Burned Area Emergency Rehabiltation Final Accomplishment Report
|
|  Yüklə 1.41 Mb.
|
MethodsNative Plant Restoration `ohi`a woodlandIn `ohi`a woodlands, restoration of native plants was concentrated in 693 plots that were established along 52 transects that spanned the burn (Figure 1). Plot density was ~ 4 plots/hectare). Transects started at the Mauna Loa Strip Road and extended perpendicular from the road in both directions (40, 220 degrees). Transects were labeled A through Z east (40 degree bearing) or west (220 degree bearing). Circular plots (15 m radius) were established at 50 m intervals along each transect. Addition of 2,000-3,000 seeds/plot of `a`ali`i (a shrub) and 400-600 seeds/plot of mamane (a tree) were conducted in all of the plots. Fifty to a hundred seeds of ko`o ko`o lau (Bidens hawaiiensis) (a shrub) and 200-300 seeds of koa (tree) were added to a subset of plots. Seeds were scattered and mixed in the soil layer (<1 cm depth) by lightly turning over the soil with a rake or potato digger. The first round of seeding into plots occurred between 8/00 to 2/01. A second round of seeding took place in 2002-2003. Recruitment from seed addition was monitored by counting the number of seedlings in 40 plots. Seedlings were categorized by height (<10, 10-50, 50-100, >100 cm height) and a subset of individuals (>10 cm height) was tagged to look at one year survivorship. Twenty monitoring plots were also established away from seeded plots to serve as unseeded control plots. Monitoring of seedling recruitment took place in 8/01 and 6/03. Outplanting of 15 native plants species was conducted in 450 plots (Table 1). Number of outplants varied between 4 to 57 indiv/plot. Average plot density was 23 indiv/plot. Outplants included 13 fire-tolerant plant species. A limited number of `ohi`a and pukeawe were outplanted in the burn. These species are considered relatively fire-intolerant, but were planted because they were the major components in the pre-burn community. Fire tolerant `a`ali`i was not outplanted because recruitment from artificial seeding into plots was expected to be high based on previous studies (Loh et al. unpubl.). Outplanting first began in December 2000 and continued to July 2003 as individuals propagated in the nursery matured and became available for outplanting in the burn. There were a total of thirty outplanting sessions. Survivorship of outplanted individuals, their vigor, and reproductive status, was monitored in 60 plots in 3/02 (halfway through the outplanting sessions) and in 4/04 (8 mo. after the last round of outplanting). Koa-`ohi`a mesic forestRestoration of native understory plants was concentrated in a 50 m x 400 m strip of burned koa forest that lay adjacent to Kipuka Puaulu SEA (Figure 1). The planting area was subdivided into 75 plots (approx 10-15 m radius) within which planting and seeding was concentrated. Eighteen native species were outplanted between 8/00 and 5/03 (Table 1). Total plant density in the 50 x 400 m strip was 2,520 indiv/ha. Plots were seeded with `a`ali`i (500-1500 seeds/plot), mamane (200-600 seeds/plot), and manele (Sapindus saponaria) (50-150 seeds/plot). Herbicide control of meadowrice grass with 1% Roundup in water was performed quarterly inorder to minimize competition from alien grasses while outplants established. Outplant survivorship of nine species was monitored in 3/02 or midway through the project. Monitoring of seedling recruitment from seed addition was not conducted. Young koa standsEstablishment of koa and native understory plants was concentrated in a 50 m x 400 m burned strip that lay ~100m west and parallel to the Park boundary adjacent to the Volcano Golf Course Subdivision. The planting area was subdivided into 60 plots (approx 15 m radius) within which planting and seeding was concentrated. Natural recovery of fire-damaged koa stands was augmented with outplants of koa and 10 other native species between 3/02 to 2/22/03 (1,327 indiv/ha). Seed addition of koa (250-900 seeds/plot), mamane (1,400-2,200 seeds/plot) and `a`ali`i (4,000-10,000 seeds/plot) was conducted in all the plots. No outplant survivorship or seedling recruitment was monitored in this site. Seed collection, seed preparation, seed storage Plant material for propagation and seeding in the burn was collected from individuals and populations located closest to the burn. Material was collected from as many individuals as possible in order to maximize genetic diversity. For common species, material was collected from >100 individuals. For less common species, material was collected from at least 25 