Natural Areas Plan: SFFA comments on the DEIR (Pt 1: Trees)
June 8, 2012 2 Comments
The Significant Natural Resource Areas Management Plan (SNRAMP - Sin-Ramp) calls for removing nearly 18,500 trees in Pacifica and San Francisco. Yet the Draft Environmental Impact Report (DEIR) claims that destroying these huge old trees will not release significant amounts of carbon dioxide (which would violate California’s AB32, regarding greenhouse gases). How can this be true?
NAP claims that they’ll replace the trees one for one, which is neither feasible nor – as we will see – what they intend to do. The Sin-Ramp specifically states that NAP wants to remove the trees in Pacifica to create an environment of coastal scrub. In other areas, they want to remove trees to prevent shading of Native plants. If they plant new trees, it defeats the purpose. They don’t have enough acreage to plant the 18,500 trees they plan to remove. (And… they’ll be planting at best, saplings and small tree-shrubs. What they’re removing are majestic trees, 100-200 feet tall. Which has more carbon?)
Other fictional claims in the DEIR: They’re only removing dead or hazardous trees. Strangely, that’s never mentioned in the Sin-Ramp; it would be odd indeed if such trees just happened to be inconveniently located next to a Native Plant area. Or that when they remove the thousands of trees, other trees unaccustomed to the exposure won’t blow over in storms. That it won’t increase erosion or run-off. That it won’t impact pollution.
The details are below, with references.
SFFA COMMENT ON TREES: “FICTIONAL PREMISES“
The Significant Natural Resource Areas Management Plan (SNRAMP) which is evaluated by the Draft Environmental Impact Report (DEIR) documents plans to destroy thousands of trees in the parks managed by the City of San Francisco in San Francisco and Pacifica. This planned tree destruction will release significant amounts of carbon dioxide into the atmosphere. The DEIR reaches the conclusion that the removal of these trees will have no impact on the environment and will not violate California State law regarding greenhouse gas emissions (AB32). This conclusion is based on these fictional premises:
- That all the trees that are removed will be replaced within the natural areas by an equal number of trees that are native to San Francisco.
- That only dead, dying, hazardous, or unhealthy trees will be removed.
- That trees are being destroyed because they are non-native and invasive.
- That tree removals will not alter wind conditions, causing the trees that remain to fail.
- That tree removals will not cause erosion or increased run-off and sedimentation
- That these tree removals will not result in reduced air quality or the loss of carbon stored in the urban forest
This comment will document that these are fictional premises. They are:
- Contradicted by the horticultural requirements of trees native to San Francisco
- Contradicted by the actual plans as documented by SNRAMP and the “Assessment of Urban Forestry Operations”
- Contradicted by the actual health status of the existing forest
- Contradicted by the actual past practices of the Natural Areas Program with respect to tree removals
- Contradicted by scientific studies and actual experience with tree removals and non-native trees
- Contradicted by the science of the terrestrial carbon cycle
1. TREES DESTROYED BY IMPLEMENTING SNRAMP CANNOT/ WILL NOT BE REPLACED
The DEIR claims that all trees removed in San Francisco will be replaced “one-to-one” by trees that are native to San Francisco. The SNRAMP supports this fictional premise by falsely reducing the number of trees that will be removed:
By not counting trees less than 15 feet tall which it intends to destroy, despite the fact that the US Forest Service survey of San Francisco’s urban forest reports that the trunks of most (51.4%) trees in San Francisco are less than 6 inches in diameter at breast height, the functional equivalent of trees less than 15 feet tall. (Nowak 2007)
- By not counting the hundreds of healthy trees that have already been destroyed by the Natural Areas Program in “natural areas” at Tank Hill, Pine Lake, Lake Merced, Bayview Hill, Glen Canyon parks, etc., prior to the approval of SNRAMP. (see pages 5-8 for details)
However, even artificially reducing the number of trees removed by the implementation of SNRAMP does not make “one-to-one” replacement a realistic goal
THE NATURAL HISTORY OF TREES IN SAN FRANCISCO
The primary reason why we know that it will not be possible to grow native trees in the natural areas in San Francisco is that there were few native trees in San Francisco before non-native trees were planted by European settlers in the late 19th century. San Francisco’s “Urban Forest Plan” which was officially adopted by the Urban Forestry Council in 2006 and approved by the Board of Supervisors, describes the origins of San Francisco’s urban forest as follows:
“No forest existed prior to the European settlement of the city and the photographs and written records from that time illustrate a lack of trees…Towards the Pacific Ocean, one saw vast dunes of sand, moving under the constant wind. While there were oaks and willows along creeks, San Francisco’s urban forest had little or nothing in the way of native tree resources. The City’s urban forest arose from a brief but intense period of afforestation, which created forests on sand without tree cover.”
THE HORTICULTURAL REALITY OF TREES NATIVE TO SAN FRANCISCO
More importantly, the reality is that even if we want to plant more native trees in San Francisco, they will not grow in most places in San Francisco because they do not tolerate San Francisco’s climate and growing conditions: wind, fog, and sandy or rocky soil, etc. We know that for several reasons:
- There are few native trees in San Francisco now. According to the US Forest Service survey of San Francisco’s urban forest only two species of tree native to San Francisco were found in sufficient numbers to be counted in the 194 plots they surveyed: Coast live oak was reported as .1% (one-tenth of one percent) and California bay laurel 2.1% of the total tree population of 669,000 trees. (Nowak 2007)
- The City of San Francisco maintains an official list of recommended species of trees for use by the Friends of the Urban Forest and the Department of Public Works. (CCSF Resolution No. 003-11-UFC) The most recent list (2011) categorizes 27 species of trees as “Species that perform well in many locations in San Francisco.” There is not a single native tree in that category.
- Thirty-six tree species are categorized as “Species that perform well in certain locations with special considerations as noted.” Only one of these 36 species is native to San Francisco, the Coast live oak and its “special considerations” are described as “uneven performer, prefers heat, wind protection, good drainage.”
- The third category is “Species that need further evaluation.” Only one (Holly leaf cherry) of the 22 species in that category is native to San Francisco.
- Finally, about 25 native trees were planted on Tank Hill to placate neighbors who objected to the removal of the trees by the Natural Areas Program (NAP). About 10 years later only 5 have survived and only one shows any growth.
SNRAMP DOCUMENTS THAT NAP DOES NOT PLAN TO REPLACE TREES
In fact, the SNRAMP documents that the Natural Areas Program (NAP) does not intend to plant replacement trees for the thousands of trees it proposes to destroy.
