ZC10046_GeoLabs_HALE_FinalReport-Rev1.pdf

April 2010

Zonge Project #10046

GEOPHYSICS INVESTIGATION FINAL REPORT

Haleakala Highway, Haleakala National Park, Island of Maui, Hawaii

April 2010

Zonge Project #10046

FINAL REPORT

GEOPHYSICAL INVESTIGATIONS OF SUBSURFACE LAVA TUBES

AND AIR-FILLED VOIDS AT HALEAKALA HIGHWAY,

HALEAKALA NATIONAL PARK,

ISLAND OF MAUI, HAWAII

Prepared for:

Mr. Robin M. Lim, V.P., P.E.

Geolabs, Inc.

2006 Kalihi St.

Honolulu, Hawaii, 96819 Tel: (800) 841-5064

Mr. Khamis Haramy

FHWA/CFLHD

12300 West Dakota Ave., Suite 210 Lakewood, Colorado 80228

Tel: (720) 963-3521

Submitted by:

Justin Rittgers, Project Geophysicist Zonge Geosciences, Inc.

7711 West 6th Avenue, Suite G Lakewood, Colorado 80214

Tel: (720) 962-4444 Fax: (720) 962-0417

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park

EXECUTIVE SUMMARY

This report presents the findings of a comprehensive geophysical survey carried out in an effort to determine the presence of shallow-subsurface hazards, specifically lava tubes and large air-filled fractures. Shallow subsurface voids beneath roadways have long been a concern for public safety, where such features can propagate or stope toward the surface and collapse unexpectedly beneath the weight of vehicles. Furthermore, subsurface voids often present unforeseen and significant safety concerns and challenges to construction projects requiring the use of heavy maintenance equipment or machinery which may generate dynamic loading of ground materials.

A proposed road improvement project along portions Haleakala Highway within the Haleakala

National Park has prompted a detailed analysis of subsurface conditions in these areas. The park contains protected ecological environments, so geophysical techniques were selected due to their non-destructive / non-invasive nature. Zonge Geosciences, Inc., under contract to Geolabs, Inc. and in close conjunction with the Federal Highways Administration - Central Federal Lands division, was tasked with conducting geophysical investigations along an approximate 4 mile section of Haleakala Highway.

The primary objective of the geophysical survey is to map the presence of lava tubes or other large air-filled voids or features of concern such as utilities and culverts as they may impact design and repaving of the existing roadway. Zonge utilized a combination of multi-antenna / multi-frequency Ground Penetrating Radar (GPR) and total-field magnetics (MAG) to meet this objective.

In general, the GPR data are sufficient for the final interpretation; however, MAG results were correlated to the GPR data and used to verify the locations of cultural features and to substantiate the final interpretations presented herein. Due to the complex nature of the subsurface geology in the areas surveyed, final interpretations utilize a priority rating system that has been developed to help characterize the severity of each anomalous feature identified in the radar data. This rating system assigns a priority of “1” to large size/magnitude and linearly trending anomalies that indicate the presence of a lava tube or large open fracture beneath existing roadways, a “2” for less significant or continuous features identified and a “3” for non-continuous, small and likely insignificant features identified in the radar data.

This report consists of sections presenting background information and site conditions / site geology, overviews of the two geophysical methods employed, data acquisition parameters and techniques, data processing and analysis, results and interpretation, and final discussion of conclusions for the investigation.

ACKNOWLEDGEMENTS

All Zonge personnel involved to any degree in the preparation of this report would like to express their most sincere appreciation to Mr. Robin M. Lim of Geolabs, Inc., Mr. Howard Forbes of the

Haleakala National Park and Mr. Matt Ambroziak of the FHWA for their logistical and technical assistance. We would also like to extend special thanks to Mr. Khamis Haramy of the FHWA for his guidance and support throughout the course of the field efforts of this investigation and for his technical review throughout the preparation of this report.

