Harvesting and Recolonization of Wild Populations of Oshá (Ligusticum porteri) in Southern Colorado Author(s): Kelly Kindscher, Leanne M. Martin, Quinn Long, Rachel Craft, Hillary Loring, Maged H. M. Sharaf and Julia Yang Source: Natural Areas Journal, 37(2):178-187. Published By: Natural Areas Association DOI: http://dx.doi.org/10.3375/043.037.0207 URL: http://www.bioone.org/doi/full/10.3375/043.037.0207
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. R E S E A R C H A R T I C L E ABSTRACT: Land managers face the challenge of conserving medicinal plants that may be threatened by harvest pressure, often with limited biological information available to inform management deci- sions. Oshá (Ligusticum porteri) is an important medicinal plant whose roots are harvested as an herbal remedy for flu, sore throat, and other illnesses. However, little is known about population structure, root production, or the capacity of oshá to recover from harvest in different environmental contexts. • We compared oshá population structure and root production within a gradient of canopy cover from meadow to forested habitat. We experimentally harvested roots of mature oshá plants and recorded oshá recolonization of pits created by root harvest. Prior to harvest, the number and percent cover of repro- ductive plants and the number of flowering stems per plot were higher in the meadow than in forested Harvesting and habitat. Canopy cover had a significant negative relationship to these variables, suggesting that oshá populations benefit from increased light availability. Average root weight per plant in meadow plots was Recolonization of three times higher than in forested plots. One year after harvest, the majority of all harvest pits across the canopy cover gradient were recolonized by oshá. Our results suggest that oshá population structure and root production are significantly influenced by canopy cover, but that plants have a high capacity Wild Populations for post-harvest recolonization under variable light conditions. These results demonstrate the need to account for environmental factors that influence population structure when addressing concerns about of Oshá (Ligusticum the overharvest of wild medicinal plants.
porteri) in Southern Index terms: Colorado, ethnobotany, medicinal plant, roots, wild-harvest Colorado INTRODUCTION in national grasslands of the Great Plains 1,4 Kelly Kindscher (Kindscher 2016). Managers of federal and Land managers are often tasked with con- other protected lands are often responsible 1Kansas Biological Survey serving species at risk of becoming rare, for conserving populations of medicinally University of Kansas endangered, or extinct. Species may be important plants. 2101 Constant Avenue threatened by a multitude of factors, such Lawrence, KS 66047 as small population sizes, anthropogenic The variable environmental conditions in alteration and destruction of habitat, in- which medicinal plant populations grow Leanne M. Martin1 vasive species, overexploitation, climate may influence their response to harvest and 2 Quinn Long change, or a combination of those or other thus, the relative threat posed by harvest, Rachel Craft1 factors (Wilcove et al. 1998; Hulme 2005). making it difficult for land managers to de- Hillary Loring1 Direct anthropogenic changes to habitat velop generalized conservation strategies. may include fragmentation, logging, or Conservation planning for medicinal plant Maged H. M. Sharaf3 development, among other things (Fahrig species therefore, requires an understand- 1 Julia Yang 2003). ing of basic population structure under different environmental conditions, and the 2Missouri Botanical Garden Land managers must often prioritize the impacts of harvesting on species viability. 4344 Shaw Blvd. conservation of plant species at risk of St. Louis, MO 63110 overexploitation. Medicinal plants are fre- Oshá (Ligusticum porteri J.M. Coult. & quently harvested from natural areas, both Rose) is one such medicinal plant species 3American Herbal Products Association in small quantities for personal use and in facing harvest pressure, and it grows in 8630 Fenton St. #918 large quantities sold to companies that mar- a variety of habitats under variable light Silver Spring, MD 20910 ket herbal products to the public (American conditions (Mooney et al. 2015). Land Herbal Products Association 2012). When managers, particularly from the US Forest • harvest of medicinal plants occurs within Service, have been recently tasked with the jurisdiction of the US Forest Service, “determining the best method of conserva- these plants are considered non-timber tion and of sustainable levels of harvest for forest products (Ticktin and Schackleton this species” (Fitzgerald 2014; Krall 2014). 4 Corresponding author: 2011; Kauffman et al. 2015). Well-known Oshá is an ethnobotanically important [email protected]; (785) 864-1529 examples include ginseng (Panax quinque- medicinal plant with large, pungent, and folius L.), black cohosh (Actaea racemosa distinctively spicy roots that are used in tea L.), and goldenseal (Hydrastis canadensis or tincture to treat influenza, bronchitis, and L.) in Southeastern national forests (Al- sore throat. Oshá roots are wild harvested brecht and McCarthy 2006; Small et al. by Native Americans, Hispanics, herbalists, Natural Areas Journal 37:178–187 2011; Souther and McGraw 2014), and and other individuals for personal use and echinacea (Echinacea angustifolia DC) commercial sale by herbal product com-
