Redwood Water Tanks Hawai’i Volcanoes National Park

National Park Service U.S. Department of the Interior

Vanishing Treasures Program

Trip Report and Preliminary Treatment Recommendations

January, 2018

National Park Service Vanishing Treasures Program

Trip Report and Preliminary Treatment Recommendations Redwood Water Tanks

Hawai’i Volcanoes National Park

Prepared by:

Randall Skeirik, R.A., Vanishing Treasures Historical Architect, National Park Service

Vanishing Treasures Program Directed by:

January 2018

Cover: One of the five remaining redwood tanks with the overhead pipe from the rain shed in the foreground.

Above: Close-up of one of the lugs that connect lengths of the steel retaining bands. Note the corrosion present in the lower band and on the threads of the upper.

Figure 1a: Partial site plan of the tank farm and rain sheds showing the 14 historic redwood tanks arrayed in two rows at the top center. The five remaining tanks (T7 - T11) are shown in blue, while the tanks that were demolished to make way for larger metal tanks are shown in pink.

Drawing from eTIC

Figure 1b: Aerial view of the tank farm and rain sheds corresponding to the area includ-ed in the site plan in Figure 1a.

T7

T8

T9 T11

T10

T7

T8

T9 T11

T10

Background and Purpose This trip report is the result of a site visit made on March 10, 2017, in response to a technical assis-tance request from Hawai’i Volcanoes National Park. The site visit, by Vanishing Treasures His-torical Architect Randy Skeirik, was to provide guidance on the preservation of wood water tanks that had been used as a component of the Park’s mid-20th century fresh water system, but have since been abandoned.

The Redwood Water Tanks Acquired at various times after the conclusion of WWII, fourteen 80,000- to 100,000-gallon redwood tanks (Figure 1a) were installed in the headquarters area of Hawai’i Volcanoes National Park to store rainwater collected by an adjacent rain shed.

Eight 100,000-gallon war surplus tank kits were purchased on Oahu in February, 1947 and, ac-cording to the 1949 Superintendant’s Report, three 80,000-gallon and one 100,000-gallon tanks were donated to the park by the Volcano House and moved from the Volcano House tank hill.

Specific entries from the 1949 Superintendent’s Report include:

pg. 93 - Memo to Director:...(4) other large tanks were partially dismantled preparatory to moving them from Tank Hill to the approved site at the rain shed.

pg. 143 - Photo Caption:

(3) 80,000 gallon tanks and (1) 100,000 gallon tank removed from VH tank hill to be re-erect-ed at rain shed.

pg. 144 - Photo Caption:

Photograph clearly shows (3) elevated tank foundations under construction for the previ-ously reported 80,000 gallon tanks...all three 80,000 gallon tanks would had to have been at the same height to operate together with the existing 100,000 gallon tanks; the elevated foundations are for the last (2) tanks erected on the outside row and the last tank erected on the inside row.

pg. 54 - Photo Caption:

The 80,000 gallon water storage tank in the background is complete except for roofing.

pg. 54 - Photo Caption:

Water storage tanks in park headquarters area.

This part of the water plant consists of (12) 100,000 gallon tanks (3) 80,000 gallon tanks, and

(3) 50,000 gallon tanks [under the rain shed].

As park needs grew, more water collection and storage capacity was needed, and nine of the 14 redwood tanks were demolished to make way for larger, metal tanks. (See Figures 1a & 1b) Ad-ditional metal tanks and a second rain shed were also constructed to the northwest of the original infrastructure. (Figures 1a and 1b)

Today, the remaining 5 redwood tanks (Figure 2) (4-100,000-gallon tanks and one 80,000-gallon

Figure 2: General view of the remaining redwood tanks. The tanks all hold 100,000 gallons except the tank on the front left which is smaller than the others, holding only 80,000 gallons.

tank) are no longer in use, and the park has con-sidered demolishing them. However, because they are now considered historic, the VT Histori-cal Architect was asked to assess the condition of the tanks, determine if it is possible for them to be preserved, and if so, to make recommendations for preservation treatments.

As it was only possible to examine the tanks from the ground, they were not inspected in detail. It was also not possible to inspect the insides of the tanks, so the following assessment and recom-mendations are based on a physical inspection of the exterior of the concrete foundations, the bottom six feet of the staves, and the lower steel bands. Based on this inspection, it appears the physical fabric of the tanks remains sound and that preservation and ongoing preservation of the tanks is possible.

In developing a preservation plan for these tanks, it is important to understand the dynamics of each tank as a system (Figure 3), and that the presence of water inside the tank is integral to the functioning of that system. Without the forces generated by the thousands of gallons of water typically stored in a tank, it is prone to failure and collapse.

The primary component of these tanks is the red-wood staves that run the full height of each tank.

The edges of each stave are slightly beveled so that they fit flush against the adjacent staves. (Figure

4) A watertight seal is achieved when the tank is

Figure 3: Cross section diagram through a tank wall shows the tank’s concrete foundation and asphalt floor. The wall of the tank is composed of redwood staves bound together with steel rods. The tank is capped with a 12-sided corrugated metal roof.

Figure 4: Cross section of typical wood tank or pipe construction showing how each stave is beveled (in this case 5.56” on the outside and 5.29” on the in-side) to accommodate the curve of the tank or pipe.

