ASSESSMENT OF TERMITE DAMAGE IN WOOD USING DIGITAL RADIOSCOPY

Transcription

1 ASSESSMENT OF TERMITE DAMAGE IN WOOD USING DIGITAL RADIOSCOPY Ronald W. Anthony Anthony & Associates, Inc., Fort Collins, Colorado USA Abstract Insects and fungal decay damage historic wood fabric world-wide. Assessing the extent of damage without causing further harm has been difficult. Most methods for investigating wood deterioration are quasi-nondestructive in that they require some penetration of the wood to gather data on the internal condition. Digital radioscopy has been shown to be effective in identifying hidden conditions but, until recently, has not been useful in quantifying the extent of deterioration. Research done by the author to quantify the loss of material was applied in to the wood members of the Mission San Miguel Arcangel, located in Southern California. The ceiling decking at Mission San Miguel consists primarily of nominal 2.5-cm thick boards throughout the nave and sacristy ceilings. These ceiling boards have extensive termite damage. Using techniques developed for x-ray examination of historic wood fabric, two ceiling boards were examined by digital radioscopy (one from the nave and one from the sacristy) and the loss of section was calculated. The damage was found to be typically a loss of approximately 50 percent of the wood material and was most severe near the wood surface. Keywords Digital radioscopy, x-ray, termite, wood deterioration, nondestructive testing 1. INTRODUCTION In the interest of saving as much historic fabric as possible and not altering or scarring historic materials as a result of investigative probes, architectural conservators look to nondestructive testing methods for the evaluation and identification of materials, conditions and alterations made to structures over time. As the answers to these questions are typically concealed in a historic structure, non-destructive techniques of investigation are a necessity, but few are available. Digital radioscopy is one of those techniques, but its use has not been wide spread. This article highlights research done by the author to improve the availability and use of radioscopy in historic structure investigation. 1071

2 2. EQUIPMENT Digital radioscopy makes use of portable, low-energy x-ray imaging systems. One imaging system used in this research is the EPIX Digital Imaging System manufactured by Logos Imaging. This system is composed of the EPIX imaging plates, the EPIX scanner, and a laptop with software to import and save the scanned image. The imaging plates are reusable, photo-stimulatable phosphor imaging surfaces, either 20 cm by 25 cm, or 20 cm by 43 cm. The plates are composed of flexible plastic sheets coated with a very thin layer of tiny storage phosphor crystals bonded together. X-ray images are created on these imaging plates as the phosphor crystals capture the energy of x-rays passing through the object of study. The energy stored in the crystals is then released by the process of scanning. Another system used in this research, the RTR-4 TM portable digital x-ray imaging system manufactured by SAIC (Science Applications International Corporation), is a fully digital imaging system that includes its own image modification tools. The system is composed of a control unit (a laptop computer), the imager, and cables to connect the imager and the controller. The imager is a compact, solid state camera with a 20-cm by 27-cm field of view. It is typically mounted on a tripod or placed directly against the surface of the object of interest, opposite from the source. As with all portable x-ray systems, access to both sides of the object of interest is required. The imager s electro-optical system captures the images and transmits them to the control unit, where they are stored as TIF images. The x-ray generating source for both systems is an XR200 x-ray source, manufactured by Golden Engineering, Inc. This source is a pulsed source, producing x-ray pulses of short duration (60 nanoseconds each) and minimal dose (3.1 milliroentgens for each pulse at a distance of 30 cm from the front of the unit), with energy up to 150 kv. The aperture size is 3 mm, and the beam produced by this source has a 40 o beam angle, so that x-rays taken about 60 cm from the source have a spread equal to the width of the imager. For each x-ray, the number of pulses can be set from One to two pulses are required to penetrate paper and four to twenty are typically used to penetrate most wood walls. 3. MODIFICATION OF DIGITAL RADIOGRAPHS TO QUANTIFY DETERIORATION Familiar to users of digital photography is the ease with which images can be edited for special effects. Digital radiographs afford the same opportunities in that, like digital photographs, information is stored as some gradient of color value in an individual cell (i.e., pixel) positioned adjacent to other such cells. This assignment of color value and position results in a bitmap. One advantage of a bitmap image is that it is possible to analyze and control the image on a pixel-by-pixel basis. Digital radiographs are records of the total and cumulative amount of radiation passing onto the imager at a specific point. Whether captured on a plate or a digital imager, this information is recorded as a color along a continuous gradient. A simple grayscale gradient records total density where black is high and white is low. Post-processing can be used to enhance the distinction by emphasizing various parts of the gradient. The resulting graphic output can be enhanced grayscale, or a color image, used for the purpose of improving visual analysis. A description of the post-processing technique is provided in reference [1]. 1072