178 Natural Areas Journal Volume 37 (2), 2017 panies (Kindscher et al. 2013). substantiate as oshá populations currently affects stage class (seedlings, juveniles, are not tracked by any state or federal con- vegetative, and reproductive plants) distri- Based on data from the American Herbal servation agencies. Additionally, there are butions (and number of flowering stems) Products Association (2012) and our per- no comprehensive management strategies of oshá populations. sonal interviews with three commercial in place for the conservation of this species. 2. Determine whether canopy cover harvesters (two small-scale and one large- affects root weights of vegetative and scale) in 2014, we believe that 1000–2000 The recent concern over conservation of reproductive oshá plants. kg of dried oshá roots are harvested each this species, along with growing market 3. Determine the rate of recolonization year from across the species’ range. At demand and potentially declining natural into harvested pits one year after harvest, present, the price of oshá roots has been supply, all underscore the need for research and whether the rate was affected by canopy increasing substantially due to abnormal that enhances our understanding of oshá cover, total plant density, or reproductive weather and the lack of federal permits, population structure, root production, and plant density in the preceding year. which is limiting the supply. Commer- the ability of plants to recover from har- cial brokers were paying $70–$90/kg in vest. Currently, there is little information METHODS 2015, with small quantities going as high regarding how oshá populations and root as $121/kg, and oshá is currently selling production respond to environmental varia- for as high as $419/kg on the Internet tion in the wild. The only study addressing Study Species (Sharaf 2016). Commercial harvesters this question to date, Mooney et al. (2005), Oshá is a slow-growing, perennial member often sell dried roots in bulk at wholesale found that oshá grows under a variety of of the parsley family (Apiaceae) and its to brokers or herbal product companies. light conditions However, the influence range encompasses the Rocky Mountains, The roots have been used in the US and of canopy cover on population structure from southern Montana and Wyoming in Mexico, are exported to Japan, Germany, and root production is not well known. the north, through Colorado, Nevada, and and elsewhere, and have been included Furthermore, there is no information re- Utah to New Mexico and Arizona, and in more than a dozen patented medicines garding whether individual pits created by significantly south into the Sierra Madre (Felger and Wilson 1995). The majority harvesting oshá roots are recolonized and, of Mexico (Scientific Authority of the of oshá on the market is wild harvested if they are, what factors might influence United States of America 2000; Terrell because commercial production has not oshá recolonization. Oshá wildcrafters we and Fennell 2009; Turi and Murch 2010). been economically viable; it is difficult to interviewed reported that plants regenerate According to herbarium specimen data that cultivate and requires the climatic regimen after harvest. Mooney et al. (2015) found we acquired from regional herbaria (Kind- of high-elevation habitats (Guernsey 2005). that oshá plants recovered after moderate scher et al. 2013), in Colorado it occurs in levels of root harvest based on data col- high-elevation sites ranging from 1829 to Given the difficulties of cultivating oshá lected from individual leaf and flowering 3567 m (6000 to 11,700 feet) (Kindscher and the considerable consumer demand for stalks in 1 m2 plots. However, root yields et al. 2013). Roots are branched taproots its medicinal properties, the wildcrafting and recolonization have not been assessed that produce multiple stems with fibrous of oshá roots has raised concerns about with larger plots or after realistic simulation collars, and large plants are rhizomatous potential threats from overharvesting of wild-harvesting, in which all roots of (Applequist 2005). It thrives in diverse soil (McKeon 1999; Scientific Authority of marketable size are harvested from pits dug types, and is often found in open meadows the United States of America 2000; West around the base of mature plants. Roots of or within groves of aspen, conifers, fir, and and Jackson 2004). Wildcrafting is the marketable size are at least pencil-sized in oak (Scientific Authority of the United harvest of medicinal and edible plants from diameter (1 cm) and 7–10 cm long; most States of America 2000; Turi and Murch wild populations (Letchworth 2001), and are much bigger. Recolonization of oshá 2010). It can be found in markets across concerns about the overharvest of natural could occur from root fragments remaining its range, and is commonly known as oshá populations of medicinal plants are wide- after harvest (Figure 1), or from rhizomes and bear root in English, and chuchupate spread (Castle et al. 2014). While oshá was originating outside of the harvested pit. in Spanish. proposed for inclusion in Appendix II of High recolonization rates would indicate the Convention on International Trade in that oshá may be resilient to harvest, at Endangered Species of Wild Fauna and least at the scale of individual plants. If Field Methods Flora (CITES) (Scientific Authority of recolonization does occur, what factors the United States of America 2000; Cech might be related to recolonization rates? Given the extensive range of oshá through- 2002), it has not been listed. However, Answering these questions would be an out much of the southern Rocky Mountains, oshá is listed as an at-risk species by the important step toward providing sound we focused our work in southern Colora- United Plant Savers, the only US nonprofit management recommendations for the do in the Cumbres Pass area of the San organization focused on medicinal plant sustainable harvest of oshá. Juan National Forest, west of Antonito, conservation (United Plant Savers 2012; Colorado. Substantial wild harvest has Castle et al. 2014). The concern about The objectives of this study were to: been observed by the authors and US declining oshá populations is difficult to 1. Determine whether canopy cover Forest Service personnel in this vicinity.