- The majority of trees over 15 feet tall designated for removal by SNRAMP (15,000 trees) are in Sharp Park. The DEIR acknowledges that these trees will not be replaced because this area will be converted to native coastal scrub.
- The DEIR makes no commitment to replace the trees less than 15 feet tall that will be removed but are not quantified by SNRAMP because they are not defined by SNRAMP as trees. There are probably thousands of trees less than 15 feet tall in the “natural areas” that will be removed and not replaced.
- Because most of the natural areas are rock outcrops and sand hills that were treeless prior to the arrival of Europeans, there is little acreage within the “natural areas” that is capable of supporting trees that are native to San Francisco: “Two native forest series…comprise approximately 17 acres, 2 percent of total vegetation [in the natural areas]” (SNRAMP, Setting, page 3-11). Obviously, it would not be physically possible to plant thousands of native trees in the small areas in which they would be able to survive.
- SNRAMP documents the intention to convert all MA-1 and MA-2 areas, comprising 58% of the total acres of “natural areas” to grassland and scrub: “Within MA-1 and MA-2, these sites [of tree removals] would then be replanted with native shrub and grassland species.” (SNRAMP, Forestry Statement, page F-3)
[Editor's note: The Sin-Ramp divides Natural Areas into three tiers by the extent to which non-native plants will be tolerated: the MA-1 are most sensitive; the MA-2 areas which are less so; and the MA-3 areas which will be least intensively managed.]
- Only MA-3 areas, comprising 42% of total acreage will continue to support the urban forest: “Within MA-3, urban forest species would be planted or encouraged (see Section 5, GR-15)” (SNRAMP, Forestry Statement, page F-3). However, the Forestry Statement also documents the intention to thin the urban forest in MA-3 areas to a basal area of 60-200 trees per acre (our estimate based on the formula for basal area in SNRAMP). That represents a significant thinning of the urban forest when compared to the tree density of the eucalyptus forest on Mount Sutro documented by UCSF as 740 trees per acre.
- The “Urban Forestry Statements” in Appendix F of the management plan contain the long-term plans for the natural areas in which trees will be destroyed. All but one of these specific plans is some variation of “conversion of some areas of forest to scrub and grasslands.” The exception is Corona Heights for which the plans are “converted gradually to oak woodland.” The Corona Heights natural area is 2.4 acres, making it physically impossible to plant thousands of oaks in that location.
- “Oak woodland” is the only vegetation goal in SNRAMP which foresees the planting of native trees. Yet, the DEIR says nothing about the potential for Sudden Oak Death (SOD) to decimate the oak population in the San Francisco Bay Area. Ironically, the DEIR acknowledges that one of the comments on the Initial Study raised this question. Yet, despite that question, the DEIR remains silent about the potential for oaks to be killed by SOD. Since the publication of the Initial Study, our local expert (Matteo Garbelotto, UC Berkeley) has reported the rampant spread of SOD and its deadly consequences: “…experts predict as many as 90% of California live oaks and black oaks could die from the disease within 25 years.”
2. THE TREES DESIGNATED FOR REMOVAL ARE NOT DEAD, DYING, OR HAZARDOUS
We have many reasons to challenge the truth of the claim in the DEIR that only dead, dying, hazardous or unhealthy trees will be removed by the implementation of SNRAMP:
- SNRAMP documents that young, non-native trees less than 15 feet tall will be removed from the “natural areas.” By definition these young trees are not dead or unhealthy because they are young and actively growing.
- SNRAMP did not designate only dead, dying, hazardous trees for removal. Trees have been selected for removal only in so far as they support the goal of expanding and enhancing areas of native plants, especially grasslands and scrub.
- The predominant non-native tree in San Francisco, blue gum eucalyptus lives in Australia from 200-400 years, depending upon the climate. (Jacobs 1955, page 67) In milder climates, such as San Francisco, the blue gum lives toward the longer end of this range. The trees over 15 feet tall that have been designated for removal are almost exclusively blue gum.
- However, there are many natural predators in Australia that were not imported to California. It is possible that the eucalypts will live longer here: “Once established elsewhere, some species of eucalypts are capable of adjusting to a broader range of soil, water, and slope conditions than in Australia…once released from inter-specific competitions and from native insect fauna…” (Doughty 2000, page 6)
- The San Francisco Presidio’s Vegetation Management Plan reports that eucalypts in the Presidio are about 100 years old and they are expected to live much longer: “blue gum eucalyptus can continue to live much longer…” (Vegetation Management Plan, page 28)
- The Hort Science “Assessment of Urban Forestry Operations” for the Recreation and Park Department states that, “the life-span of the blue gum, the most common eucalyptus species, is unknown.” In other words, although they have lived in San Francisco more than 100 years, they have not lived in San Francisco long enough to know how long they will live here.
- The Natural Areas Program has already destroyed hundreds of non-native trees in the past 15 years. We can see with our own eyes that these trees were not unhealthy when they were destroyed.
- The claim that only unhealthy and/or hazardous trees will be destroyed in the natural areas is contradicted by the “Assessment of Urban Forestry Operations” of the Recreation and Park Department, July 2010.
- Neither written plans nor EIRs are required to remove hazardous trees. The City has the right and an obligation to remove hazardous trees when they are identified as such by qualified arborists.
TREES HAVE BEEN DESIGNATED FOR DESTRUCTION SOLELY TO BENEFIT NATIVE PLANTS
The DEIR claims that only dead, dying, hazardous trees will be removed from the natural areas. This claim is contradicted by the SNRAMP that the DEIR is supposedly evaluating. Not a single explanation in the SNRAMP of why trees have been selected for removal is based on the health of the trees.
- Lake Merced: The explanation for removing 134 trees is “To maintain and enhance native habitats, it is necessary to selectively remove some trees.”
- Mt. Davidson: The explanation for removing 1,600 trees is: “In order to enhance the sensitive species habitat that persists in the urban forest understory and at the forest-grassland ecotone, invasive blue gum eucalyptus trees will be removed in select areas. Coastal scrub and reed grass communities require additional light to reach the forest floor in order to persist “
- Glen Canyon: The explanations for removing 120 trees are: “to help protect and preserve the native grassland” and “to increase light penetration to the forest floor”
- Bayview Hill: The explanation for removing 505 trees is: “In order to enhance the sensitive species habitat that persists in the urban forest understory and at the forest-grassland ecotone, invasive blue gum eucalyptus trees will be removed in select areas.”