TABLE OF CONTENTS

1.0 Problem Statement

2.0 Introduction

3.0 Survey Locations and Conditions

3.1 General

3.2 Haleakala Highway

4.0 Technical Approach

4.1 Method Description

4.1.1 Ground Penetrating Radar

4.1.2 Total Field Magnetics

4.2 Data Acquisition

4.2.1. Ground Penetrating Radar

4.2.2. Total Field Magnetics

4.3 Data Processing and Analysis

4.3.1. Ground Penetrating Radar

4.3.2. Total Field Magnetics

4.3.3. Review of Results from Previous Work Performed

5.0 Results and Interpretations

5.1 General

5.1.1 Ground Penetrating Radar

5.1.2 Total Field Magnetics

5.2 Haleakala Highway

6.0 Summary and Conclusions

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 1

1.0 PROBLEM STATEMENT

Subsurface voids beneath roadways have long been a concern for public safety, where such features can propagate or stope toward the surface and collapse unexpectedly beneath the weight of vehicles. Furthermore, subsurface voids often present unforeseen and significant safety concerns and challenges to construction projects requiring the use of heavy maintenance equipment or machinery which may generate dynamic loading of ground materials.

A large-scale road improvement project is proposed for the main park road, Haleakala Highway, within Haleakala National Park which is a located on the northwest flank of the now dormant shield volcano, Haleakala. Due to the geologic setting, the park area is extremely prone to geologic hazards including large open fractures and air-filled lava tubes that have previously been encountered in road-cuts along Haleakala Highway, so concern about large and shallow subsurface voids along the proposed project alignment is a serious one. Geophysical surveys were carried out in an effort to help detect such shallow voids beneath Haleakala Highway and aid in the planning of safe road improvement efforts.

The geophysical surveys were carried out in sensitive ecological environments using non-destructive / non-invasive techniques in an effort to minimize environmental impacts of the geotechnical investigation portion of the project and to help protect endangered endemic species of flora and fauna.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 2

2.0 INTRODUCTION

Zonge Geosciences, Inc. (Zonge), under a single contract to Geolabs, Inc. (Geolabs), acquired geophysical data for a road improvement project located along within Haleakala National Park

(Figure 1). Data was collected along an approximate 4 mile section of Haleakala Highway. Field work was conducted under Zonge project 10046 from March 29th to March 31st, 2010.

The main objective of the geophysical investigation is to:

Evaluate shallow-subsurface conditions beneath the existing roadway and map the locations of anomalous features of interest/concern such as lava tubes or open tension fractures in a manner that allows the separate identification of such features from that of culverts and utilities crossing beneath the roadway.

To meet the objectives of the survey a combination of two geophysical techniques was deployed, where Zonge utilized multi-antenna / multi-frequency Ground Penetrating Radar (GPR) and total-field magnetics (MAG). GPR was utilized to image lithology and features that exhibit the characteristics of lava tubes and other large/shallow voids of concern. MAG results were correlated to the GPR data and used to verify the locations of culture (culverts and utility lines/vaults) and to substantiate the final interpretation of features imaged in GPR the data and presented herein.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 3

3.0 SURVEY LOCATIONS AND CONDITIONS

3.1 GENERAL

The project area is situated within Haleakala National Park. Haleakala Highway is generally located on the northwest flank of the now dormant shield volcano, Haleakala (Figure 1). The near surface geology throughout the survey area is mainly comprised of a volcanic ash layer that varies in thickness (0 to greater than 15 feet thick in most locations) on top of relatively young basaltic lava flows. Younger pāhoehoe and 'A'ā flows are present on top of this ash layer in various areas. Typical problematic geologic features found in this region include open tension fractures, open ring fractures, air-filled blisters and air-filled lava tubes.

All of the GPR and MAG surveys were conducted over paved surfaces along Haleakala Highway. Weather was mostly sunny (as seen in the inset photo) with consistently heavy wind throughout the course of the survey. Overall, weather did not pose any issues with the project. During data acquisition, Haleakala Highway had a consistent level of moderate traffic due to park visitors and tour busses, making traffic control coordination and data acquisition and quality control an integral portion of the field effort.

3.2 HALEAKALA HIGHWAY.

The Haleakala Highway project area extends approximately 4 miles along the open portion of the existing two-lane and undivided road from mile-marker 11 located near the park’s visitor center to mile-marker 15 located approximately half-way up the mountain (Figure 2). To aid in the accurate positioning of data, road stations 100+00 through 301+00 were located and painted every 50 feet by surveyors along the majority of the roadway segment surveyed with GPR and MAG. The portion of roadway surveyed with GPR and MAG corresponds approximately from road station 92+40 to 307+20.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 4

FIGURE 1 – Map identifying general location of the Haleakala Highway, Haleakala National Park project site. Project area is defined by the red box and is also shown on the inset detail. Images taken in April 2010 from Google Earth.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 5

FIGURE 2 – Haleakala Highway survey data coverage (in red) for GPR and MAG data collected from approximate road stations 92+40 to 307+20. Road stations that were marked in the field are shown here as blue triangles along the majority of the survey limits. Background geotiff satellite image obtained using Dapple, April, 2010.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 6

4.0 TECHNICAL APPROACH

4.1 METHOD DESCRIPTION

This section is intended to describe the general theory for the two methods used and to provide information regarding data acquisition parameters and the survey logistics in detail. Additional details on the theory of methods are available on our website: www.1800GEOPHYSICS.com.