Also shown is a typical lug that is used to hold and tension the steel bands.

filled and the staves absorb water, swelling tightly against each other.

The structural capacity to resist the horizontal hydrostatic fluid pressure that is generated by the thousands of gallons of water contained in each tank is provided by a series of steel bands that en-circle each tank and prevent the staves from being pushed outward.

The mild-steel bands on the HAVO tanks are either 3/4” or 5/8” inches in diameter and are formed from five individual sections that are threaded at each end and joined by steel lugs. (See Cover Page and Figures 4 & 5) Because the hydro-static pressure is greatest at the bottom of a filled tank, the bands at the bottom are spaced more closely (six inches apart), with the spacing of the bands increasing as you progress up the tank, first to eight inches and finally to twelve inches at the top.

At ground level, a concrete footing both provides a firm foundation and holds the bottom end of the staves in place. (See Figures 3, 7, & 8) The bottom two bands, spaced only 3 inches apart, press the lower end of the staves against the foundation, effectively locking them in place. (See Figure 3) At the tops, the staves rely on water pressure to push them outward and hold them tight against the upper bands. While it was not possible to examine the roof structure, it is likely that the roof framing and the outward thrust of the roof rafters are helping to hold the tops of the staves in place in the absence of water pressure in the now empty tanks.

Existing Conditions -General A physical examination of the tanks from the ground indicated that they largely remain in good condition.

One significant exception is the mild-steel bands which are uniformly failing. Mild steel is not an alloyed steel and therefore does not contain sig-nificant amounts of anything other than iron. The lack of alloying elements, such as those found in stainless steels, means that the iron in mild steel is subject to oxidation (rust) if not properly coated.

Given the humid subtropical climate of HAVO coupled with the presence of corrosive sulfur dioxide from volcanic gases, the steel in the tank bands is showing significant corrosion, and in some places have rusted completely through.

(Figure 6) The black plastic that encases all of the steel bands may have been an attempt to “coat” the steel and protect it from corrosion, but given the ubiquity of the corrosion, this was obviously ineffective.

Existing Condition-Foundations The reinforced concrete foundations are two feet thick at their base and extend eight to twelve inch-es below grade. One foot up from the bottom, the foundation steps back to become just one foot in thickness and continues up for an additional foot. (See Figure 3) “J”-bolts, which secure a 2 x 6 plate, are embedded in the top of the foundation.

The purpose of the 2 x 6 plate is not clear, but, by swelling when wet, it may help to provide a water-tight seal at the point where the wood staves meet the concrete foundation.

Figure 6: The damp climate, coupled with the pres-ence of sulfur dioxide from volcanic gases, has caused the mild-steel in the tank bands to corrode.

Here the band has rusted completely through.

Figure 5: Close-up view of one of the lugs that con-nects the 5 sections of each steel band. The threaded connection with the nut (circled) allows the bands to be tensioned so that they compress the wood staves and resist the outward thrust of the water.

Despite some surface erosion, the concrete ap-pears to remain in good condition. Soil deposi-tion has raised the grade around the tank foun-dations, brining it up to and over the level where the foundations step back. This has allowed moss and/or grass to grow right up to the base of the tanks creating a situation where the end grain of the staves has become susceptible to prolonged exposure to moisture. (Figures 7, 8 & 9)

Because the tops of all the tanks must be at the same height in order for them work as a system, the smaller 80,000-gallon tank (T8) has been placed on an earthen platform that is supported on the south by a low stone retaining wall. (Figure 10)

Figure 9: Close-up of the base of a tank showing where soil had covered the foundation’s step-back allowing grass to grow on top of the concrete. This condition can potentially accelerate the deteriora-tion of both the wood staves and the concrete.

Figure 8: Changes in soil levels have covered the bottom halves of the footings allowing grass to grow right up to the base of the tanks.

Figure 7: At the base of the tanks, the concrete foot-ing steps back approximately at grade level. Soil deposition has both raised the grade around the tanks and allowed soil to accumulate on top of the footing, supporting the growth of both grass and moss. This will accelerate the deterioration of both the concrete footing and wood staves.

Figure 10: Tank 8 sits on an earthen platform re-tained by a rubble stone wall

Figure 11: Close-up of tank staves and bands. Note that because the tank is empty there is no horizontal hydrostatic pressure to push outward and hold the steel bands in place. Nails have been used to pre-vent the bands from slipping down.

Figure 12: Wood is most vulnerable to water pen-etration at the end grain, yet even in areas where grass was growing directly under the ends of the staves, an icepick would penetrate less than an eighth of an inch, indicating that the staves remain in good condition.

Recommended Treatment-Foundations Grade should be adjusted around the foundations to restore a condition like that shown in Figure 3, where the soil level is held below the level of the step-back. At the same time, it is important to slope the grade away from the base of every tank so that water will drain away from the foundations and not pool against the foundation. All accu-mulated soil, grass, and moss should be removed from the top of the step-back. This should be kept clear and clean so that the bottom ends of the staves are not exposed to additional moisture and can quickly dry when they are wetted.