Volume 37 (2), 2017 Natural Areas Journal 179 Figure 1. An oshá root that was cut below the soil surface with a shovel during harvesting and resprouted the following year.
Our study site was exposed to seasonal cover within and between habitat types. Within each plot we recorded counts and open range grazing by cattle with low to Soils in both the meadow and forested percent cover for specific stage classes of moderate stocking rates, though there was habitats are classified as a Leighcan-Frisco oshá: seedling, juvenile, vegetative, and limited browsing observed on oshá plants. association, which are moist soils that occur reproductive. The total number of flower- We collected oshá population data along on 5%–60% slopes (Soil Survey 2015). ing stalks in each plot was also recorded. a canopy-cover gradient on two sides of The continuity of slope, aspect, and soils Seedlings were designated as plants with a US Forest Service road transecting a between these habitats allowed for a more cotyledons and plants of comparable size, north-facing slope. The downhill side of controlled test of canopy cover effects on given that cotyledons senesce. Juveniles the road had significant mixed spruce–fir oshá populations. were single- or double-stemmed plants tree canopy cover that was never logged that were distinctly larger than seedlings. (hereafter “forested habitat”). The uphill In August of 2012 we established two Vegetative plants had three or more stems side of the road was an anthropogenic parallel transects spaced 10 m apart on and obtained a size equivalent to repro- meadow (hereafter “meadow”) with greatly each side of the road. Each transect con- ductive plants, but were not flowering reduced canopy cover from contracted log- sisted of 10 m × 3 m plots (n = 20) with in 2012. Reproductive plants displayed ging of a mixed spruce–fir tree community a minimum buffer of 2 m between plots, inflorescences or evidence that a flow- in 1991. Rather than uniform differences thus establishing 40 plots on each side of ering stalk had formed during the 2012 in canopy cover among habitats, logging the road for a total of 80 plots. All plots growing season. An individual oshá plant and the subsequent 21 years of spruce and were located between 3100- and 3200-m can be challenging to discern due to its fir growth resulted in a gradient of canopy elevation. rhizomatous root system. Therefore, we
180 Natural Areas Journal Volume 37 (2), 2017 assigned the criteria that one individual the field, had fulfilled significant contracts divided by the number of mature plants plant is 50 cm or less in diameter, and that with several herbal product companies. The harvested). Second, to obtain estimates a plant is treated as a separate individual if other two wildcrafters were smaller scale relevant to consumer markets in which oshá it is greater than 20 cm from the edge of and used the roots for their own work or is sold as dried root, we determined dry another individual. If a plant was greater for herbal classes. All wild harvesters with root weights from fresh root weights that than 50 cm in diameter, it was considered whom we interacted made efficient use of we harvested. We present estimates for the to be two plants. Although these criteria their time in the field by harvesting only meadow and forest habitat separately for may be subjective, they were used con- the largest mature plants. However, they descriptive purposes, because the habitats sistently across plots, and imposing them generally refilled their harvest pits, which serve as a proxy for canopy cover (average was necessary in order to quantify both were sometimes 40-cm deep. They also (± 1 SE) canopy cover was 16.8% (± 1.9%) root production and aboveground variables deliberately left smaller plants because they in meadow plots, and 52.1% (± 2.0%) in in a standardized manner that realistically planned to harvest in the same location forested plots). However, we were unable mimics wild harvest (i.e., a 50-cm radius in the future, or they simply felt that this to test differences statistically as habitat seemed consistent with the size of pits was the right thing to do for ensuring the is not replicated. To measure the ratio of observed in areas where wildcrafters had persistence of populations. Wildcrafters fresh to dry root weight, we weighed 10 harvested oshá roots). selected oshá plants based on large size individual, randomly chosen, fresh roots and ease of access (i.e., in open areas, not of variable sizes and allowed them to dry Harvest intensity treatments were imposed next to large rocks, logs, or fallen trees before reweighing them. Dry root weight to study long-term effects of harvest where it is difficult to dig). We simulated averaged 37% of fresh weight for the roots on oshá populations. Here we describe those methods in our 33% and 66% harvest we harvested at the Cumbres site, which these treatments because they affected treatments. However, all mature plants were is consistent with previous findings that root weight