4.1.1 Ground Penetrating Radar

GPR is a geophysical method that can use a range of frequencies of electromagnetic (EM) energy (typically on the order of tens of kilohertz to a few gigahertz depending on the system used) to image interfaces within the subsurface. GPR systems typically use a single frequency and a single transmitter/receiver antenna pair, however, modern state-of-the-art systems such as that used for the

Haleakala Highway project utilize multiple frequencies and multiple antennas for more comprehensive data coverage and improved resolution while maintaining depth of investigation.

GPR systems repeatedly create an electromagnetic pulse which propagates in all directions away from the transmitter antenna. The majority of this energy propagates downward as a plane-wave into the material beneath the system where the impulse encounters interfaces between materials. The interface is defined by contrasting dielectric constants (a ratio of two physical properties inherent to all materials or solutions - electrical permittivity and magnetic permeability). At each interface, a portion of the energy is reflected back towards the system where the reflection is recorded by the receiver antenna, some measurable time after the transmitter pulse. The remaining portion of energy propagates through the interface and continues on to greater depths/distances until the next interface is reached. This reflection/transmission process is repeated until all of the energy is reflected or attenuated.

The ratio of dielectric constants for two materials gives the reflection coefficient of that interface.

The value of this ratio dictates the polarity of reflected wavelets. The polarity of wavelets that have reflected off of specific types of targets such as air-filed voids or metal culverts can be predicted and can help to constrain the interpretation of such features imaged.

The depth of investigation of a GPR survey is influenced by several things, primarily by the electrical conductivity of materials being investigated. Conductive materials (i.e., soils with high clay content or soils saturated in conductive fluids such as brine water) readily attenuate EM energy and limit the depths to which useful reflections can be detected/imaged. The depth of investigation can be increased by using a system with a lower center frequency; however the resolution of the resulting data inherently is diminished. The GPR system used for the Haleakala Highway project is able to maximize depth of investigation while maintaining resolution by simultaneously utilizing two frequencies.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 7

4.1.2 Total Field Magnetics

Total field magnetometry surveys employ magnetometers to measure the magnitude of Earth’s magnetic field and any disturbances and variations to that regional field caused by various sources including local geology, ferrous objects, buildings and electrical currents within nearby transmission lines, radio towers or high-power ignition systems. Magnetometer sensors are extremely sensitive and can often be affected by electrical storms, solar wind storms and passing vehicles or even people wearing clothing with ferrous metal (buttons and buckles). Therefore, it is very important that the operator is aware of local sources of noise and is demagnetized by removing all metal from their person before acquiring data with a magnetometer.

The detection capability of a given magnetometer generally depends on the size (and ferrous content depending on target type) of the causative body and the level of ‘noise’ in a given survey area. As an example, a 55-gallon drum is easily detected at a depth of 10 feet if the interference due to other bodies or buildings is minimal. Recorded values are plotted as line graphs or as 2D plan-view maps with a color scale, and anomalous highs and lows are interpreted to help determine their source, source depth and occasionally the size/orientation of that source.

4.2 DATA ACQUISITION

For the GPR and MAG surveys, the ability to pause each system during acquisition allowed for traffic to regularly pass the survey crew and avoided any negative impacts on data quality for either method due to passing vehicles in near proximity to sensors. Nearby vehicles were not detected with the GPR system do to effective shielding built into the antennas, yet the system was still paused for passing of vehicles. The MAG system is very sensitive to these and other objects however, so the MAG system was kept at least 150 feet ahead of the chase-car and was paused while passing vehicles approached and receded.

In an effort to maintain manageable-sized data files and avoid the loss of significant amounts of data in the event of a system crash, Haleakala Highway was split into 20 line segments of arbitrary length (usually dictated by road stations painted on the road and site conditions). These segment start/end points were used for both GPR and MAG surveys. Table 1 presents the road stations that correspond to the start / end for each line segment for the Haleakala Highway survey.