Existing Condition-Staves The staves form the wall of the tank, support the roof of the tank, and contain the water when the tank is filled. As such, staves are a primary struc-tural component of the tank. (Figure 11) The staves of the HAVO tanks, made from clear red-wood, possess a natural decay resistance. When inspected from the ground, the accessible por-tions were found to be good condition with no apparent, significant issues. (Figure 12)

However, it has been reported by park staff that the staves have experienced deterioration on their inside faces. Because the 2017 inspection involved only an exterior assessment, this could not be confirmed. If true, it will be necessary to establish the full extent of the deterioration.

Recommended Treatment-Staves As far as they could be inspected, the staves were found to be in good condition on the exterior. To keep them that way, the foundations should be cleared as previously described to help keep the staves dry and to allow them to dry more quickly when wetted.

Additionally, where moss, mildew, fungus, or li-chens are growing on the wood (Figures 12 & 13), major problem, some staves were showing areas of raised grain, and a more detailed inspection may reveal more evidence of this type of UV deg-radation.

If UV degradation is found to be a significant problem, there is a treatment that has been devel-oped specifically for unfinished redwood that will protect it from UV radiation. (See Attachment 1) If, upon close inspection, significant surface loss is noted, it would be worthwhile to test this treat-ment on the tanks.

This UV protectant has only been applied in arid environments, where it has been found to last up to five years. Its longevity in a subtropical climate has not been established

For as long as the UV treatment does remain in place, it will also serve to prolong the effectiveness of the borate treatment because it will consoli-date the wood’s surface, helping to prevent the borate from leaching out. For this reason, the UV treatment should only be applied after the liquid borate treatment.

On the interior, if decay is found but does not extend beyond the surface so that the staves are not structually compromised, the condition could be addressed with an application of a liquid bo-rate. This will help to prevent further decay and, because the inside face of the staves are protected from the weather, this should prove to be a rela-tively longterm treatment.

If some staves are found to be structurally com-promised, it will be necessary to determine which staves in each tank are damaged and, since the exact form of the roof structure is not known, whether or not they are involved in supporting the roof structure. Since no damage is expressed on the exterior, and because the tanks will be preserved empty, it may be possible to retain the damaged staves that do not help support the roof.

While the proper treatment cannot be determined until actual site conditions have been estab-lished, it may be possible to stabilize deteriorated, non-structural staves with an epoxy consolidant and leave them in place.

Deteriorated structural staves will need to be replaced or supplemented. Exact treatments cannot be recommended until actual conditions have been established, but replacement staves may the surface should be cleaned with a 10% bleach solution using a natural bristle brush. In areas where regrowth is likely to occur, especially at their bottom ends, the staves can be treated with a liquid borate product such as Boracare. It will be particularly beneficial to treat the end grain of the wood, getting the wood to absorb as much of the borate as possible.

Because the staves are not painted, the borates are likely to leach out of the wood fairly quickly and the treatment may need to be reapplied as frequently as once a year. Monitoring for fungal growth or deterioration will help establish the frequency for which reapplication will be most beneficial.

Additionally, ultraviolet (UV) radiation will break down the lignin in the wood, resulting in surface loss. This is often expressed as “raised grain”.

(Figure 14) While this was not observed to be a

Figure 13: As seen here, close proximity of vegeta-tion, including grass, shrubs, and trees can create microclimates that prevent drying and encourage the growth of moss and fungus on the wood staves.

The area around the tanks should be kept clear of vegetation and the grass should be kept short.

Figure 14: Ultraviolet degradation is one cause of raised grain, as seen here.

This would then let air in at the base. These openings would need to be screened to prevent the entry of animals and insects.

At the tops of the tanks, additional vents could be created in the soffits, or a minimally instrusive vent could perhaps be installed at the peak of the roof, or vents could be fabricated to run along the roof hips similar to a ridge vent. Like the vents at the base, these openings should be screened to prevent the entry of birds, animals or insects.

Existing Condition-Steel Bands Five-part steel bands encircle each tank, spaced from 6 to 12 inches apart. (See Figures 3, 5, 6, 11 & 16) As already noted, many sections of the steel bands are in poor condition. The humid, sub-tropical environment coupled with the cor-rosive effects of sulfur dioxide from volcanic gases have had a devastating impact on the mild steel of the band sections, many of which have begun to corrode at multiple points around the circumfer-ence of the tanks (Figures 5, 6, 16 & 17).

Currently, on at least one tank, the two bottom bands - which are critical for holding the bottoms of the staves in place against the concrete founda-tion - have both rusted completely through. (Fig-ure 17) While this tank appears to remain stable, the failed bands have the potential to structurally compromise this tank.

be available through the original manufacturer, In-ternational Tank and Pipe Co., or through salvage companies. Replacement staves should match the original in both size and species. A possible alter-native would be to provide supplemental struc-tural support on the interior of the tank where it would not be visible. The design of any supple-mental supports would be dependent on actual site conditions and the structure of the roof itself.

The possible presence of deterioration on the interior of the tanks suggests that the tanks are not adequately ventilated. Because the tanks are being preserved empty, methods for venting the tanks to equalize atmospheric conditions inside and out should be explored - especially if inte-rior deterioration is found. Currently, the tanks appear to only be vented through small (approx-imately 2 inch diameter) holes that are located in the soffit - one to either side of each roof hip.