- McLaren: The explanation for removing 805 trees is: “In order to enhance the sensitive species habitat that persists in the urban forest understory and at the forest-scrub-grassland ecotone, invasive trees will be removed in select areas. Coastal scrub and grassland communities require additional light to reach the forest floor in order to persist.”
- Interior Greenbelt: The explanation for removing 140 trees is: “In order to enhance the seasonal creek and sensitive species habitat that persists in the urban forest understory, invasive blue gum eucalyptus trees will be removed in select areas.”
- Dorothy Erskine: The explanation for removing 14 trees is: “In order to enhance the grassland and wildflower community, removal of some eucalyptus trees is necessary.”
In not a single case does the management plan for the Natural Areas Program corroborate the claim made in the DEIR that only dead, dying, diseased, or hazardous trees will be removed. In every case, the explanation for the removal of eucalypts is that their removal will benefit native plants, specifically grassland and scrub. In other words, the explanation provided by the DEIR for tree removals in the natural areas is a misrepresentation of the SNRAMP which it is supposedly evaluating.
THE TREES THAT HAVE ALREADY BEEN DESTROYED IN THE “NATURAL AREAS” WERE NOT DEAD, DYING, OR DISEASED
Although it’s interesting and instructive to turn to the written word in SNRAMP for the Natural Areas Program to prove that the DEIR is based on fictional premises, the strongest evidence is the track record of tree removals in the past 15 years. The trees that have been destroyed in the “natural areas” in the past 15 years were NOT dead, dying, or diseased.
Hundreds of trees have been removed in the natural areas since the Natural Areas Program began 15 years ago. We’ll visit a few of those areas with photographs of those destroyed trees to prove that healthy, young non-native trees have been destroyed. This track record predicts the future: more healthy young trees will be destroyed in the future for the same reason that healthy young trees were destroyed in the past, i.e., because their mere existence is perceived as being a barrier to the restoration of native grassland and scrub.
- The first tree destruction by the Natural Areas Program and/or its supporters took the form of girdling about 1,000 healthy trees in the natural areas about 10 to 15 years ago. Girdling a tree prevents water and nutrients from traveling from the roots of the tree to its canopy. The tree dies slowly over time. The larger the tree, the longer it takes to die. None of these trees were dead when they were girdled. There is no point in girdling a dead tree.
- Many trees that were more easily cut down without heavy equipment were simply destroyed, sometimes leaving ugly stumps several feet off the ground.
- About 25 young trees were destroyed on Tank Hill about 10 years ago. We can see from those that remain that the trees—which were planted around the same time—were young. They don’t look particularly healthy in the picture because they were severely limbed up to bring more light to the native plant garden for which the neighboring trees were destroyed. All of the trees would have been destroyed if the neighbors had not come to their defense. About 25 oaks were provided to the neighbors by NAP to plant as “replacement” trees. Only 5 are still alive. Only one has grown. The remainder are about 36” tall and their trunks about 1” in diameter, as when they were planted.
- About 25 young trees were destroyed at the west end of Pine Lake to create a native plant garden that is now a barren, weedy mess surrounded by the stumps of the young trees that were destroyed. These trees were destroyed after all the trees in Stern Grove/Pine Lake had been evaluated by Hort Science. The trees that were cut down to create this new native plant garden had not been judged to be hazardous. They were cut solely for the purpose of expanding the native plant garden.
- About 25 trees of medium size were destroyed at the southern end of Islais Creek in Glen Canyon Park about 6 years ago in order to create a native plant garden. They were replaced with shrubs.
- Many young trees were recently destroyed in the “natural area” called the Interior Greenbelt. These trees were destroyed in connection with the development of a trail, which has recently become the means by which the Natural Areas Program has funded tree removals with capital funding.
- In 2008, the Public Utilities Commission completed a seismic upgrade to the water tank on Mt. Davidson. Because the existing pipe to the tank from the reservoir was “located under sensitive habitat areas” according to the PUC announcement of the project, the pipe was relocated at the insistence of the Natural Areas Program and its supporters. The relocation of the pipe through the non-native forest required the destruction of approximately 100 healthy, mature trees and substantially increased the cost of the project. Only five replacement trees were planted.
There was nothing wrong with any of these trees before they were destroyed. Their only crime was that they were not native to San Francisco. There are probably many other trees that were destroyed in the natural areas in the past 15 years. We are reporting only those removals of which we have personal knowledge.
THE CLAIM THAT ONLY UNHEALTHY AND/OR HAZARDOUS TREES WILL BE DESTROYED IN NATURAL AREAS IS CONTRADICTED BY THE “ASSESSMENT OF URBAN FORESTRY OPERATIONS” OF THE RECREATION AND PARKS DEPARTMENT
The “Assessment of Urban Forestry Operations” of the Recreation and Park Department was conducted by the professional arborists of Hort Science and published in July 2010. It states that:
- No risk assessments of trees in parks, squares, and golf courses for health and safety hazards had been conducted in San Francisco with the exception of Stern Grove and Park Presidio Blvd at the time the report was published. The hazards identified in those two assessments had been only partially mitigated by the time the report was published.
- All tree maintenance conducted in San Francisco’s parks is reactive, i.e., done in response to specific requests for tree removals or pruning. There was a backlog of “some 450” such requests at the time the report was written.
- The “Assessment” recommends that trees be evaluated in 18 parks considered “high priority.” None of these 18 parks are natural areas. In the few parks that contain natural areas, only the “park perimeter streets” will be evaluated.
- There is no reforestation in San Francisco’s parks, squares and golf courses outside of Golden Gate Park. The number of trees removed in parks and squares exceeds the number of trees planted. The ratio of removals to plantings is significantly higher in golf courses, particularly Sharp Park.
These observations by certified arborists and written in consultation with the Recreation and Park Department contradicts these claims in the DEIR for the Natural Areas Program:
- The trees in the natural areas had not been evaluated for health or safety when they were designated for removal by SNRAMP in 2006. Therefore, the DEIR cannot claim that the trees designated for removal in the natural areas are unhealthy and/or hazardous.
- There is no reforestation effort outside of Golden Gate Park. Therefore, the DEIR cannot claim that all trees removed in the natural areas will be replaced.