4.2.1. Ground Penetrating Radar

GPR data were acquired with a multi-antenna, dual-frequency GPR system manufactured by Ingegneria Dei Sistemi, S.p.A. (IDS) of Pisa, Italy. The RIS MF Hi Mod #4 system is equipped with a bank of four dual-frequency antennas, each of which contain 200 MHz and four 600 MHz antennas, in order to image both deep and shallow features, respectively (shown in the inset image on the next page). This GPR system was selected for this investigation for two main reasons: First, the system’s ability to simultaneously acquire two independent frequency sets of GPR data representing two depths of investigation/resolutions; and second, the system’s ability to collect several profiles of data simultaneously with a single pass down a given traffic lane, generating a wide swath of data

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 8 coverage rather than a single profile. Lastly, the IDS system offers useful 3D features of the integrated IDS software (GRED 3D) and lends the ability to tie features beneath the road across both lanes, ultimately helping to identify linearly trending anomalies indicative of the presence of lava tubes and other similar subsurface features.

The inset image, taken at the Haleakala Highway project area, shows the Hi Mod #4 with four antenna ‘boxes’ as configured for both projects.

For the GPR survey, data was collected along the center of each traffic lane from mile post 11 and mile post 15 corresponding to road stations 92+40 to 307+20 (see Figure

2). The GPR system was pushed by hand in the downhill traffic lane, and was pulled by a long nylon strap behind a field vehicle in the uphill traffic lane. This approach to data acquisition allowed for nearly full coverage of the entire width of traffic lanes surveyed, where traffic lanes are typically less than ten feet wide and the system records a swath of data approximately six feet wide (see inset photo). Flaggers and a ‘chase’ car identified with lights, and signage were used for safety and traffic control purposes.

Positional data were collected using an integrated wheel-odometer system built into the GPR system that actively encodes distance along each GPR record. Supplemental positional data were also acquired with the use of a Real Time Kinematics corrected Differential Global Positioning

System (RTK-DGPS) and with the use of fiducials which were manually ‘tagged’ in the data while beside notable landmarks and features of interest along the roadway during data collection. GPS data were acquired using Trimble 5800 system for both the rover and base station units. The combined positioning approach allowed for the accurate integration of GPR and MAG data sets and location of identified features (either geologic or cultural). Final decisions regarding positioning of the data were made in the field by the Project Geophysicist based on experience, site conditions, and compliance with Park regulations.

4.2.2. Total Field Magnetics

The MAG survey was conducted using a Geometrics G858 cesium vapor magnetometer system with a single sensor mounted on a handheld counterbalanced staff. The system was configured to incorporate integrated RTK GPS (shown in the inset image on the next page) during data acquisition for accurate data positioning. Total-field MAG data were acquired in the center of both traffic lanes along Haleakala Highway from mile post 11 to mile post 15 corresponding to road stations 92+40 to 307+20 (see Figure 2). Data were acquired nominally at 1-foot in–line spacing (every second).

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 9

For the MAG portion of the surveys, positional data were also acquired with the use of a Real Time Kinematics corrected Differential Global Positioning

System (RTK-DGPS) and with the use of fiducials which were manually ‘tagged’ in the data while beside notable landmarks and features of interest along the roadway during data collection. Each positioning approach allowed for the accurate integration with GPR data and placement of identified features (either geologic or cultural) onto plan and profile sheets.

Final decisions regarding positioning of the data were made in the field by the Project Geophysicist based on experience, site conditions, and compliance with Park regulations.

Haleakala Highway.

File/Segment Start Station End Station

1 92+40 (mm11) 104+00 2 104+00 116+00 3 116+00 126+00 4 126+00 137+00 5 137+00 148+00 6 148+00 158+00 7 158+00 169+00 8 169+00 179+00 9 179+00 189+00

10 189+00 201+00 11 201+00 212+00 12 212+00 222+00 13 222+00 232+00 14 232+00 243+00 15 243+00 253+00 16 253+00 265+00 17 265+00 275+00 18 275+00 285+00 19 285+00 295+00 20 295+00 307+20 (mm15)

TABLE 1 – Start/End road stations for line segments along Haleakala Highway.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 10

4.3 DATA PROCESSING AND ANALYSIS

4.3.1. Ground Penetrating Radar

Normally, GPR data is viewed in the field and signal parameters adjusted to best fit the needs of the investigation. Often, raw signal data is of suitable quality such that significant data processing may not be required, and field determination and interpretation may be made based on raw collected data. This was often the case during this investigation; however, processing was still carried out to ensure the best possible results. Data were reduced using the integrated IDS GRED, Version

02.01.008 for Windows. GPR analysis for both projects consisted of the following:

• Picking the surface arrival as "time-zero" and adjusting each radar trace to a common surface arrival – this is critical for time-depth conversion.