(Figure 15)

In order to improve ventilation, it would be ide-al to create openings at the base of each tank and also at roof level to allow air to be drawn up through the full height of the tanks. It might be possible, for instance, to use a concrete core bit to create a number of holes in the foundation between the bottom of the staves and the point where the foundations step out. Connecting holes would need to be drilled down through the as-phalt floor to connect with the foundation holes.

Figure 16: General view of tank bands showing areas of corrosion. Corrosion is widespread with a number of bands having rusted completely through.

In addition, all of the threaded ends have corroded to the extent that the bands will need to be cut for them to be removed.

Figure 15: Looking straight up the outside of the tank, small holes can be seen in the roof soffit - one to either side of each roof hip. These appear to be the only means of exchanging air between the inside and outside of the tanks. This photo also shows the bottom of two intact sections of the wood gutter.

Soffit

Gutter

Bottom edge of Fascia

Vent (typ)

In addition, the threaded ends on all of the bands have corroded, making it impossible to turn the nuts that are used to tension or remove the bands.

(Figure 5) As a result, in order to tension, or to remove and replace any band sections, they will have to be cut.

Despite the poor condition of the bands, the lugs all appear to remain in good condition and can be retained and reused, provided that they can be separated from the corroded bands without damage.

Recommended Treatment-Steel Bands Each band is made from 5 separate, approximate-ly 20-foot-long sections that are joined with bolt-ed connections at steel lugs. (Figure 5) Concern has been expressed that when a band corrodes all the way through, it may snap, flying out from the tank with the potential to cause injury or damage.

Because there is no water in any of the tanks, and therefore there is no outward pressure, the bands are no longer in tension and it is highly unlikely that they will snap. Instead, they will more likely just break apart and remain in place as long as the nails that are holding them remain intact.

Nevertheless, the corroded bands will need to be replaced and eventually all of the remaining bands will rust through and require replacement.

Replacement is particularly urgent for the tank that has lost both of its bottom two bands (Figure

17) because this may structurally compromise the tank. These bands should be replaced as quick-ly as possible. Replacement bands are available from the original manufacturer, National Tank and Pipe Company (now International Tank and Pipe Company) of Clackamas, Oregon.

As noted above, in order to preserve the tanks, it will be critical to maintain tight compression of the staves at the base. This is accomplished by tightly fastening the bottom two bands, pressing the staves between the bands and the concrete foundation.

It will also be important to create some form of compression ring on the interior of each tank at the top end of the staves (see Recommended Treatments-Roof, below) and to then provide com-pression against this ring using a tightened band around the outside of the tank. By compressing the staves at the top and the bottom of each tank, the staves will be locked in place, holding the tanks together and preventing collapse.

Because the tanks no longer contain water, mul-tiple, closely spaced steel bands are no longer necessary. Because strength in tension is not required by the majority of the bands on an empty tank, alternatives to expensive, and vulnerable, mild steel bands may be available. For instance, because the bands on these tanks are covered with black plastic, it may be possible to replace all of the non-structural bands with black plastic tubing. A firm called Netafim produces a black polyethylene tubing is both UV and acid resistant, and that can withstand heat, direct sun and harsh chemicals. It is available in 3/4 and 1 inch tubing diameters that should closely match the diameter of the plastic housings on the existing bands and it comes in coil lengths of up to 1000 feet. (http:// www.netafimusa.com/greenhouse/products/sys-tem-components/polyethylene-tubing/) A short length of threaded rod (preferably stain-less steel) could be epoxied into each end of each section of tubing to receive the tensioning nuts, thereby closely replicating the appearance of the existing bands. If the hollow tubing proves too insubstantial, it could be filled with something like sand to provide additional weight.

While the majority of the bands can be replaced with non-functional replicas, it will be critical

Figure 17: On one tank, both of the bottom two bands have rusted through. Because these two bands compress the staves against the concrete foun-dation, this tank may be structurally compromised.

Compression at the bottom end of these staves should be reestablished as quickly as possible.

to have functioning steel compression bands in place at the top and bottom of each tank, and to have perhaps three additional lightly tensioned steel bands equally spaced between those top and bottom bands.

Finally, where non-functional bands are used, they will not be in tension around the tank so they will tend to slip down. The problem of loose bands is currently addressed by supporting the bands with nails driven into the staves at regular intervals. (Figure 11) In places where the nails have come loose, the bands have begun to slip downward. (Figure 18)

While the nails currently in use are not overly intrusive when viewed from a distance, it may be possible to find something that will hold the bands more securely yet be less visible. One possibility could be black plastic coated nail straps similar to the one shown above. The black plastic coating on the nail straps would blend with the black plastic on the bands, helping to reduce their appearance. Nail straps are available from compa-nies such as Dottie. (http://catalog.lhdottie.com/ viewitems/nail--straps/plastic-coated-nail-straps).

(Figure 19) Whatever type of fastener is ultimately used, galvanized metals should be avoided since the zinc in the galvanizing will react with chemi-cals in the redwood and stain the wood. Stainless steel and aluminum are the preferred metals to use for fasteners in redwood.

Existing Condition-Roofs Each tank is covered by a 12-sided pyramidal roof.

(Figure 20) Without access to ladders or scaffold-ing, it was only possible to asses the roofs from the ground or with photographs. Further, without access to the interiors of the tanks, there is no way to know the nature or condition of the roof struc-ture. Cross section drawings of the tanks (Figure

3) only show the roofs conceptually without any hint of their actual construction.