3. THERE IS NO EVIDENCE THAT NON-NATIVE TREES ARE “INVASIVE”
The DEIR also justifies the destruction of thousands of trees on the grounds that they are non-native and “invasive:”
Further, most of the trees within the Natural Areas are nonnative and most are also invasive. The invasive forests within the Natural Areas are predominantly eucalyptus, although cypress, pine, and acacia also occur. (DEIR, page 456)
In fact, there is no evidence that any of these trees are “invasive.” Although, the California Invasive Plant Council has classified eucalyptus as “moderately invasive,” there is no scientific evidence to support this claim. According to the US Forest database of plants and trees, “It [Blue gum eucalyptus] does not spread far and rarely invades wildlands.”
William Russell (USGS) and Joe McBride (UC Berkeley) (Russell and McBride 2003) used aerial photos of Bay Area parks taken over a 60 year period from 1939 to 1997, to study changes in vegetation types. They studied photos of 3 parks in the East Bay (Chabot, Tilden, Redwood), 2 parks in the North Bay (Pt Reyes, Bolinas Ridge), and one on the Peninsula (Skyline).
These photos revealed that grasslands are succeeding to shrubland, dominated by native coyote brush and manzanita. (They also noted that this conversion increases fire hazards.) Eucalyptus and Monterey pine forests actually decreased during the period of study. In those cases in which forests increased in size, they were native forests of oaks or Douglas fir. In other words, they found no evidence that non-native trees are invading native trees or shrubs in open spaces in the Bay Area.
The California Invasive Plant Council classifies Acacia dealbata (Silver wattle) as “moderately invasive” and the impact of Acacia melanoxylon (Black acacia) as “limited” and adds “impacts are low in most areas.” In fact, acacia does not spread unless it is cut down when it then resprouts vigorously from the roots unless it is poisoned repeatedly or the roots are dug out of the ground with heavy equipment.
Neither Monterey cypress nor Monterey pine are invasive. Even the California Invasive Plant Council agrees with that assessment. And both are California natives with fossil evidence that they existed on the San Francisco peninsula in the distant past.
On Mt. Davidson, plans to destroy 1,600 trees over 15 feet tall include many Monterey cypresses. In this particular “natural area,” it is therefore not accurate to say that “most” trees that will be removed are invasive. Table 6.2-1 in SNRAMP claims that only .10 acres of Mt. Davidson are forested with Monterey cypress. This is not accurate. Acres of Monterey cypress on Mt. Davidson are much greater. Since these species are also native to California and have existed in San Francisco in the past, it is an exaggeration to call them non-native.
The final EIR must provide scientific evidence that the trees that will be destroyed by SNRAMP are invasive or it must delete this justification for their destruction.
4. TREE REMOVALS WILL CHANGE WIND PATTERNS, CAUSING TREE FAILURES
The DEIR does not evaluate wind impacts of removing thousands of trees in the natural areas because:
“This section does not address wind impacts in certain Natural Areas because trees targeted for removal are isolated individuals or small groups scattered throughout these Natural Areas, and this removal is not expected to have noticeable wind effects.” (DEIR, page 243)
It is patently false that “trees targeted for removal are isolated individuals or small groups scattered throughout” the natural areas. Here are a few examples of the large number of trees that will be removed from small areas (SNRAMP, Forestry Statement, Appendix F-14-F-17):
• Mt. Davidson: 1,000 trees will be removed from MA-1c (3.5 acres)
• Glen Canyon: 100 trees will be removed from MA-2e (.6 acres)
• Sharp Park: 1,476 trees will be removed from MA-2j (5.6 acres)
• Corona Heights: 10 trees will be removed from MA2C (less than .01 acre)
• Bayview Park: 140 trees will be removed from MA-1d (.02 acres)
• McLaren Park: 600 trees will be removed from MA-2b (9.9 acres)
• Interior Greenbelt: 100 trees will be removed from MA-2a (1 acre)
Tree removal on this scale cannot be done piecemeal, taking only a few individual trees on separate occasions. The game of “pick-up-sticks” is a good metaphor to understand the problem. Felling one tree will impact those in close proximity. Whether intended or not, neighboring trees are likely to be felled by the falling tree.
Removing that individual tree will not be possible unless its neighbors are also removed. One can’t pick up that felled tree when other standing trees surround it. Felled trees must be dragged out. As a qualified arborist said, when expressing his opinion of the proposed selective method of tree removal, “I don’t have tweezers to pick these trees out of the forest.”
Leaving the felled trees on the ground until they are all destroyed is not an option because the restoration objective is to plant the bared ground with native plants, which can’t be accomplished if the ground is covered with dead trees.
Furthermore, destroying a large number of trees slowly, over time substantially increases the cost of such tree removals. Even if it were physically possible to remove them piecemeal, it would not be a responsible use of the limited resources of the Recreation & Park Department which frequently justifies the poor quality of its service and the maintenance of San Francisco’s parks on the grounds that their budget is insufficient.
Even if it were true that only “small groups of trees” will be removed, it does not follow that tree removal would not include wind-toughened edge trees nor that removals “would not result in increased wind hazards or expose trees…to high winds.” In fact, most of the tree removals will occur on the edge of the existing forest, which is consistent with the stated goal of the removals to expand the adjacent native scrub and grassland (SNRAMP, Forestry Statement, pages F-8-F-11):
• Mt. Davidson: “Additional removals will occur…on the eastern edge of the forest.”
• Bayview Park: “Tree removal will focus on the existing edge of forests…”
• McLaren Park: “…removal will occur along forest edges…”
• Interior Greenbelt: “Tree removal will focus on the eastern border and the western tip of this Natural Area…”
In addition to these narrative descriptions of the location of tree removals, SNRAMP contains detailed maps of the natural areas in Section 6 that indicate the location of the tree removals. These maps reveal the vulnerability of the remaining trees as a consequence of some of the tree removals.
In their “Assessment of Urban Forestry Operations” for the Recreation and Park Department, Hort Science reminds us of the vulnerability of the trees that remain after their neighbors are gone and they are exposed to more wind than they have developed defenses against: “As individual trees die or fall, it exposes remaining trees to higher wind loads and increases the overall failure rate.” (page 27)
In December 2011, Hort Science provided us with a specific example of such an occurrence in their “Stern Grove-Pine Lake Park, Parkside Square tree risk assessment.” This report was written as an update of Hort Science’s comprehensive assessment of all trees in Stern Grove-Pine Lake in 2003, in preparation for finally removing the hundreds of trees that had been evaluated as hazardous 8 years before. Here is what Hort Science found at the “West end of the park, near Wawona and 33rd Ave:”
“This area had a number of trees removed by the Natural Areas Program. Subsequently a large Monterey pine failed at the edge of the newly exposed woodland. Concern was expressed about the exposed nature of the edge and potential for additional failures. This area of Pine Lake Park is exposed to westerly winds. There is still, however, significant tree canopy at street-edge. Pines have been declining for some time. Tree #1057 is posted for removal. Mid-slope is a standing dead pine #347, also recommended for removal.”