3 Identification of the presence of decay in wood has proved to be straightforward, but quantifying it was more challenging. A typical radiograph of a timber structure shows higher density materials, such solid wood or nails, as darker and lower density, decayed wood as lighter, as expected. Decayed wood is also characterized by a mottled appearance and the lack of clear growth rings. While these visual characteristics are relatively easy to identify, determining the extent of remaining sound wood was not as easily determined. In an effort to try to quantify loss of section due to decay or insect damage, it was decided to begin with a simple case, a timber cut into a stepped block with known thicknesses at each step. As such, this was a situation with discrete boundaries separating the sections, which ranged in thickness from 6.25 cm to 25 cm. Radiographs were taken using a constant sourceto-object distance of 60 cm. Successive radiographs were taken with different x-ray pulses ranging from 16 to 100. To find a user-friendly method to quantify the range of tones in these radiographs, we elected to use a readily available software package, Adobe Photoshop Elements. This package (along with most other photographic-enhancement software) has a histogram function that identifies the range of tones in a radiograph by graphing the number of pixels at each color intensity level (which are grey tones only for the radiograph). The number of pixels is represented on the vertical axis while the horizontal axis ranges from the darkest values (at zero) to the brightest values (255 for this package). One of the advantages of using the histogram function to quantify deterioration in wood is that it can be used with the Rectangular Marquee Tool to select a portion of the image to analyze. Without selecting a portion of the radiograph, the histogram function displays the histogram of the entire radiograph. For smaller areas, each section can be outlined with the Rectangular Marquee Tool, and then the histogram function can be employed. For each radiograph, this was done, using the section at 100, 75, 50 and 25 percent of the initial cross section, and taking the mean and median values for each histogram (Figure 1). This approach resulted in excellent correlation (R 2 values were all 0.99 or above) between the mean value and the percent of remaining cross section at each pulse level. However, this information was useful only for one species, using one particular setup and only for these thicknesses. It was felt that in a typical field situation, a more realistic method to determine loss of section might include an identification of intact wood versus decayed wood and a comparison between them. To this end, the ratio of the median histogram value for sound wood (the 25-cm thick section) to the median histogram value for other thicknesses was calculated, and plotted against the percent of remaining section. This created a graph in which all the functions collapsed into a single function (Figure 2). The ability to have a single function is important because it shows that the number of pulses used when taking an x-ray are not a factor when trying to determine loss of section in a timber as long as ratios are used, rather than raw values. However, there were questions about the applicability of this graph to other situations. 1073

4 Figure 1: Graph of histogram function (from Adobe Photoshop Elements) versus percent of remaining section for stepped-block radiographs Figure 2: Graph of average corrected histogram ratio (for all pulses) versus percent of remaining section for radiograph of the stepped block, with additional points The primary concern was for the hot spot effect. This effect is visible in most radiographs (see Figure 5 below), where the top and bottom of the radiograph are darker than the more central areas, even when the thickness of the member is uniform throughout. It was also apparent when the histogram function was used on smaller sub-sections across any portion of the stepped-block radiographs representing only one thickness, with higher values found closer to the center of the hot spot. To address this problem, the radiographs were examined visually, to determine the probable center of the hot spot. It was then decided to use only the top 15 percent of the