A visual inspection from the ground revealed white-finished corrugated aluminum panels with matching white metal hip caps. At least one of the tanks, Number 7, has a hatch in the roof to pro-vide access to the interior of the tank. (Figure 21) It stands to reason that the other tanks also have roof hatches, but they were not observable from the ground.

Figure 19: In place of shiny silver nails, something like this black plastic coated nail strap could be used to less conspicuously lock the bands in place.

Figure 18: Without the swelling and outward pres-sure that result from filling a tank with water, the circumfrance of an empty tank will shrink, releasing pressure on the steel bands. Where the nails support-ing the bands have failed, the lugs and bands are slipping downward (arrows).

Figure 20: Each tank has a 12-sided pyramidal roof that is covered with white corrugated metal panels.

The tanks originally had wood gutters, sections of which remain on some tanks.

Viewed from the ground, the corrugated alumi-num roofing panels appear to be a relatively light gauge and, in some cases, the edges and corners appear to be coming loose. (Figure 21)

Some tanks retain sections of their original wood-en gutters. (Figure 22) As can be seen in Figure 15, the wooden box gutters were installed directly against the fascia of the roof’s eave. The gutter sections that remain are in poor condition, are choked with debris, and are supporting healthy vegetative growth. (Figure 22) On at least two tanks, the gutters have detached and are dangling

Figure 24: Although a smaller tank in a different location, this clearly illustrates what can happen to an empty tank. With no water in the tank, there is no outward pressure on the staves. As the staves dry and shrink, gaps will open between the staves and, once there is enough space for one stave to fall inward, the rest will quickly follow.

Figure 21: Tank 7 has an access hatch through the roof to provide entry to the tank interior. Logically every tank would need such a hatch. It may be that other hatches are not visible because they are closed (this hatch is open), and/or because they were not located in a position where the viewing angle al-lowed them to be seen from the ground. The corru-gated aluminum roofing panels appear to be a light gauge and some may be working loose (circled).

Figure 22: Some tanks retain sections of the original wood gutters. What remains is in poor condition, is choked with debris, and is supporting healthy veg-etative growth. Visible here is the gutter’s “down-spout” (circled). As might be expected, the rain that fell on the tank roofs was also collected and chan-neled into the tanks.

Figure 23: The remaining gutters are in poor con-dition and some, as seen here, are detached and dangling precariously above the ground.

precariously. (Figure 23) The “downspouts” from the gutters (See Figure 22) drop only a few inches before making a 90-degree turn to conduct the rainwater into the tanks.

Recommended Treatments - Roof To help with the long-term preservation of the tanks, it will be important to maintain the tank roofs to keep rain water out of the tanks so that the interiors stay dry. To that end, the roofs will require regular inspection and maintenance to repair damage and deterioration as it occurs.

Venting to allow interior and exterior atmospher-ic conditions to equalize will also be important.

With the tanks preserved empty, it will also be critical to assure that there is adequate outward pressure on the tops of the staves to prevent them from collapsing inward. (Figure 24) Because all sloped roofs develop an outward thrust, it should be possible to use the roof structure to help achieve this objective.

Hip Rafter (typ)

Align topmost band with 2x12s Band must compress staves against 2x12s.

2 x 12 Hip Rafter

Corrugated Metal Roofing

Fascia Soffit

(2) 2x12x8’ cut to match radius of tank and fitted between each pair of hip rafters. Attach to rafters with skewable angles.

Redwood Stave

Skewable Steel Angle

Replacement band (not re-quired to be structural, typ.)

A

A

(2)-2x12x8’ cut to fit interior radi-us of tank (full, uncut size shown dashed). Typical for full circumfer-ence of tank. See Section A-A for possible connection to rafters.

Figures 24a & b: Figure 24a (top) shows a hypothetical roof fram-ing plan - based on the assump-tion that 2x12 hip rafters are used. It is impossible to surmise the nature of the framing be-tween the hip rafters. Figure 24b (left) illustrates how the rafters may relate to the stave wall of the tank. If possible, 2x12s could be cut to match the radius of the tank and installed inside the tank between the hip rafters. These should align horizontally with the topmost steel band in order to compress the tops of the staves.Section A-A

Not to scale

Hypothetical Roof Framing and Conceptual Design for Installation of a Compression Ring at the Top of the Tank - no scale

Roof Hip Rafter Plan Not to scale

Because these tanks have not collapsed, they may already have some internal structure that is pro-viding the necessary outward thrust. Without the ability to inspect and assess the tank’s interiors and roof framing, it is impossible to know.

If they do (See Figure 25), no changes may be required. If they do not, then support (a “com-pression ring”) should be added that will provide an outward force to counter the inward force from the topmost steel band on the outside of the tank. This will effectively sandwich the tops of the staves between the band and the newly installed compression ring, locking them in place. (See Figure 24b)

Without knowing the exact construction of the roof framing, it is impossible to create a specific design for this internal compression ring - if it is needed. Figures 24a and 24b serve to illustrate the concept, and may serve as a point of departure for a design based on the actual conditions. All of the internal framing should be inspected, and neces-sary repairs made, before adding any additional framing.