• Applying separate band-pass filters to the 200-MHz and 600-MHz data

• Background removal to remove multiples and ringing caused by subsurface layering and the surface reflection

• Applying a smoothed, multi-point gain function to the records to highlight subtle anomalies at depth

• Time-Depth conversion using a dielectric constant of 10 and a calculated velocity of 0.1 meters per nanosecond propagation speed (this velocity was applicable for all data collected)

• Rescaling of line segment distances (rubber-sheeting) via registering fiducials and feature locations to Plan and Profile sheets.

Data analysis involved viewing each of the eight parallel radargrams per lane (divided into sections for ease of viewing) and highlighting observed anomalies. In general, lava tube anomalies were expected to be characterized by clear hyperbolic diffractions similar to those observed over concrete pipes or culverts; areas where geologic bedding appeared to be discontinuous and collapsed;

and areas where the GPR energy reverberates between two reflectors (a phenomenon referred to as

“ringing”).

Anomalies that were isolated beneath the travel path of a particular antenna housing (i.e. – not observed in a parallel profile) were discounted as being a target of concern. Clear, broad anomalies extending the breadth of a lane or the entire roadway were given higher priority as being potentially caused by lava tubes or similar structures.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 11

4.3.2. Total Field Magnetics

MAG data were first exported from the data logger using the program MagMap2000. Next, data drop-outs and spikes were removed and linear interpolations were performed prior to exporting a text file containing all recordings along with the associated positional information. This data was then imported to Oasis Montaj, where further positional smoothing was performed for GPS data with low positional accuracy (as a result of tall trees with thick canopy that block satellite and base station communications with the GPS rover unit). Finally, a low pass filter was performed on the data to remove regional trends in the data and to help highlight short-wavelength features in the MAG data that are expected from culverts, utilities and even large air-filled voids in the right conditions.

The filtered data were then gridded using the Kriging algorithm and plotted as a 2D color-contour map in plan-view to visually identify MAG anomalies. All MAG anomalies identified were cross-checked with the GPR data. Moreover, all anomalous GPR features of interest such as hyperbolas imaged at possible utilities or possible voids were cross-checked in the MAG data to help constrain final interpretations presented herein for each project.

4.3.3. Review of Results from Previous Work Performed

In addition to the original scope of work described within the contract agreement for this project, a review of results from a previously performed GPR survey at the Haleakala Highway project site was performed by Zonge. This previous work was performed by another geophysical services company several years ago, and hard copies of those results were furnished to Zonge by the FHWA.

This previous work was performed using a single channel and single frequency system, where data was acquired along a single profile in both the uphill and downhill traffic lanes of the Highway.

While the general nature of these previously collected data match Zonge’s data (e.g., suggest highly variable geology exhibiting areas of highly fractured rock), the data quality from this previous work was questionable for several reasons. First, there are long segments of data exhibiting high levels of attenuation in one profile but not on the adjacent profile (suggesting instrument power malfunctions as apposed to attenuative geology). Second, there is strong banding and processing artifacts still present in the profiles. Lastly, the positioning of data is extremely difficult to determine making it impossible to correlate the two radar surveys. For example, each radar profile is labeled as to indicate that data coverage starts and ends at mile posts, suggesting that each figure presents a mile (5280 ft) of data. Yet, each profile is greatly different in length and not even approximately a mile in length (i.e., 4300 ft, 5000 ft, 5350 ft, 5500ft).