5 inch and 5-inch sections, which appeared to have approximately the same relationship to the center of the hot spot as the entire portion of the 10-inch and 2.5-inch sections. To make this a relatively uniform measurement, a grid function was superimposed on the radiograph prior to measurement. While there is still some curvature in the graph, it does appear to partially correct for the hot spot problem, with all histogram ratios following approximately the same trend line. To completely eliminate the hot spot problem, additional research into image enhancement would be required. To validate the potential relationship, other radiographs were used (with other species, pulse numbers, and x-ray setups), and the histogram ratio calculated for wood with a known loss of section. This was done for radiographs of simulated termite damage, a decayed log, and other sections of the stepped block. The graph of the collapsed function (Figure 2), with the other calculated points shows the validity of this approach. Finally, a linear function, showing what might be the effect if the hot spot problem was completely eliminated was plotted as the heavy black line on Figure 2. The equation for this line is: RS% = (111 x HR) - 11 (1) Where: RS% is the percent of remaining section; HR is the histogram ratio (the median value of the histogram for intact wood divided by the median value of the histogram for damaged wood). 4. USE OF THE TECHNIQUE AT MISSION SAN MIGUEL ARCHANGEL Mission San Miguel Arcangel was founded on July 25, 1797, one of 21 founded by the Franciscans along El Camino Real (the Royal Way) during the period of Spanish colonialism. Set near the banks of the Salinas River in modern-day California, the church was constructed between 1816 and The interior finish was completed in 1823, with frescoes directed by Spanish artist Esteban Munras but painted by the local Salinas Indians who were neophytes at the Mission (Figure 3). The frescoes, largely untouched since their original painting, are one of the most important historic features of the Mission today. The Mission was considered moderately well-preserved by many for decades, but a severe earthquake in 2003 caused extensive damage, moving adobe walls laterally over a foot and leaving large cracks in the walls at corners and above doors and windows. In consultation with Rainer & Zebala Partners and with the assistance of Mr. Douglas Porter of the University of Vermont, Anthony & Associates, Inc. investigated the condition of a selection of viga and ceiling boards at the Mission. The investigation focused on the general condition of wood vigas and ceiling boards because of visible deterioration to these historic elements. In addition to the digital radioscopy work, the field investigation consisted of visual observations, probing, moisture content measurements and resistance drilling to assess the condition of the wood. A general discussion of that investigation is provided in the companion paper by Porter and Anthony [1]. 1075

6 Figure 3: The painted ceiling and upper portions of the side walls and retablo of the church at Mission San Miguel Arcangel Digital radiographs were made of other wood elements in the church in addition to the vigas and decking. The radiograph in Figure 4 was taken of a portion of the elevated pulpit on the north wall. It shows well developed grain (growth rings) in the boards, a few termite galleries, and original nails. To investigate the decking, an initial image was taken of a thick piece of ceiling decking that had failed. The end view (Figure 5) shows that this wood is riddled with termite galleries. This was verified through resistance drilling, a quasi-non-destructive technique that has proven to be useful for quantifying deterioration in large timbers. The radiograph of this decking (Figure 6) shows an amorphous character to the image with no clear grain structure from the growth rings. An indication of the extent of the termite damage is the lack of well-defined individual galleries. This is because the three-dimensional piece of wood, riddled with termite galleries, is reduced to a two-dimensional image on the radiograph so the galleries at the front of the piece overlap with those in the center and the back. It should be noted that although radioscopy could be used to determine the presence of, and quantify, termite damage, the technique would be more cumbersome for a comprehensive investigation of many large timbers than resistance drilling. 1076

7 Nails Termite galleries Figure 4: Radiograph showing a few termite galleries in a hardwood support and mostly sound wood in the softwood members. A small slice of the top step is on the bottom of the photograph, and the diagonal boards with original nails can be seen on the top Figure 5: End view of the failed ceiling decking timber showing the entire cross section riddled with termite galleries 1077