These trees were cut down in order to expand the native plant garden around Pine Lake. Hort Science had evaluated all trees in Stern Grove/Pine Lake for hazards about one year before these trees were removed. We know those trees were not hazardous, because they had not been judged to be hazardous by the Hort Science evaluation done in 2003.
In other words, as a result of trees removed at the west end of Stern Grove by the Natural Areas Program, a large Monterey pine fell across the path around the lake. Hort Science doesn’t mention where the tree fell, but park visitors remember this failure well. It was not a tree that had been previously designated by Hort Science as hazardous. It became hazardous because its wind break was compromised by the removal of trees by the Natural Areas Program.
The removal of 1,600 trees over 15 feet tall on Mt. Davidson will substantially increase windthrow hazards. Although the DEIR denies this risk, the Forestry Statement in SNRAMP acknowledges it:
“Because of this, removal of edge trees on the northwest side of the park (MA-1c and MA-2c) could increase the rate of windthrow within the stand. Substantial tree removal in these areas should not occur. A significant number of mature trees should remain at the park edge to minimize the effects of wind on this stand.” (SNRAMP, Appendix F, page F-11)
Despite this warning that “substantial tree removal…should not occur” in MA-1c and MA-2c areas, just three pages later in Table F-1, SNRAMP reports these tree removals in these areas on Mt. Davidson:
In other words, SNRAMP predicts tree failures on Mt. Davidson from removal of as many as 82% of all trees on 3.5 acres on the northwest side of Mt. Davidson, yet it plans to remove 1,200 trees from those areas anyway.
The Forestry Statement of SNRAMP also attempts to minimize the danger of windthrow on the grounds that the trees are not near residential neighborhoods: “In general, potential windthrow hazard to people is minimal because there are no residential areas near the stands where the tree removals will occur.” (Forestry Statement, Appendix F-11). This irresponsible excuse for endangering the public does not acknowledge that there are people visiting these parks and taking their lives in their hands by doing so. The death of a park visitor to Stern Grove in 2008, is apparently insufficient testimony to the way the public is being endangered by these tree removals.
Either the planned tree removals must be decreased to reduce the risk of tree failures caused by windthrow or the final EIR must acknowledge the significant risk of these tree removals.
5. TREE REMOVALS WILL INCREASE RUN-OFF, RESULTING IN EROSION AND LANDSLIDES
The DEIR concludes that the implementation of SNRAMP will not cause increased sedimentation, reduced water quality, erosion, or increased run-off, for example:
“The potential for erosion would be less than significant through implementation of the GR-12a (revegetate steep slopes) and GR-12b (phased invasive species removal to reduce erosion), erosion control measures and the erosion and sediment control BMPs described in M-HY-1.” (DEIR, page 374)
We will examine each of these assumptions in the light of scientific studies and our actual experience with the Natural Areas Program.
Revegetating steep slopes will not prevent erosion and increased run-off
The DEIR claims that increased run-off and erosion will be prevented by revegetating areas in which non-native plants and trees are eradicated. This claim is based on these erroneous assumptions:
- That native plants will quickly occupy the bare ground on which they are planted.
- In the 15 years in which the Natural Areas Program has been engaged in its enterprise, it has not successfully vegetated the bare ground created by eradicating non-native plants and trees. Denuded areas are quickly occupied by annual grasses that die back to leave bare ground during the dry season.
- That grassland and dune scrub and non-native trees are equally capable of absorbing run-off and stabilizing soil.
- This assumption is contradicted by the following scientific studies:
- “Results indicate that smoothing of precipitation intensities may translate into overall greater stability of hillslopes under forest canopies. In general, peak intensities of through-fall were damped in intensity and lagged in time relative to peak intensities of rainfall. Damping and lagging of rainfall intensity at both study sites generally increased modeled slope stability relative to openings (areas with no canopy).” (Keim & Skaugset 2003)
- “The reinforcement of the main body of a dike by a grove of trees is much higher and effective in comparison to the reinforcement of the top soil layer by a grass sward. The increase in stability against landslides was found to be at least ten times higher.” (Lammeranner & Meixner 2009)
- This assumption is contradicted by the following scientific studies:
LEAVING TREE STUMPS IN THE GROUND WILL NOT PREVENT EROSION
The DEIR also claims that the removal of trees will not result in erosion because: “…tree removal would be selective, would be implemented gradually over several years, would involve limb-by-limb removals, and would leave tree stumps and root balls intact.” (DEIR, page 364) These claims are inconsistent with SNRAMP, incredible, and/or contradicted by scientific studies:
- As we have already discussed, trees have been selected for removal by SNRAMP in large groups wherever they shade native plants. Some of these groups are as large as 1,000 trees on 3.5 acres (Mt. Davidson). Such removals cannot be accurately described as “selective.”
- It is simply not believable that 18,500 large trees will be removed “limb-by-limb.” What public entity would ever be in a position to pay for such a laborious removal? How is it even physically possible to remove 15,000 trees in Sharp Park “limb-by-limb?”
- Nor is it believable that 18,500 trees will be taken down piecemeal over a long period of time. This would be both physically difficult and prohibitively expensive.
- Leaving “tree stumps and root balls intact” does not prevent erosion. There is considerable scientific evidence that erosion results when the roots die:
- “The immediate effect of deforestation is, therefore, favorable, but adverse effects become evident when root systems decay and when a drop in evapo-transpiration causes a rise in the ground water table.” (Brown & Sheu 1975)
- “Measurement of the decline in tensile strength of small roots in coastal British Columbia after death of the parent tree indicates that over half the strength is lost within 3 to 5 years after cutting.” (O’Loughlin 1974)
- “Soil strength increased linearly as root biomass increased. Forests clear-felled 3 years earlier contained about one-third of the root biomass of oldgrowth forests.” (Ziemer 1981)
- “Decay of tree roots subsequent to logging was found to cause a reduction in the shear strength of the soil-root system.” (Wu, McKinnell & Swanston 1979)
The DEIR’s assumption that increased run-off and erosion will not result from the implementation of SNRAMP does not take into account that the potential for both run-off and erosion are significantly increased by the steepness of slope. Some of the planned tree removals will occur in very steep terrain:
These are only examples of the steepness of slopes in many of the natural areas. The DEIR should be morally and legally obligated to evaluate the steepness of all of the natural areas in the context of the potential for increased run-off and erosion resulting from the removal of non-native trees.