Lastly, it is very difficult to determine the methodology or criteria used in that company’s interpretation. The majority of ‘anomaly picks’ appear to be that of culverts or utilities crossing under the survey area. Furthermore, some features were selected as possible lava voids that appeared to not trend beneath the roadway while other anomalous features were not selected or labeled otherwise. As a result, anomalies identified in these older results can not be located and verified or discounted in Zonge’s GPR results presented herein.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 12

5.0 RESULTS AND INTERPRETATIONS

5.1 GENERAL

Results of the geophysical investigations are first discussed in general terms, providing example figures and discussing the format of final presentation of results. Example figures present radargrams that contain highlighted features which correspond to various subsurface culture and anomalous geologic structures mapped along the segment of roadway surveyed. All GPR and MAG data have been correlated and evaluated to develop final interpretations based on both data sets.

These final interpretations include a priority/confidence rating, or “anomaly classification,” which assigns a “1” a “2” or a “3” for each anomalous feature identified:

• A classification of “1” is given to a large amplitude and linearly-trending feature that suggests the possible presence of a significant and shallow void that may be of concern to design and construction engineers.

• A classification of “2” is assigned to features that are smaller in size/extent, do not suggest the presence of a large air-filled void or are deeper and likely not of concern to design and construction engineers.

• A classification of “3” is assigned to features that are small in size and do not suggest the presence of a significant air-filled void, but are likely geologic bedding or small fractures likely not of concern to design and construction engineers.

5.1.1 Ground Penetrating Radar

Overall ground penetrating radar (GPR) data quality for this investigation was very good. The GPR system deployed used 200- and 600-MHz antennas located within the same housing to maximize our ability to resolve small targets in the shallower subsurface, as well as larger targets at depth (as seen in Figure 8 which shows example data collected at separate nearby project on the island of Hawaii).

GPR is analogous to 2D seismic reflection where a series of traces along a given survey line produce a two-dimensional cross-section referred to as a radargram. Interfaces and objects can be detected and mapped with a series of pulses/traces. Point reflectors, such as small cobbles, fractures, voids and utilities oriented perpendicular to a survey line, produce diffraction patterns that are hyperbolic in shape (see Figure 9). Geologic horizons are often mapped as continuous horizons within a radargram, where dark and light-colored bands represent the positive and negative polarity of wavelets or series of wavelets or “wiggle-traces”. Continuous horizons of a given polarity are formed as energy reflects off of continuous interfaces between layers with contrasting dielectric constants (see Figure 10).

These horizons allow for geological layers, structures or features in the subsurface such as cross-bedding within a sand dune to be imaged.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 13

FIGURE 8 - An example GPR profile highlighting the system’s ability to resolve both small/shallow and deep targets from a separate yet nearby project site located on the island of Hawaii. This figure demonstrates the ability of the 600 MHz antenna to resolve individual bars at the base of a buried reinforced slab (located at the Hawaii Volcanoes National Park entrance station on Crater Rim Dr.).

FIGURE 9 - Example GPR profile highlighting a typical hyperbola caused by metal culverts which were easily identified by their high amplitude and symmetry along Haleakala Highway.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 14

FIGURE 10 - Example GPR profile showing a typical signature caused by geologic interfaces which can be seen as continuous horizons that buldge and dip along Haleakala Highway.

Figure 11 presents a final example of radar data collected over a culvert near mile marker 15 as a

3D volume, where linear interpolation has been performed between adjacent 2D radargrams obtained with the multi-antenna GPR system. The 3D volume has been sliced at various depths to demonstrate the linearity of anomalies in adjacent radar grams typically seen from features such as culverts and fractures throughout the project. Furthermore, Figure 11 demonstrates a capability of the GRED processing software which allows the user to render the variable size, depth, location and orientation of pipes or other features in 3D. Rendering of culverts or other features as seen in Figure

11 is based on hyperbolas that are identified and ‘picked’ by the user in multiple adjacent 2D radargrams. 3D visualization and interpretation was a very valuable tool in the development of final interpretations presented herein.

While GPR data quality was excellent for the Haleakala Highway geophysical investigation, its ability to resolve small targets such as air-filled fractures and tiny air-pockets actually made interpretation of the data very time-intensive due to the numerous anomalies identified. Horizontally smoothing or stacking GPR traces may have removed some of the smaller (distracting) features observed in the GPR data but may have introduced artifacts or partially masked anomalous structures which may be of interest. The dense data collected during this investigation allowed the data interpreters to discount anomalies that did not occur on immediately adjacent GPR profiles. Without this dense GPR data coverage, many, many, more anomalies would have been highlighted as anomalies of possible concern.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 15

FIGURE 11 – Several views of an example 3D volume of GPR data, sliced at various depths to reveal a typical hyperbolic and linearly-trending feature in the radar data caused by metal culverts along Haleakala Highway. The figure also shows the interpreted location/size/orientation of the causative culvert rendered in 3D (shown as a dipping red tube) generated with the use of the GRED software developed by IDS.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 16

5.1.2 Total Field Magnetics

While data quality for the MAG survey is very good due to few sources of signal interference, the highly magnetic nature of the local geology along Haleakala Highway is such that there are large amplitude fluctuations in the local magnetic field over short distances (referred to as geologic noise).