8 Possible decay Well developed termite galleries Figure 6: Radiograph of failed ceiling decking timber showing well developed termite galleries and lack of uniform wood grain. Possible decay is displayed in upper right Initially, it was thought that digital radioscopy would allow for identifying all of the painted ceiling decking that had been attacked by termites. This information would be of use to the paint conservator to generate a plan for cleaning or restoring the painted boards. After examining several of the boards visually and with simple probing, it was determined that the extent of damage was wide-spread and likely affected virtually all of the ceiling decking. Moisture stains and stains from other residue were visible on the painted surface of much of the decking. Simply touching the painted surface could result in paint chipping off of the decking. The decking was visually examined from the attic. Some of the decking was found to have no visible damage on the unpainted surface. Other decking had termite galleries on the upper unpainted surface. This was the case not only for the nominal one inch (2.5 cm) decking but also for the occasionally thicker (approximately 6 to 13 cm thick) decking. The reason for using the uniformly-spaced thicker decking is unknown and may have had something to do with roof or ceiling support prior to the later installation of steel trusses. Concern over the condition of the thicker decking timbers was due to the extent of visible termite damage on the upper surfaces when viewed in the attic and from the failed decking timber in the image shown earlier. It was assumed that termite damage was extensive in the decking but whether the damage was concentrated just below the painted surface or was uniform throughout the cross section of the decking was unknown and of some importance to the paint conservator. Digital radioscopy was used to answer this question. Rather than attempting to take radiographs of decking in place, two samples of decking, one from the nave, the other from the sacristy, were examined. These samples had a range of visual termite damage and were x-rayed to determine whether the loss of section due to termite damage could be quantified. Using the methodology described above, an estimate of the sound and damaged wood was made. A calculation of the histogram functions in small 1078

9 areas of the radiographs provided the location of the termite damage and an estimate of remaining section (since all other factors of the x-ray set-up remained constant). The decking was initially x-rayed through the 2.5-cm thickness of the boards. The decking sample from the sacristy had the most variability in internal damage, with one end displaying relatively little damage from termites both visually and from the radiograph, while the other end displayed significant damage. Comparing the histogram function of the same central area in the radiographs of the two images gave an estimate of approximately 50 percent of the wood remaining. In addition to taking x-rays of the decking through the board face, x-rays were also taken through the narrow face of the board (Figure 7). The setup shows how timbers were used to prevent bleeding of x-ray energy around the edges of the narrow decking, which would have resulted in brightness that makes it difficult to analyze the image. The corresponding radiograph (Figure 8) clearly shows the presence of termite galleries near the outer faces of the ceiling board. A histogram comparison of this board showed that material along the edges (where the termite galleries were concentrated, just beneath the paint) contained approximately 56 percent of the material in the center of the piece. Figure 7: Radioscopy setup for taking an image through the width of the ceiling decking from the nave. Heavy timbers were placed above and below the board to prevent bleeding of the x-rays and a hotspot in the resulting image 1079

10 Termite galleries Figure 8: Radiograph corresponding to the photograph in Figure 8. Note that the termite galleries are concentrated near the upper (painted) and lower surfaces Based on this image, we do not know that the center of the decking is sound but we do know that is has about twice the wood per unit volume as the edges. Using this information in conjunction with the results of the images taken through the face of the boards, we can conclude that approximately 50 percent of the wood (at least in these samples) has been lost to termite attack. Further, the most severe loss of material is just below the outer surfaces, including the painted surface. 5. SUMMARY The ceiling decking at Mission San Miguel Arcangel consists of nominal 2.5 cm-thick boards except for three thicker lines of timbers that extend longitudinally through the nave ceiling and one line in the sacristy ceiling. The ceiling boards have extensive termite damage. Using techniques developed for x-ray examination of historic wood fabric, two ceiling boards were examined by digital radioscopy (one from the nave and one from the sacristy), and the damage was typically a loss of approximately 50 percent of the wood material and was more severe just below the painted surface and the (upper) unpainted surface. In some cases, only the paint remained where the insect galleries were near the surface. This investigation demonstrated that digital radioscopy is a tool for quantifying the extent of deterioration in wooden objects where other methods of examination may cause damage to the sensitive fabric. 6. REFERENCES [1] Historic Fort Collins Development Corporation, Anthony & Associates, Inc., and City Visions, Inc., 2005, Advances in Digital Radioscopy for Use in Historic Preservation, NCPTT Grant Number MT NC-07, National Center for Preservation Technology Training, 119p. [2] Porter, D.W. and Anthony, R.W. Nondestructive Evaluation at Mission San Miguel Achievements and Limitations, Proceedings of the SACoMaTiS 2008 International RILEM Conference, On Site Assessment of Concrete, Masonry and Timber, Varenna, Italy, September 1-2, 2008 (in press). 1080