The potential for increased run-off and erosion is greatly increased by steep slopes. The DEIR has not considered that many of the planned tree removals will occur in very steep locations. Some of these locations are directly uphill of densely populated residential neighborhoods which are in the direct path of both run-off and landslides caused by erosion. Yet, the risks to these residential neighborhoods have not been considered by the DEIR. The residential neighborhoods surrounding Mt. Davidson are particularly vulnerable to increased run-off, erosion and landslides.
On May 23, 2012, the State of California sued the US Army Corp of Engineers to challenge a national policy “requiring the removal of virtually all trees and shrubs on federal levees.” Donald H. Gray, Professor of Civil and Environmental Engineering at the University of Michigan, explains why California is fighting this federal requirement in this summary of his literature search about the role trees play in stabilizing soil:
“In the long run, cutting of trees on slopes leads to a gradual decrease in mass stability as a result of the decay of roots which previously acted as tensile reinforcements on the slope. Root decay can also lead to the formation of pipes in slopes which promote internal or seepage erosion. The removal of tree canopy results in the loss of interception and evapo-transpiration which tends to promote wetter and less secure slopes. Canopy removal also results in less attenuation in the delivery rate of rainfall to the ground surface.”
The City and County of San Francisco should consider the implications of this suit. If the State of California is willing to sue to keep trees on its levees in order to prevent erosion and flooding, what are the prospects that the City and County of San Francisco can successfully defend itself against a legal challenge to its plans to remove 18,500 mature trees from the parks managed by the City of San Francisco?
The City and County of San Francisco is particularly vulnerable to legal challenges from the City of Pacifica regarding its plans for Sharp Park. SNRAMP plans the removal of over 15,000 trees over 15 feet tall in Sharp Park. In many management areas 75% of the trees will be removed. These trees will be replaced by dune scrub. The majority of these trees will be removed from the steep watershed at the eastern end of the park. The park slopes from 750 feet above sea level at its eastern end to sea level at its western end. The golf course, archery course, Laguna Salada, and horse pond are downstream from this steep watershed.
Tree removals will violate Pacifica’s logging ordinance. The DEIR claims that the City of San Francisco is exempt from this law, but provides no explanation for or evidence to support this claim. The final EIR must explain why San Francisco is not subject to Pacifica’s laws.
The final EIR must provide evidence that it is physically possible to remove tens of thousands of trees from a steep watershed without causing sedimentation, erosion, and landslides.
The final EIR must provide evidence that the endangered species that exist in Sharp Park will not be harmed by increased sedimentation, erosion, and landslides resulting from the removal of 75% of the trees in the watershed.
The final EIR must evaluate the risk of increased run-off, erosion and landslides. It must substantiate its baseless claims that the removal of thousands of trees will not increase this risk, using scientific studies. If the final EIR cannot provide scientific evidence that these tree removals will not increase these risks, it must mitigate these risks by decreasing plans for removal in natural areas where the risks are great because of steepness and/or the proximity of residential properties endangered by the tree removals.
6. THE IMPLEMENTATION OF SNRAMP WILL RESULT IN A SIGNIFICANT LOSS OF STORED CARBON AND INCREASED AIR POLLUTION
The urban forest of San Francisco stores 196,000 tons of carbon and adds to that accumulated store of carbon at an annual rate of 5,200 tons per year according to the US Forest Service survey. (Nowak 2007) About 25% of the annual rate of sequestration and the accumulated storage of carbon are accomplished by the blue gum eucalyptus, the chief target for destruction by SNRAMP. When a tree is destroyed, it releases the carbon that it has accumulated throughout its lifetime into the atmosphere as Carbon Dioxide (CO₂) as it decays. Carbon Dioxide is the predominant greenhouse gas that is causing climate change.
Since greenhouse gases are regulated in California by a law that commits the state to reduce greenhouse gas emissions, the Draft Environmental Impact Report (DEIR) for the Natural Areas Program (NAP) goes to great lengths to make the case that destroying thousands of trees will not violate California law. The DEIR’s claim that the implementation of SNRAMP will not contribute to greenhouse gas emissions is based on:
a) Fabricating facts by misrepresenting scientific studies. The facts are:
- Grassland in San Francisco does NOT lower ground temperature
- Grassland does NOT store more carbon than forests
The DEIR confuses the RATE of carbon sequestration with the total accumulated carbon storage in the plant or tree as it continues to grow. While a young tree may sequester carbon at a faster RATE while it is growing rapidly that does not alter the fact that a mature tree stores more carbon over its lifetime as the carbon accumulates.
Replacing mature trees with ANY plant or tree will never compensate for the loss of the carbon stored in the trees that will be destroyed. Managing the forest by thinning and reforestation does NOT compensate for the loss of carbon stored in mature trees
GRASSLAND IN THE SAN FRANCISCO BAY AREA DOES NOT LOWER GROUND TEMPERATURE
The DEIR claims:
“According to a study presented at the American Geophysical Union’s meeting, grasslands above 50 degrees latitude reflect more sun than forest canopies, thereby keeping temperatures lower by an average of 0.8 degree Celsuis.” (DEIR, page 457, cited study )
This statement in the EIR does not apply to the San Francisco Bay Area and the reference used to support it misrepresents the cited study:
- The entire continental United States, including the San Francisco Bay Area, is below 50 degrees latitude. In other words, this statement—even if it were true—does not apply to the San Francisco Bay Area.
- The statement is taken out of the context of the article. The entire sentence in which this statement appears actually says, “Grassland or snowfields, however, reflected more sun, keeping temperatures lower. Planting trees above 50 degrees latitude, such as in Siberia, could cover tundras normally blanketed in heat-reflecting snow.” It does not snow in the San Francisco Bay Area. Therefore, this statement does not apply to the San Francisco Bay Area.
- The article being quoted by the DEIR is NOT the scientific study, but rather a journalistic article in The Guardian, a newspaper in England, in which the author of the study has been misquoted and his study misrepresented.