Example MAG figures present color contour maps where blues and greens (i.e., cool colors) are relative lows in the total magnetic field and yellows and reds (i.e., warm colors) are relative highs in the total magnetic field. These fluctuations, as seen in Figures 12 and 13, are likely a result of non-homogeneous distributions of magnetic minerals such as magnetite and hematite within the variable basaltic lava flows throughout the areas surveyed. The highly variable characteristic of the magnetic field along each roadway demonstrates why the identification of subsurface voids impossible based on MAG data alone. A typical MAG signature expected from a subsurface void would be similar in spatial extent yet much smaller in amplitude in comparison to the geologic noise seen along each roadway surveyed. While electric utilities are easily identified in the MAG data, the signatures of some culverts are often obscured as well by geologic noise in the MAG data. Although, most culverts and other utilities were verifiable in the MAG plots for specific locations/features identified in the GPR data (see example in Figure 13). With the addition of MAG data, interpreters were confident of the separation of GPR features due to culture from those due to geology/structure.

FIGURE 12 - An example showing MAG data collected along Haleakala Highway, highlighting typical signatures caused by the many metal culverts along the surveyed portion of the roadway.

Most of these features are easily identified by their high amplitudes which stand out from background levels, however geologic noise in the area can exhibit similar characteristics, as seen in the figure.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 17

FIGURE 13 - An overview of MAG data collected near mile post 15 showing the highly variable nature of the total magnetic field typical for Haleakala Highway and surrounding areas due to geologic noise.

5.2 HALEAKALA HIGHWAY

Results for the Haleakala Highway project are presented here in Figures 14 through 16 as radargrams which show each class 1 anomaly identified for the project, followed by lists of all anomalies (class 1, 2 and 3) organized by classification (see Tables 2 through 4). These tables provide the locations of each anomaly, an approximate depth to the top of each anomaly, an approximate width of the feature identified (for class 1 and class 2 anomalies only), and a brief description/interpretation of each anomalous feature. Normally, these features would also be plotted on plan and profile sheets for graphical presentation of anomaly locations, however Plan and Profile sheets were not available at the time this report was drafted.

Overall, there were no extensively large voids detected by the geophysical surveys (i.e., air-filled voids much larger in diameter than depth of burial) detected within the upper 5 feet beneath existing roadways. However, the overall nature of the data indicates that the entire project area exhibits highly fractured and variable geology that may contain discontinuous (discrete) voids or series of thin air-filled gaps between geologic layers that could pose a risk of non-catastrophic collapse if overloaded.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 18

FIGURE 14 – Class 1 anomaly identified 37ft from the start of File 3, Haleakala Highway (anomaly indicated with purple crosshair and circle). GPR Channels 1 and 4 shown here for comparison.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 19

FIGURE 15 – Class 1 anomaly identified 49ft from the start of File 15, Haleakala Highway (top of anomaly indicated with purple crosshair).

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 20

FIGURE 16 – Class 1 anomaly identified 814ft from the start of File 17, Haleakala Highway (anomaly indicated with purple crosshair and red boxes).

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 21

Haleakala (File distances increase up-station) File # Distance From Start (ft) Depth (ft) Approx. Width (ft) Station (ft) Confidence/Priority Description

3 37 8 1.5 11637 1 Strong hyperbola, clear and sized like culvert, but nothing in MAG, trends across road (Lava Tube?)

15 49 4 & 7 2 & 2 24349 1

A6 - Diffractions from collapse above a large, positive hyperbola in the 200MHz data at 2m depth, same location (Lava Tube?)

17 814 9.9 7 27314 1 Broad, deep but faint hyperbola (Lava Tube or infilled void?)

TABLE 2 – All Class 1 anomalies identified for Haleakala Highway.