individuals. Fewer than 10 individuals were used as source material for locally rare native tree species, alani (Meliocope radiata), and kolea (Myrsine lessertiana) that were outplanted in mesic koa forest. Fruits were generally pulped and removed from the seed before broadcasting into the burn or for plant propagation. Seeds of `a`ali`i remained in their capsules for broadcasting in the burn. Seeds were stored in refrigerators with the exception of the legumes, which were stored at room temperature before they were husked, and `a`ali`i. Plant propagation was conducted in temporary park greenhouses that were constructed for the project. A 100 x 24 ft hoop house, and two 20 x 20 ft nurseries were constructed by March 2001. Twenty-four 4x12 ft benches were built and, combined with 30 benches provided by the park, were devoted to propagation of native plants for outplanting into the burn. Many of the species had not been propagated or outplanted in the park in recent years (e.g. Santalum paniculatum, Charpenteria obovata, Melicope radiata). Plant propagation techniques had to be developed or refined for the rehabilitation project. Propagation techniques, and sanitation protocols developed for the project are currently being summarized in a separate technical report (Mcdaniel pers. comm.). Alien Plant Control Work crews conducted systematic sweeps on foot and removed alien woody species, Himalayan blackberry (Rubus argutus), Himalayan raspberry (Rubus ellipticus), strawberry guava (Psidium cattleianum) and faya tree (Myrica faya) from the entire burn. Individuals were either manually uprooted or stems were cut and Garlon 3a (10% in water) applied to stumps. Treatment was conducted annually between 2001-2003. Monitoring Natural Recovery in the Burn Additional monitoring to evaluate community response to fire and rehabilitation activities was conducted in permanent plots. Sixty 30 x 20 m plots were established in the burn, forty inside planting plots and twenty outside the planting plots to document vegetation changes following fire and management. Half of the plots were located in an area that had been twice burned (2xburn), once in 1977 and again in the Broomsedge Burn. The remainder was established in an area that had only burned once in the Broomsedge Burn (1xburn). Ten additional 30 x 20 m control plots were established in adjacent unburned `ohi`a woodlands. Inside the plots, vegetation cover, native seedling recruitment and tree survivorship were recorded, following modified protocols described in the Fire Monitoring Handbook (NPS 1992). Modifications to the FMH protocols included 1) reduction of plot size (20 x 30 m), and 2) trees categorized by basal diameter rather than DBH. Monitoring was conducted in summer 2001, one year after the fire, and at the end of the project in 2003. Also, recovery of native `ohi`a (n=206) and koa (n = 97) individuals were monitored along 6 transects that spanned the burn. Fire severity (char height, scorch height, percent canopy scorch) on individuals was quantified within the first month following fire, and individual survivorship (as evidenced by regrowth at the base or along the trunk and branches) was monitored one year later. Data Analysis Survivorship of outplanted species was summarized as the percent of live individuals among the total number of individuals that were monitored across all sample plots within a community type (`ohi`a woodland, koa mesic forest). Seedling recruitment from seed additions into plots was summarized as the average number of individuals (by height class, or across all height classes) per plot present in seeded and unseeded plots. Extremely low recruitment in unseeded plots prevented statistical comparison between seeded and unseeded plots. Survivorship of tagged seedlings was summarized as the percent of live individuals among the total number of individuals that were monitored across all sample plots. To evaluate natural recovery of plants in the burn, a one-way ANOVA (SYSTAT ver. 9) was used to compare plant abundances among the 1xburn, 2xburn and unburn plots. For cover and density data, species were grouped and analyzed by life form category (exotic herb, exotic grass, exotic woody, native herb, native woody, native grass). Count data were log-transformed and cover data were arcsine transformed. Tukey multiple comparison tests were performed to identify significant treatment responses across sites. Survivorship of `ohi`a, and koa was summarized as the percent of live individuals (as evidenced by basal or epicormic live foliage) among the total number of individuals that were