- The day after this article appeared in The Guardian (and also in the New York Times), The Guardian published an op-ed (which also appeared in the New York Times) by the author of the scientific study, Ken Caldeira in which he objected to the misrepresentation of his study:
“I was aghast to see our study reported under the headline “Planting trees to save planet is pointless, say ecologists.” (December 15). Indeed, our study found that preserving and restoring tropical forests is doubly important, as they cool the earth both by removing the greenhouse gas carbon dioxide from the atmosphere and by helping produce cooling clouds. We did find that preserving and restoring forests outside the tropics does little or nothing to help slow climate change, but nevertheless these forests are a critical component of Earth’s biosphere and great urgency should be placed on preserving them.” (Caldeira 2006)
As if this misrepresentation of the facts weren’t bad enough, we find in Appendix A of the DEIR that this isn’t the first time that someone has informed the authors of the DEIR that this statement is not accurate. One of the public comments submitted in 2009 in response to the Initial Study quotes Ken Caldeira’s op-ed in the New York Times. Yet, two years later, the DEIR persists in repeating this misrepresentation of Professor Caldeira’s (Stanford University) research.
GRASSLAND DOES NOT STORE MORE CARBON THAN FORESTS
The DEIR also claims:
“Research studies have concluded that grassland and scrub habitat could act as a significant carbon sink.” (DEIR, page 457, cited studies )
Once again, the cited study does not support the statement in the EIR:
- Again, the statement has been taken out of context. The entire sentence reads, “We conclude that grasslands can act as a significant carbon sink with the implementation of improved management.” This sentence appears in the abstract for the publication. (Conant 2001)
- One wonders if the authors of the DEIR read the entire article or just the abstract. The point of the study is that land management techniques such as fertilization, irrigation, introduction of earthworms, plowing and fallow methods, etc., can improve the sequestration of carbon in the soil of croplands and pastures. This is obviously irrelevant to the Natural Areas Program, which is not engaged in agriculture or pasturage.
- However, the study is relevant in one regard. It reports that when forest is converted to grassland, no amount of “management techniques” compensates for the loss of the carbon in the trees that are destroyed: “Though more than half of the rain forest conversion studies (60%) resulted in increased soil Carbon content, net ecosystem Carbon balance…decreased substantially due to the loss of large amounts of biomass carbon.” (Conant 2001)
The second study cited in support of the claim about carbon storage in grassland reports that increased levels of Carbon Dioxide in the air increases carbon accumulation in the soil. This study tells us nothing about the relative merits of grassland and forests with respect to carbon storage. (Hu 2001) Another study reports a similar relationship between global warming and carbon storage in trees: “…warmer temperatures stimulate the gain of carbon stored in trees as woody tissue, partially offsetting the soil carbon loss to the atmosphere.” (Melillo 2011)
THE DEIR CONFUSES THE RATE OF CARBON SEQUESTRATION WITH THE TOTAL ACCUMULATED STORAGE OVER THE LIFE OF THE TREE
The DEIR claims that because a young tree, growing at a faster rate than a mature tree, sequesters carbon at a faster rate than a mature tree, it follows that replacing mature trees with young trees will result in a net carbon benefit. This is NOT a logical conclusion, as illustrated by this graph from the US Forest Service survey of San Francisco’s urban forest (Nowak 2007).
This graph tells us that although trees sequester carbon faster when they are very small, the large, most mature trees are also sequestering carbon and they store far more carbon than the smaller trees. This is as we would expect, because the total amount of carbon stored within the plant or tree is proportional to its biomass, both above ground (trunk, foliage, leaf litter, etc.) and below ground (roots).
Even IF it were possible replace the non-native trees with native trees—and it’s NOT–the native trees would be significantly smaller than the trees that will be destroyed. The few trees that are native to San Francisco are ALL small trees, compared to the trees that will be destroyed. The Natural Areas Program reports that they have planted 8 species of native trees in the “natural areas” since 2008. Of those 8 species, only one (Red Alder) is classified as a tree by the USDA plant database. The other 7 species are classified as “tree/shrub,” indicating their small stature and low branching habit. Since the amount of carbon stored within the tree is proportional to its biomass, the native trees would never sequester as much carbon as the trees that will be destroyed by the implementation of SNRAMP.
In its zeal to exonerate SNRAMP from releasing carbon stored in the trees it proposes to destroy, it contradicts itself, i.e., that SNRAMP proposes to destroy all non-native trees less than 15 tall. These are the very same young trees that the DEIR says are capable of sequestering more carbon than mature trees. If, indeed, carbon storage could be preserved by a forest of exclusively young trees—and it CAN’T—what is the point of destroying all the young non-native trees?
THE DEIR DOES NOT ACCOUNT FOR THE LOSS OF THE CARBON IN THE TREES THAT WILL BE DESTROYED
If we were starting with bare ground, it might be relevant to compare carbon sequestration in various types of vegetation, but we’re not. We’re talking about a specific project which will require the destruction of thousands of non-native trees. Therefore, we must consider the loss of carbon associated with destroying those trees. It doesn’t matter what is planted after the destruction of those trees, nothing will compensate for that loss because of how the trees will be disposed of.
The fate of the wood in trees that are destroyed determines how much carbon is released into the atmosphere. For example, if the wood is used to build houses the loss of carbon is less than if the wood is allowed to decompose on the forest floor. And that is exactly what this project proposes to do: chip the wood from the trees and distribute it on the forest floor, also known as “mulching.”
As the wood decomposes, the carbon stored in the wood is released into the atmosphere as carbon dioxide: “Two common tree disposal/utilization scenarios were modeled: 1) mulching and 2) landfill. Although no mulch decomposition studies could be found, studies on decomposition of tree roots and twigs reveal that 50% of the carbon is lost within the first 3 years. The remaining carbon is estimated to be lost within 20 years of mulching. Belowground biomass was modeled to decompose at the same rate as mulch regardless of how the aboveground biomass was disposed” (Nowak 2002)
Furthermore, the process of removing trees releases stored carbon into the atmosphere, regardless of the fate of the destroyed trees: “Even in forests harvested for long-term storage wood, more than 50% of the harvested biomass is released to the atmosphere in a short period after harvest.” (Anderson 2008)
The DEIR claims to have run a model of carbon loss resulting from the project in Sharp Park: “The model returns the CO₂ emission rates for all equipment deliveries, and worker activity involving on-road and off-road gasoline and diesel fuel use.” (DEIR, page 455). The CO₂ emissions resulting from the destruction of 15,000 trees over 15 feet tall in Sharp Park is conspicuously absent from their analysis.
