ery from geotechnical disasters. The responses were remarkably consistent, with 90% to 93% of the DOTs agreeing or generally agreeing with all three state- ments. This implies that the state DOT geotechnical leaders understand the purpose and value of data visualization even if they may not yet have the tools and expertise to take full advantage of it. VISUALIZATION USERS Almost 90% of the state DOTs report that their geotechnical engineering staff is using GDV tools, and nearly 70% report that their staff geologists use these tools. Between 15% and 25% of the DOTs said that managers, planners, designers, laboratory staff, and other engineering disciplines use GDV tools, but fewer than 5% report that first responders to geo- technical disasters use these tools. Based on the types of software that the state DOT geotech- nical leaders identified as being used in their organizations, it can be assumed that most geotechnical personnel are pro- ficient with one or more applications (e.g., spreadsheet, bor- ing log generator, instrumentation software). However, with the exception of those few state DOTs that use a centralized database, most geotechnical personnel likely do not have easy visual access to all of the available geotechnical data. VISUALIZATION ISSUES Responses to the study survey indicate that usage of a geo- technical data interchange standard within the individual state DOTs is limited. Use of a geotechnical data interchange stan- dard would simplify the collection, processing, and retrieval of geotechnical data for rapid visualization. A data interchange format is a specification of a structured data file based on the open-source Extensible Markup Lan- guage (XML) produced by the World Wide Web Consortium. The human and machine-readable file contains the data in a specified order, a description of the data, and a description of the data sequence within the file. The advantages of a data interchange format are that the data generator knows how to deliver the data; the software developer understands how to read the data; multiple users can read and use the same data; and data quality is maintained at every step. This concept has been successfully applied by the EPA in its Staged Electronic Data Deliverable (SEDD) stan- dard. The land surveyors and architects have used landXML, an XML-based data interchange format, to collect, process, and share surveying and civil engineering data since 2000. At least three XML-based data interchange formats have been proposed for geotechnical data, including: • Data Interchange for Geotechnical and Geoenvironmen- tal Specialists (DIGGS, www.diggsml.com) • Spatial Data Standard for Facilities, Infrastructure, and Environment (SDSFIE, www.sdsfieonline.org) • Association of Geotechnical and Geoenvironmental Specialists (AGS, www.ags.org.uk). The SDSFIE format is a U.S. Department of Defense stan- dard; AGS is a standard developed in the United Kingdom (UK); and DIGGS was developed by a group that includes the FHWA, U.S. Army Corps of Engineers, EPA, USGS, the UK Highway Agency, and 11 state DOTs. Among the issues raised during the study interviews was that reliability and scale are critical considerations for the use of GDV. Using flawed visualization tools or data to make decisions about geotechnical hazard mitigation or disaster response may make the situation worse rather than better. The data underlying any visualization, geotechnical or otherwise, must meet the reliability criteria of completeness, correctness, timeliness, and integrity: • Completeness refers to the percentage of relevant infor- mation contained in the data. For example, one might catalog all of the exposed joints in a rock mass, but the data set may be incomplete because of hidden joints. • Correctness refers to the accuracy and consistency of the data. The depth to groundwater, for example, is typ- ically measured at a few locations to the nearest 0.1 foot with some degree of measurement uncertainty. The few measurements are then projected between locations, adding another level of uncertainty. • Timeliness refers to the age of the data. The passage of time can add uncertainty to the most complete and cor- rect measurements. • Integrity refers to the steps taken to ensure that the data remains complete and correct. The criteria may not require 100% completeness or correct- ness; it is difficult to collect and maintain any type of data,

25 and especially geotechnical data, without some level of uncer- tainty. However, the inherent uncertainty in the data must be understood by the user and, preferably, be displayed in the visualization. For example, it is not uncommon to see error bars or error bands on an x-y plot, but it is rare to see a con- tour plot or LiDAR image with a visual or textual expression of the uncertainty associated with the plot or image. The issues of geotechnical data reliability extend to the software and hardware used to visualize the data. For exam- ple, x-y-z contouring software provides a surface projection based on a finite number of measurement points. A value taken from an arbitrary point on the surface has an associated uncertainty that is a combination of data and projection uncer- tainties. Hardware can also be problematic: For example, if the resolution of the user’s screen or printer is much different than the resolution of the underlying data, the user may “see” distinctions that are not supported by the underlying data or may miss important information. Scale issues arise in GDV because geotechnical data are measured at scales ranging from particle size to satellite images. Although no one is likely to attempt to integrate geo- technical data from the entire scale range into a single visu- alization, it is possible that someone could be interested in, say, site-specific to corridor-level visual integration. In such a case, the user must be aware of problematic factors in larger scale visualizations—e.g., that the level of uncertainty associ- ated with corridor level data may overshadow more precise, site-specific data.