With the staves safely locked in place, either by existing framing or by the new compression ring, the existing roofing should be systematically inspected for loose and damaged panels. These should be reattached, repaired, or replaced as necessary.

The roof hatch on each tank should be located and inspected. All hatches should be watertight, Figure 25: Although a scale model, this shows a tank with both a conical roof and framing that is flush with the top of the tank. Framing of this type, if it exists in the HAVO tanks, would help to prevent the inward collapse of the staves.

Figure 26: The effects of “splash” are clearly visible here. The line of gravel on the ground that runs along the base of the tank shows where the water from the roof hits the ground. That water splashes back against the bottom of the tank, wetting the bot-tom ends of the staves. This allows moss and fungus to grow on the wood, leading to deterioration.

be firmly attached, and have a mechanism for latching them in the closed position. All deficien-cies, in any of these areas should be addressed.

As noted above in the section on wood staves, venting opportunities at the roof level should be explored. Hip vents, or a vent at the roof peak would provide the best ventilation and should be incorporated, if possible. Otherwise, methods for discretely increasing the size of the soffit vents should be explored.

Finally, the gutters should be restored. Without gutters, rainwater will run directly off the roof, hitting the ground at the base of the tanks. This water will splash back against the wall of the tank, contributing to the deterioration of the concrete foundation and wetting the bottom ends of the staves. (Figure 26) Wetting the staves in this way will encourage moss and fungal growth on the wood, leading to deterioration of the bottom ends of the staves.

In order to preserve the historic appearance, the new gutters should be constructed of wood, although pressure-treated wood attached with screws might be a good alternative to untreated wood and nails. Additionally, the gutter should be lined with metal with soldered connections to create a waterproof liner that will protect and preserve the wood.

Along with the restoration of the gutters, down-spouts must be created to conduct the water down to the ground and away from the base of

Figure 27: Existing internal drains or exterior pipes could be utilized in place of a traditional downspout to carry the water collected by the gutters down to the ground. How this is accomplished will need to be determined on-site, based on actual conditions.

the tanks. Given that the original “downspouts” directed the water into the top of each tank, an al-ternative must be found since we will want to keep the interiors of the tanks dry.

One possibility may be the creation of an internal downspout, using the route of the original down-spout down from the gutter and into the tank, but then continuing a pipe down to the floor of the tank to connect with an existing drain . (Figure

27) It may also be possible to use an existing pipe that runs the full height of the exterior of the tank.

(Figure 27) Either of these alternatives would pre-serve the historic appearance of the tanks while also providing an effective method of preventing splash and preserving the bases of the tank staves.

Any solution will require that the new downspout be adequately sized to carry all of the water col-lected.

Conclusion As noted above, the recommendations presented here were developed based only on an inspection of the tanks from the ground. The tops of the tanks, their roofs, and their interiors were not ac-cessible. Actual conditions, especially those of the roof structure, may be very different from those conjectured here. Therefore, it is important to understand that some of these recommendations may not be applicable, and others may need to adapted to meet actual conditions.

Nevertheless, treatments such as grade adjust-ments, cleaning of mold and fungus, and borate treatments, as well as replacement of critical steel bands can and should proceed as quickly as pos-sible.

In all cases, the intent should be clear that the objectives are to preserve as much of the original fabric as possible, to preserve the existing/historic appearance of the tanks, and to use appropriate substitute materials when it is in the best interest of the resource. All work should meet the Secre-tary of the Interior’s Standards for Preservation.

(See Attachment 2)

Attachment 1: UV Protectant

Note: Reference to removal of existing finishes is not relevant in this case.

Re dw oo d

Pr es er va �o n at

S co y’ s C as tle

De at h

Va lle y

N a� on al

P ar k

Pr oj ec t p ar tn er s i nc lu de

Cu ra �o n, D ea th

V al le y

N a� on al

P k

At la s M at er ia l T es �n g

So lu �o ns

Co er va �o n

As so ci es

Va ni sh in g

Tr ea su re s P ro gr am

(N

PS

Do ug la s P or te r, Pr in ci pa l I nv es �g or

An ge ly n

Ba ss A rc hi te ct ur al

C on se rv

Li is a

Re im an n

Ch ris

D oo le y

Sc ho ol o f E ng in ee rin g, U ni ve rs ity o f V er m on t

Co op er a� ve

A gr ee m en t N o:

H 8W

Ta sk

A gr ee m en t N o:

J8 W

Re dw oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k

W ea th er in g of w oo d is ch ac te riz ed b y ch ro m op ho ric

(c ol or c ha ng es a t t he su rf ac e.