MANAGING THE FOREST BY THINNING AND REFORESTATION DOES NOT COMPENSATE FOR THE LOSS OF CARBON STORED IN THE TREES THAT WILL BE REMOVED
The DEIR claims that improving the health of the urban forest by thinning and reforestation with young trees—which will NOT be physically possible–will result in a net benefit of carbon storage.
In fact, the more open canopy of an urban forest with less tree density results in greater growth rates. (EPA 2010) Although more rapid growth is associated with greater rates of carbon sequestration, rates of storage have little effect on the net carbon storage over the life of the tree. (Nowak 1993) Net carbon storage over the life of the tree is determined by how long the species lives and how big the tree is at maturity. These characteristics are inherent in the species of tree and are little influenced by forest management practices such as thinning. (Nowak 1993)
More importantly, even if there were some small increase in carbon storage of individual trees associated with thinning, this increase would be swamped by the loss of the carbon in the trees that will be destroyed.
THE DESTRUCTION OF THOUSANDS OF TREES WILL INCREASE AIR POLLUTION
According to the US Forest Service survey of San Francisco’s urban forest, “It is estimated that trees and shrubs [of San Francisco] remove 260 tons of air pollution (CO, NO2, O3, PM10, SO2) per year with an associated value of $1.3 million (based on estimated national median externality costs associated with pollutants). Trees remove about 19 percent more air pollution than shrubs in San Francisco.” (emphasis added)
The DEIR provides us with no information about the increase in air pollution which will result from removing thousands of trees over 15 feet tall, untold numbers of trees less than 15 feet tall, thinning the remaining urban forest in the natural areas from approximately 740 trees per acre to less than 200 trees per acre and replacing all those trees with grassland and shrubs which are significantly less capable of reducing air pollution.
The final EIR must correct the following errors of FACT in the DEIR:
- The final EIR cannot claim that all non-native trees that will be destroyed will be replaced with an equal number of native trees because that is neither consistent with the SNRAMP, nor is it physically possible.
- The final EIR cannot claim that all non-native trees that will be destroyed are dead, dying, diseased, or hazardous because they are NOT and the claim contradicts the SNRAMP.
- The final EIR must evaluate the risk of failure of the trees that remain after removal of thousands of trees
- The final EIR must evaluate the risk of increased run-off, erosion and landslides
- The citations used to make bogus claims regarding carbon sequestration must be removed because they are not relevant and they have been misrepresented by the DEIR.
- The DEIR’s presentation of the terrestrial carbon cycle must be corrected because it is inaccurate:
- RATES of carbon sequestration must not be confused with the total accumulated stored carbon in mature trees.
- The final EIR cannot claim that there will be a net carbon benefit of the proposed tree destruction because that claim is inconsistent with the science of the terrestrial carbon cycle
The DEIR has not quantified the carbon stored in the current landscape; has not quantified the carbon released by the planned tree destruction; has not quantified the carbon stored in the resulting grassland and scrub. The claimed “qualitative analysis” does not tell us how much carbon will be released into the atmosphere by the implementation of SNRAMP.
As required by CEQA and California Law AB 32, the final EIR must quantify the loss of carbon resulting from the destruction of thousands of healthy trees, compare that loss to the resulting vegetation (grassland and scrub) and mitigate for the net loss of carbon that is the inevitable outcome of the implementation of SNRAMP.
—, “Urban Forest Plan, City & County of San Francisco,” Urban Forestry Council, Department of the Environment, April 2006.
—, “Resolution adopting the 2011 San Francisco Street Tree Recommendations List,” Resolution No. 003-11-UFC, 2011
—, “Vegetation Management Plan,” The Presidio Trust, 2004
—, Environmental Protection Agency, “Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008,” April 15, 2010., EPA 430-R-10-006
Anderson, J., et. al., “The Potential for Terrestrial Carbon Sequestration in Minnesota, A Report to the Department of Natural Resources from the Minnesota Terrestrial Carbon Sequestration Initiative, February 2008.
Brown, C.B., Sheu, M.S., “Effects of deforestation on slopes,” Journal Geotechnical Engineering Division (ASCE), 1975, 101(2): 142-165
Caldeira, Ken, “Planting trees is far from pointless.” The Guardian, December 16, 2006.
Conant, Paustian, Elliott, “Grassland Management and Conversion into Grassland Effects on Soil Carbon,” Ecological Applications, 11 (2) 2001, 341-355.
Doughty, Robin, The Eucalyptus, 2000
Jacobs, M.R. Growth Habits of the Eucalyptus, 1955
Keim, R.F., Skaugset, A.E., “Modeling effects of forest canopies on slope stability,” Hydrological Processes, 2003, 317: 1457-1467
Lammeranner, W., Meixner, H, “Woody plants on flood protection levees: A contradiction?, Published by Nat’l Resources and Applied Life Sciences, Dept of Civil Engerineering and Nat’l Hazards, Institute of Soil Bioengineering and Landscape Construction, Vienna, Austria, 2009
Melillo, J., Butler, S., Johnson, J., Mohan, J., Steudler, P., Lux, H., Burrows, E., Bowles, F., Smith, R., Scott, L., Vario, C., Hill, T., Burton, A., Zhouj, Y, and Tang, J. Soil warming carbon-nitrogen interactions and carbon-nitrogen budgets. PNAS, May 23, 2011
O’Loughlin, C.L., “The effects of timber removal on the stability of forest soils,” Journal Hydrology, 1974, 13(2):121-134
Russell, William H., McBride, Joe R., “Landscape scale vegetation-type conversion and fire hazard in the San Francisco bay area open spaces,” Landscape and Urban Planning, Volume 64, Issue 4, August 15, 2003, pages 201-208.
Nowak, David, “Atmospheric Carbon Reduction by Urban Trees,” Journal of Environmental Management, (1993) 37, 207-217
Nowak, David, et.al., “Effects of urban tree management and species selection on atmospheric carbon dioxide,” Journal of Arboriculture 28(3) May 2002
Nowak, David, et al., “Assessing urban forest effects and values, San Francisco’s urban forest,” USDA Forest Service, Resource Bulletin NRS-8, 2007. http://nrs.fs.fed.us/pubs/9660
Wu, T.H., McKinnell, W.P., Swanston, D.N., “Strength of tree roots and landslides on Prince Wales Island, Alaska,” Canadian Geotechnical Journal, 1979 16: 19-33
Ziemer, R.R., “Roots and the stability of forested slopes,” Proceedings of International Symposium on Erosion and Sediment Transport in Pacific Rim Steeplands, 1981, No. 132: 343-361