monitored along transects. Average, minimum and maximum char height, scorch height and percentage of crown scorch of individuals were summarized for each species. Results Native Plant Restoration Treatment Approximately 3000 worker days (1239 volunteer days) were spent on restoration efforts in the burn. The majority of the time was devoted to collecting seeds, propagating individuals, sowing seeds and outplanting in the burn. Eight hundred twenty-eight planting plots were established, 75 in koa-`ohi`a mesic forest, 693 in `ohi`a woodland, and 60 in young koa stands. Nearly nineteen thousand (18,798) individuals of 30 species were outplanted and 2,735,100 seeds of five species were sown in the burn. Survivorship was high among outplants despite no supplemental watering in the field. Average survivorship across all 16 species planted in fire-damaged `ohi`a woodlands was 81% (Table 3). Survivorship among species ranged from a high of 96% for koa trees, to 60-85% for most other species and a low of 32% for `emoloa grass (Eragrostis variabilis) and 33% for pua kala or Hawaiian poppy (Argemone glaucum). The low survivorship of the latter two species is not unexpected since the poppy is an annual species that dies after seed production, and the emoloa was difficult to relocate amid the tall alien grasses (outplants were not flagged). First year outplant survivorship for nine species monitored in mesic koa forest was much higher (>90%) (data not shown). However, monitoring was done midway through the outplanting program and numbers of planted individuals were low for some species (e.g. < 10 indiv). More monitoring is needed to reliably determine outplant success for all species planted in the koa mesic forest. No monitoring was conducted for individuals outplanted in young koa stands. This environment is very similar to `ohi`a woodlands, and outplant survivorship is expected to be similar to survivorship in plots located in fire-affected `ohi`a woodlands.
There was significant seedling recruitment from seed additions into plots for koa, mamane, `a`ali`i, and ko`o ko`o lau located in fire-affected `ohi`a woodlands (Figure 2). Seedling survivorship of tagged individuals >10 cm height between 8/01 to 6/03 was 35% for mamane, 69% for koa, 98% for `a`ali`i, and 100% for ko`o ko`o lau. Recruitment from seed addition into plots was observed but not quantified for manele (Sapindus saponaria), mamane, and `a`ali`i in mesic koa forest, and for mamane and `a`ali`i in young koa stands.
Individuals of ten outplanted species and `a`ali`i were reproductively mature (seed production or root suckers) at the time of the last sampling (4/04) (Table 3). These included 4 tree species (mamane, mamaki, naio, koa by root suckers), 3 shrubs (`a`ali`i, nae nae, ko`o ko`o lau), a lily (`uki`uki), pua kala or the Hawaiian poppy, and 2 grasses (Eragrostis variabilis, Deschampsia nubigena). Seed production was widespread among outplanted ko`o ko`o lau and grasses, but less common for other outplanted species. Outplanted individuals of naio (Myoporum sandwicensis) and mamaki (Pipturus albidus), small trees, were producing fruit in mesic koa plots but not in dry `ohi`a woodland plots. Future monitoring will determine whether seed production will lead to successful seedling establishment in the plots. Alien Plant Treatment Ten percent (311 worker days) of work time devoted on the projects, was spent searching and removing disruptive alien species that threatened native plant recovery in the burn. Over seven thousand (7,409) individuals of Himalayan blackberry (Rubus argutus), Himalayan raspberry (Rubus ellipticus), strawberry guava (Psidium cattleianum) and faya tree (Myrica faya) were found and removed between Summer 2001 to Summer 2003. Spot application of herbicide on meadowrice grass (Erharta stipoides) was used to reduce competition to native outplants in mesic koa forest. Work crews consisted of personnel form the Division of Resources Management, contract employees through cooperative agreements, Youth Conservation Corps, and volunteer biologists. Plant community Response to fire Three years after fire, exotic grasses had re-established rapidly in the 1x and 2x burn areas (Figure 3). Percent cover of exotic grasses was within two-thirds of adjacent unburn areas. Both species richness and cover abundance of exotic herbaceous species were much greater in the burn than in the adjacent unburn. In contrast native woody and herbaceous species had significantly less plant cover in the burn relative to adjacent unburn areas. Density of mature native shrubs, pukeawe and `a`ali`i, was