(File distances increase up-station)

File # Distance From Start (ft) Depth (ft) Approx. Width (ft) Station (ft) Confidence/Priority Description 2 182 2.5 1 10582 2 Hyperbola across channels 2-4; (culvert based on MAG?)

4 912 5.3 1 13512 2 Hyperbola only seen on channels 3&4 (nothing in MAG) 9 158 4.7 2 18058 2 Negative hyperbola, channels 2&3 only, however.

10 482 3-6.5 NA 19377-19387 2 Lots of fractures 10 946 5.7 1.5 19846 2 Hyperbola, negative polarity. Mainly on uphill-side of road.

13 636 3.9 2 22836 2 A5 - only observed on channels 1-2, so classified as 2.

However, polarity is right, shape is right for air-filled void

14 51 4.4 1.5 23251 2 Ringing hyperbola. Best seen in center of up-hill lane 14 489 3.5 1 23689 3 Diffractions either from collapse or tight fracture zone

15 141 6.6 2 24441 2 Broad hyperbola on channels 1-4. Most of the interpreted fractures are more vertical, this is wider

15 799 3.7 1 25099 2 Likely fractures, but very bright. Only seen on channel 1-3.

Channel 4, 200MHz data, looks more like collapse

16 556 7.7 2 25856 2 A7: Good sized anomaly, could be an air filled void. However, in a pretty fractured zone

16 575 5.7 2 25875 2 A9 - Good sized anomaly, not as deep. Very fractured area.

Only seen on uphill-side of road

16 585 5.9 1 25885 2 A10 - smaller and weaker amplitude but trends across road.

TABLE 3 – All Class 2 anomalies identified for Haleakala Highway.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 22

(File distances increase up-station) File # Distance From Start (ft) Depth (ft) Station (ft) Confidence/Priority Description

3 548 3.7 12148 3 Small Culvert or small void?

5 352 3.5 14052 3 Very small, polarity change and diffractions.

5 914 8.4 14614 3 Weak, deep hyperbola channel 1-3 (1 is weak).

6 445 3.8 15245 3 Fracture across ch1-3, weak in 4 6 525 5.7 15325 3 Fracture across ch1-3, weak in 4 6 846 9.8 15646 3 Culvert 7 64 3.4 15864 3 Fracture? Only on Channels 1-3. In a very fractured area 9 903 3.5 18803 3 Box Culvert or repair? Likely not a void.

11 109 1.6 20209 3 Small diffraction, negative hyperbola; Strongest response in center of uphill-lane 11 117 1.6 20217 3 Small diffraction, negative hyperbola 13 180 6.7 22380 3 Weak hyperbola, only channels 1-3. Ground very fractured in proximity - likely fracture 13 729 4.9 22929 3 Fracture zone? Possible collapse. Doesn't look like air-filled void 14 489 3.5 23689 3 Diffractions either from collapse or tight fracture zone 16 565 0 25865 3 A8: don't really see anything there, so class 3. Highly fractured zone 17 881 3.5 27381 3 Fracture? Hyperbola is incomplete, jumbled, but across Ch1-3 17 898 3.4 27398 3 Fracture? Hyperbola is incomplete, jumbled, but across Ch1-3 17 908 4.2 27408 3 Fracture? Hyperbola is incomplete, jumbled, but across Ch1-3 19 603 1.8 29103 3 Long fracture (near vertical, extends very deep?)

TABLE 4– All Class 3 anomalies identified for Haleakala Highway.

Final Report – Geolabs, Inc – Haleakala Highway, Haleakala Nation Park 23

6.0 SUMMARY AND CONCLUSIONS

Overall, the geophysical surveys conducted along Haleakala Highway within Haleakala National

Park where successful in mapping subsurface geology and detecting and differentiating between culture (i.e., utilities and culverts) and air-filled voids beneath existing roadways.

In general no anomalies indicating the presence of voids of significant concern (i.e., air-filled voids/lava tubes larger in diameter than depth of burial within the upper five feet of the subsurface) were detected beneath existing roadways or parking lots surveyed for either project. While no singular voids of significant concern were detected, both project areas exhibit highly fractured and variable geology that could pose the risk of accumulative collapse (collapse of several thin air-filled gaps between layers) if overloaded. As with any project, the utmost caution should be taken while working or excavating with heavy machinery in either of these project areas. Lastly, the combination of GPR and MAG data allowed for a high level of confidence in the results and interpretations presented herein for each project.