As th e lig ni n is ch an ge d by ex po su re to li gh t, th e ch em ic al pr od uc ts a re c he m ic ly un st ab le a nd d iss ol ve re ad ily in w er

Sm l a m ou nt s o f w at er a re en ou gh to w as h aw ay e te rio ra �o n pr od uc ts o n e su rf ac e, e xp os in g ne w w oo d to lig ht so th at th e pr oc s co n� nu

D A

M A

G E

T Y

P O

L O

G Y

S

U B

S T

R A

T E

S

U R

F A

C E

E R

O S

IO

N

E x p o su re t o U

V l ig h t re su lt s in t h e b re ak d o w n o f th e li g n m ri x o f w o o d re su lt in g i lo ss o f su rf ac e co h er en ce in cr ea se d f ri ab il it y d o si o n E ar ly w o o d i s es p ec ia ll y s u sc ep ti b le t si o n an d t h e p ro ce ss i s o ft en a cc o m p an ie d b y et ac h m en t o f w o o d f ib er s th at a p p ea r as l it tl e

“h ai rs e w o o d s u rf

In F ig w o o d s u rf ex p o se d t o s u n li g h t ar e n o ti ce ab ly e ro d ed w h il e sh ad ed s u rf ac es a re s m o l ar g el y i n ta ct U n ag n if ic at io n d am ag e to t h e w o o d s tr u ct u re a g re g at io n o f w o o d f ib er s fr o m t h e su rf ac e ar e ev id en t

(F ig

I n F ig er o si o n i s p ro n o u n ce d a ro u n d t h e sy em f sm al l ch ec k s th at h ev el o p ed o n t h e en d g ra in el l as a lo n g e ar ly w o la y er s o f th e w o o d s u rf ig

Se ve ra l s tr at eg ie s e xi st s f or th e cl ea r o r s em itr an sp ar en t fi ni sh in g of w oo d in cl ud g:

ol id an t t re at m en t t o im pr ov e su rf ac e re sis ta nc e to w ea th er g

Us e of in or ga ni c pi gm en ts

(s em i-‐t ra ns pa re nt d ec k st ai ns t ha t a bs b

Th e ad di �o n of

U V st ab ili ze rs to c le ar c oa �n gs

St ab ili ze rs in th es e fin ish es in cl ud e ab so rb er s t ha t c on ve rt li gh t e ne rg y to c al or ic e ne rg y he a nd

H

AL

S th at tr ap ra di ca ls fo rm ed b y ph ot o-‐ de gr ad a� on I n th is pr oj ec t p ro pr ie ta ry fin ish es w er e te st ed a s w as

B ut va r B -‐9

(a p ol yv in yl b ut yr al u se d pr im ar ily in o bj ec t er va �o n ap pl ic a� on s f or it s m ec ha ni ca l p ro pe r� es d ep th o f p en et ra �o n, li gh t a nd h t ab ili ty a nd re ve rs ib ili ty /r e-‐ tr ea ta bi lit y) oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k

He th y ce ll st ru ct ur e th er ed d et er io ra te d ce ll st ru ct e ol id at ed w ith

B -‐9 oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k

Be fo re v ar ni sh re m ov

Du rin g va rn ish re m

A� c on so lid a� on

At

S co y’ s C as tle w he re th e ex is� ng c oa �n g( s) h ad fa ile d co m pl et el y, a pp lic a� on o f t he ol id an t r es ul te d in v isi bl e im pr ov em en t o f t he su rf

Dr y cl ea ni ng

(w ith a n yl on -‐b ris tle b ru sh l e� a su rf ac e re la �v el y fr ee o f o ld v ar ni sh ut di sg re ga te d as th e re su lt of

U V ex po su e co ns ol id an t t re at m en t r es ul te d in th e bi nd in g of d et ac hi ng w oo d fib er s t o th e su rf

N ot e th at th e su rf ac e di sp la ys p re fe re n� al e ro sio n of e ar ly w oo d la ye rs oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k

Be fo re c on so lid c on so lid

B -‐9 re m e ea rly st ag es o f f ai lu re e m br i� le d co a� ng s d ev el op c ra ck s t ha t a dm it w at er to e bs tr at e, p ro m o� ng d im en sio na l i ns ta bi lit y as so ci at ed w ith c ha ng es in m oi st ur e co nt t, d su bs eq ue nt d et ac hm en t o f t he c oa �n e co ns ol id an t fi lls th es e cr ac ks r ed uc in g m oi st ur e in fil tr a� e co ns ol id an t c an b e eff ec �v el y re m ov ed w ith ou t a dv er se ly a ffe c� ng th e ex is� ng c oa �n c ar ef ul ly se le c� ng so lv en ts

(f or m ix in g an d cl ea ni ng t ha t d o no t d iss ol ve th e or ig va rn ish (e s) oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k

Tr tm en t P ro ce du re s:

C le an g

1.

D

B ru sh g ov e lo os e va rn ish a nd d eb ris w ith a c oa rs e, n yl br ist le b ru sh B ru sh in g sh ou ld b e in th e di re c� on f e gr ai n.

2.

Ac et on e

Sp ra y* ov e lo os e va rn ish th at d oe s n ot c om e off re ad ily w ith a b ru sh b y ap pl yi ng a ce ne ro ug h a lo w -‐p re ss ur e sp ra y bo le O nc e th e ac et on e ha s d rie d, b ru sh a w ay th e va rn ith a n yl on b ru sh A ga in b ru sh in th e di re c� on o f t he g ra in R ep ea t a s n ec es sa

*T hi s m et ho d of c le an in g is on ly a pp ro pr ia te w he re th e ex is� ng fi ni sh h as fa ile d an d m us t b e re m ov ed e n� re oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k tm en t P ro ce du re s:

C le an g

4.

et

C le an g

Lo os e em be dd ed d irt b ird d ro pp in gs m ild ew d ot he r a cc re �o ns c an b e re m ov ed er a nd a c oa rs e ny lo n-‐ br ist le b ru sh u se d e di re c� on o f t he g ra in T o re m ov e m ild ew us e a

2%

V ul pe x so lu �o n in w at er a nd ri e ou gh oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k tm en t P ro ce du re s:

C on so lid

M ak e su re su rf ac e is fr ee o f d us t, de br is an d lo os e fib s.