significantly higher in the unburn (8 indiv/plot for `a`ali`i, 148 indiv/plot for pukeawe) than in the 1x burn (<1indiv/plot for `a`ali`i and pukeawe), and 2x burn (<1 indiv/plot for a`ali`i, and 2 indiv/plot for pukeawe). Few fire-affected `a`ali'i and pukeawe survived the burn by resprouting at the base of fire damaged individuals. Average seedling recruitment (<10 cm height) of `a`ali`i was 20-fold greater in the 1xburn (19 indiv/plot) and 2xburn (21 indiv/plot) than in the unburn (1.5 indiv/plot). There was no significant difference in small seedling recruitment (12-65 indiv/plot) of pukeawe between the 1x, 2x burn and unburn. Survival by resprouting was common for ohelo (Vaccinium reticulatum) in the 1x burn (62 indiv/plot) and 2x burn (9 indiv/plot). Among the 97 fire-damaged koa trees that were monitored, greater than 95% were top killed. Char height ranged between 0.7 to 3.1 m height (average char height = 1.8 m), scorch height ranged between 2.6 to 8.0 m height (average scorch height = 5.2 m), percent crown scorch was between 66 to 100% (average % crown scorch = 90%). One year after fire, 55% of individuals had resprouted (92% basal, 8% epicormic). Among fire-damaged `ohi`a (n= 206), char height ranged between 0.4 to 2.8 m height (average char height = 2.3 m), scorch height ranged between 2.0 to 5.1 m height (average scorch height = 4.1 m), percent crown scorch was between 23 to 100% (average % crown scorch = 87%). One year after fire, 68% of individuals had resprouted (resprouting was 86% basal and 26% epicormic among live individuals). Discussion Rehabilitation of the Broomsedge Burn was the first attempt by park managers to restore fire-affected dry `ohi`a woodland communities at Hawai`i Volcanoes National Park. Field and laboratory experiments conducted between1993-2000 had identified fire-tolerant native plant species for restoration into burn sites (Loh in prep.). Techniques for getting species established through artificial seed banks had also been tested. The Broomsedge Burn provided managers the opportunity to apply the knowledge gained in past experiments to restoration of a large burn. Establishment of native plants by outplanting and seeding Native plant establishment through outplanting and seeding appeared successful based on early monitoring results. Outplant survivorship was >80% in fire-affected dry `ohi`a woodlands and >90% in koa mesic forest. This is relatively high survivorship when compared to survivorship or native species planted in other parts of the park, many of which required additional watering in the field (Belfield pers. comm.). High survivorship may be partially due to more favorable moisture conditions. The burn is located on the upper elevational extent of the dry `ohi`a woodlands that transitions into mesic forest. Outplant success is expected to be considerably lower in restoration projects conducted in drier parts of the `ohi`a woodlands. Also, the time that elapsed between when individuals were first outplanted and last monitored varied between 8 mo and 3 years. Consequently, some mortality of recently outplanted individuals may occur in the future. Successful seedling recruitment from artificial seeding into the burn was evident for the five species (mamane, `a`ali`i, koa, ko`o ko`o lau, manele) where seeds were placed in plots. One year survivorship of tagged individuals (mamane, `a`ali`i, koa, ko`o ko`o lau) was 35-100%, and suggests that some seedlings will continue to grow to reach reproductive maturity. Seedling survivorship in the burn was relatively high compared to research sites in other parts of the dry `ohi`a woodlands (Ainsworth unpubl., Vayadati pers. comm.), and maybe partially due to more favorable moisture conditions in the burn. There were other species (e.g. `ili`ahi, na`e na`e) that might have successfully established from artificial seeding, based on observations in experimental burns and life history characteristics, but the amount of seeds available was insufficient for distributing across plots. By 2004, individuals of ten outplanted species and `a`ali`i had grown to reproductive maturity in the burn. Nine species, including `a`ali`i, showed either fruit production or regeneration by root suckers (koa) in `ohi`a woodland plots. Two additional tree species (mamaki, naio) were producing fruit in mesic koa forest plots. These are positive steps toward the long-term establishment of native species in the burn. Particularly for `ohi`a woodland, where a grass-fire cycle has established with the invasion of fire-adapted grasses, establishment of a soil seed bank of fire-tolerant