Dr y-‐ br us h ba re w oo d r v ac uu m /s po ng e va rn ed rf ac

Cu t i n an y ha rd w ar e us in g a tr im b ru sh A ny

B -‐9 th ad ve rt en tly a pp lie d to m et al h ar dw ar e ca n be re m ed e th an ol th e w oo d ev en ly w ith th e co ns ol id an t.

Ad d a he av ie r ap pl ic a� on to a ny a re as th at a re p ar �c ul ar ly d

Al lo w

B -‐9 to d ry fo r m in ut es b ef or e ap pl yi ng e se nd c oa t.

oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k tm en t P ro ce du re s:

C on so lid ol id an t:

3.

5%

B -‐9

(g /m l) in

E th an ol

(E

TO

H) w

T in uv in by v ol o f B -‐9

B-‐ m l or

0g

Et ha no l

ET

O H)

0m l

Ti nu vi n

.8 l

A.

dd d ry

B -‐9 slo w ly to e th an ol

(E

TO

H) nd ix th or ou gh ly

(d o no t l et th e

B-‐ c lu p)

B.

llo w to si t f or m in ut

C.

dd

T in uv in a nd m ix th or ou gh oo d

Pr es er va �o n at

S co y’ s C as tle

D ea th

V al le y

N a� on al

P k c on sid er in g th e fe as ib ili ty o f u sin g th e tr ea tm ts sc rib ed in th is re po rt in o th er c on te xt s, it im po rt an t t o re ca ll th e pr oc es s a rc hi te ct er va to rs fo llo w ed in d es ig ni ng th e tr ea tm en ts d to a pp ly th em a pp ro pr ia te ly to d iff er en t s ite s d iff er en t m at er ia ls an d co nd i� on s.

It sh ou ld ot be ss um ed th at th e tr ea tm en ts d ev el op ed h er e ca n be u ni fo rm ly a pp lie d at a ny si te I ns te ad ny tr tm en t a t f ut ur e sit es sh ou ld b e pr ec ed ed y:

nd i� on a ss es sm en t, to u nd er st an d th e de te rio ra �o n co nd i� on s t ha t ca id at e tr ea tm en ts w ill n ee d to a dd ss

Fi ni sh es a na ly sis t o de te rm in e w hi ch o f t he e xi s� ng fi ni sh es a re to b e pr es er d sin se le c� on t o se le ct tr ea tm en t p ro du ct s t ha t w ill a dd re ss d et er io ra �o n w ith ou t pr om isi ng h ist or ic fi ni sh ea tm en t t es �n g, to d et er m in e w he th er a nd h ow tr ea tm en ts c an b e ap pl ie d to m in im ize ad rs e im pa ct s t o hi st or ic m at er ia ls

Pi lo t t es �n g, to so lv e pr ob le m s p os ed b y fie ld im pl em en ta �o n of th e tr ea tm en ts a ta bl ish th e de sir ed tr ea tm en t a es th e� c

Attachment 2: Secretary of the Interior’s Standards for Preservation

Preservation as a treatment When the property's distinctive materials, features, and spaces are essentially intact and thus convey the historic significance without extensive repair or replacement; when depiction at a particular period of time is not appropriate; and when a continuing or new use does not require additions or extensive alterations, Preservation may be considered as a treatment.

Standards for Preservation

1. A property will be used as it was historically, or be given a new use that maximizes the retention of distinctive materials, features, spaces and spatial relationships. Where a treatment and use have not been identified, a property will be protected and, if necessary, stabilized until additional work may be undertaken.

2. The historic character of a property will be retained and preserved. The replacement of intact or repairable historic materials or alteration of features, spaces and spatial relationships that characterize a property will be avoided.

3. Each property will be recognized as a physical record of its time, place and use. Work needed to sta-bilize, consolidate and conserve existing historic materials and features will be physically and visually compatible, identifiable upon close inspection and properly documented for future research.

4. Changes to a property that have acquired historic significance in their own right will be retained and preserved.

5. Distinctive materials, features, finishes and construction techniques or examples of craftsmanship that characterize a property will be preserved.

6. The existing condition of historic features will be evaluated to determine the appropriate level of intervention needed. Where the severity of deterioration requires repair or limited replacement of a distinctive feature, the new material will match the old in composition, design, color and texture.

7. Chemical or physical treatments, if appropriate, will be undertaken using the gentlest means possi-ble. Treatments that cause damage to historic materials will not be used.

8. Archeological resources will be protected and preserved in place. If such resources must be dis-turbed, mitigation measures will be undertaken.

The Guidelines for the Treatment of Historic Properties illustrate the practical application of these treatment standards to historic properties.

https://www.nps.gov/tps/standards/treatment-guidelines-2017.pdf

The Guidelines for the Treatment of Cultural Landscapes apply these treatment standards to historic cultural landscapes.

https://www.nps.gov/tps/standards/four-treatments/landscape-guidelines/index.htm