species before another wild fire occurs is required to ensure the long-term persistence of re-introduced species. The assumption, based on observations in previous wildfire and research burns, is that future fires may top-kill individuals, but these fire-tolerant species will survive and proliferate through vegetative regrowth and seedling recruitment from the soil seed bank in response to fire. Prevention of alien woody species establishment Four of the most ecologically disruptive alien species were prevented from establishing in the burn. Over 7,000 individuals of Himalayan raspberry, Himalayan blackberry, faya tree, and strawberry guava were removed between 2001-2003. The intent of the search and control effort was to prevent these species from becoming established during the interim that followed the burn, when the area is largely unoccupied by plants and most vulnerable to invasion by alien species. By 2003, plant cover in the burn had not yet recovered to adjacent unburn levels. Future monitoring will determine whether these four disruptive alien woody species will continue to invade the burn, and whether search and control efforts need to be continued. Natural recovery of preburn community after fire Natural recovery of `ohi`a was relatively high compared to documented recovery in other fire-affected `ohi`a woodland sites in the park (Tunison et al. 1995). Individuals were top-killed, but at least 2/3rds of individuals were able to regenerate from resprouting at the base of trees. `Ohi`a survival through resprouting after fire varied considerably between 0% to 49% in other areas of the park and survival appeared partially dependent on burn severity and tree size (Tunison et al. 1995). Occasional seedlings were observed in the burn, but numbers were insufficient to detect in plots, and it is doubtful that seedlings will contribute much to the future recovery of `ohi`a stands. This may lead to continued stand reduction with recurrent fire in the future. Pukeawe, the dominant native shrub in the preburn community, was also heavily impacted by fire. Almost all individuals damaged by fire in plots were top-killed, and no vegetative resprouting was observed. Future recovery will depend on recruitment from the existing seed bank. Seedling occurrence was highly variable in burn plots. Seedlings grow extremely slow and have difficulty surviving in competition with exotic grasses after fire (Hughes and Vitousek 1993). Future monitoring will determine the extent to which individuals survive and grow in the presence of the exotic grasses that are recovering rapidly in the burn. `A`ali`i and koa are two fire-tolerant species that were present in the preburn communities. Most 'a'ali'i individuals were top killed and very little vegetative resprouting at the base of stems was observed. Seedlings were evident and in greater abundance in burn plots than in adjacent unburn plots. A`ali`i shrub densities and cover increase, primarily through seedling recruitment, in response to fire (D`antonio et al. 2000, Tunison et al. 1995). Fifty-five percent of koa resprouted following fire-damage in mesic koa forest. Koa is a disturbance adapted tree that regenerates through vigorous seedling recruitment and root suckers (Vogl 1969, Spatz 1973). Prolific recruitment from the natural seed bank and root suckers was evident but not quantified in this burn. Conclusions Rehabilitation of native plants in the Broomsedge Burn serves as the model for restoration in other fire-affected dry `ohi`a woodlands in the Park. Over fifteen thousand acres, or two-thirds of dry`ohi`a woodland have been affected by wildfire in the older section of Hawai`i Volcanoes National Park. The remaining unburn areas have been invaded by alien grasses and have the potential to burn which would result in further conversion of `ohi`a woodland to alien grass savanna. Since the Broomsedge Burn, three wildfires (Kupukupu (2002), Panau Iki (2003), Kipuka Pepeiao (2004)) have affected nearly 3,000 acres of dry or transitionally dry `ohi`a woodland. These include previously burned areas and large areas where fire had not historically been documented (Tunison unpubl.). Restoration of fire-tolerant native species has begun in sections of the Kupukupu burn, and plans are being developed to re-introduce fire-tolerant species in Kipuka Pepeiao. Based on the results of this project, the following management recommendations can be made for future restoration of other fire-affected dry ohia woodlands in HAVO:
Long-term monitoring is also needed to evaluate the effectiveness of the two vegetated fuel breaks that were established in the Broomsedge Burn. The concept of building dense vegetation to slow or stop fire spread is new to HAVO managers. Microclimate changes that enhance fire potential have been associated with the conversion of `ohi`a woodlands to alien grass savannas (Friefelder et al. 1998). Currently, individuals planted in the vegetated strips are too small and immature to change microclimate conditions and reduce fuel loads sufficiently to slow fire spread. Monitoring will determine whether the vegetation increases and microclimate changes occur in the future. The ability of the mature fuel breaks to resist fire remains to be tested. Finally monitoring natural recovery in burns should continue. The relatively high survivorship of `ohi`a in the Broomsedge Burn compared to previous studies underscores the need to understand the role site conditions, life history, fuel conditions, and burn severity have in determining successional outcomes. By understanding how these factors influence native species recovery, managers will be able to better predict potential recovery following fire and prioritize sites for management. Literature Cited D`Antonio, C.M. and P.M. Vitousek. 1992. Biological invasions, the grass-fire cycle and global change. Ann. Rev. Ecol. Sys. 23:63-87. D`Antonio, C.M., F.R. Hughes, M. Mack, D. Hitchcock, and P.M. Vitousek. 1998. The response of native species to removal of invasive exotic grasses in a seasonally dry Hawaiian woodland. Journal of Vegetation Science 9:699-712. D`Antonio, C.M., J. T. Tunison, and R.K. Loh. 2000. Variations in impact of exotic grasses and fire on native plant commmunities in Hawaii. Journal of Australian Ecology 25:507-522. Freifelder, R., P. M. Vitousek, C.M. D’Antonio. 1998. Microclimate effects of fire-induced forest/grassland conversion in a seasonally dry Hawaiian woodlands. Biotropica 30:286-297. Higashino, P., and L. Cuddihy. 1988. Checklist of vascular plants, Hawaii Volcanoes National Park. Cooperative Park Studies Unit Technical Report No. 64. University of Hawaii at Manoa, Honolulu, Hawaii. Huenneke, L. F., and P. M. Vitousek. 1990. Seedling and clonal recruitment of the invasive tree Psidium cattleianun: implications for management of native Hawaiian forests. Biological Conservation 53:199-211. Hughes, F, P.M. Vitousek, and T. Tunison. 1991. Alien grass invasion and fire in the seasonal submontane zone of Hawai`i. Ecology 72(2):743-746. Hughes, R.F. and P.M. Vitousek. 1993. Barriers to shrub reestablishment following fire in the seasonal submontane zone of Hawai`i. Oecologia 93:557-563. Loh, R. K. 2004. Early plant establishment following experimental removal of invasive Morella faya (Wilbur) Ait. stands in a Hawaiian forest. Dissertation, University of Hawaii at Manoa, Honolulu, Hawaii. Loh, R. K., A. Ainsworth, T. Tunison, C. M. D`Antonio. In prep. Testing native species response to prescribed fire in Hawaii Volcanoes National Park. National Park Service. 1992. Western Region Fire Monitoring Handbook. US Department of Interior. Parman, T.R. 1976. The effects of fire on a Hawaiian montane ecossytem. In Proc. Of the First Conference in Natural Sciences. Hawaii Volcanoes National Park. C.W. Smith (ed). University of Hawaii, Honolulu. Smith, C.W. 1985. Impact of alien plants on Hawaii s native biota. In Hawaii`s Terrestrial Ecolsystems: Preservation and Management. C.P. Stone and J.M. Scott, eds. University of Hawaii Press. 584 pp. Spatz. 1973. S ome findings on vegetative and sexual reproduction of koa. Island Ecosystem IRP U.S. IBP Technical Report No. 4. University of Hawaii at Manoa, Honolulu, Hawaii. Tunison, J.T., R.K. Loh and J.A.K. Leialoha. 1995. Fire effects in the submontane seasonal zone, Hawai`i Volcanoes National Park. Technical Report 97. Cooperative Parks Resources Studies Unit. University of Hawai`i, Honolulu. Yoshinaga, A. Y. 2002. Seed storage practices for native Hawaiian Plants. http://www2. Hawaii.edu/scb/seed/seedmanual.html. Vitousek, P. M., and L. R. Walker. 1989. Biological invasion by Myrica faya in Hawaii: Plant demography, nitrogen fixation, and ecosystem effects. Ecological Monographs 59(3):247-265. Vogl, R.J. 1969. The role of fire in the evolution of the Hawaiian flora and vegetation. pp 5-60 In Proceedings Tall Timbers FireEcology Conference (9). T  able 1. Natives species for introduction into burned communities. X denotes establishment by outplanting, x denotes establishment by artificial seeding, * denotes fire-tolerant species.
Table 2. Number of outplants and seeds added to the burn by species.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