estimates. Assessment of the harvested in the 100% harvest treatment roots dried to approximately one-third of long-term resilience of oshá populations to regardless of size and ease of access. their fresh weights (Guernsey 2005). We different harvest intensities is ongoing and then multiplied fresh root weights by 0.37 will be evaluated in the future. The harvest Harvesting entailed digging as much to estimate the cumulative dry root weight percentages of mature plants (both vegeta- root as possible from mature plants us- for each habitat. Next, to calculate the av- tive and reproductive) were 0%, 33%, 66%, ing standard shovels as the wildcrafters erage dry root weight per harvested plant and 100%, with 20 replicate plots of each used. Every effort was exerted to harvest for each habitat, we divided the cumulative harvest intensity across the canopy-cover all roots of marketable size from mature dry root weight by the total number of gradient. One of every three mature plants plants. Roots were harvested such that mature plants harvested. To estimate the was harvested in the 33% treatment, two of the main roots underneath the plant were dry root weight that could potentially be every three mature plants were harvested in dug as completely as possible, and roots harvested from each habitat type (if all the 66% treatment, all mature plants were that spread laterally were collected if they mature plants had been harvested from harvested in the 100% treatment, and no were easily retrieved. Harvest of a plant every plot), we multiplied the average dry plants were harvested in the 0% control. ceased when diminishing returns of roots root weight per harvested plant by the total Because a minimum of six mature oshá were reached. Size of the pits, thus, varied number of mature plants counted. Since plants was required to differentiate between depending on quantity of roots associated each habitat contained 40 plots (3 m × 10 2 these treatments, prospective plot locations with a particular plant, up to a maximum m; 30 m ), our estimates for each habitat 2 along each transect with fewer than six of a 50-cm radius. Although the pits are based on a cumulative area of 800 m . mature plants were omitted from the study were not of systematic size, our methods We divided this number by the number of and included in the buffer between plots. were necessary due to differences in the plots (40) to generate an average estimate quantity of roots per plant, and because of dry oshá roots per 30-m2 area, and 2 Our harvest methods mimicked those of we were trying to mimic wild harvesting. we estimated dry root mass per 1 m as wildcrafters. To ensure that our methods Soil removed during harvest was replaced well. This information at different scales were realistic, we met with many people in pits to simulate the practices of more could be very useful for determining the who were knowledgeable about the meth- conscientious wildcrafters, such as those economic potential of oshá stands growing ods of oshá wildcrafters. First, we met with we interviewed. in different environments, which will help US Forest Service personnel in both the us better understand the economic under- San Juan and Rio Grande National Forests Root Production Estimates pinnings of harvest pressure. in southern Colorado, who showed us areas where wildcrafters had harvested oshá. In We measured and recorded the combined Individually Tagged Pits addition, we visited with herbal product fresh weight of harvested roots at the plot companies who buy oshá roots. Finally, level for two reasons. First, we wanted to We estimated initial recolonization rates in we visited with three wildcrafters who estimate root production, for which we the meadow and forest habitats at Cumbres shared their techniques with us. One of the calculated the average root weight per Pass to determine whether pits where plants wildcrafters, with whom we spent time in harvested plant (total root weight per plot had been dug in 2012 were recolonized by
Volume 37 (2), 2017 Natural Areas Journal 181 oshá in 2013. For this purpose, we marked across the canopy cover gradient because percent cover of reproductive plants (slope the center of harvested pits with a metal it could be a potentially confounding factor = −0.01; adjusted r2 = 0.24), and number of tag and recorded the x and y coordinates affecting oshá population structure at our flowering stems (slope = −0.02; adjusted r2 of each tag within a plot. The pits from site. To do this, elevation data from a 1/3 = 0.36) had a strong negative relationship 125 harvested plants were tagged within arc-second DEM (USGS NED n38w107 to canopy cover (Table 1, Figure 2). The each habitat. This data on 250 individual 1/3 arc-second 2013 1 × 1 degree ArcGrid) number of vegetative plants (slope = −0.00; oshá plants allowed us to assess the extent was overlaid in a GIS by GPS coordinates adjusted r2 = 0.00) and cover of vegetative of recolonization and whether factors such taken at the southwest corner of each plot plants (slope = −0.00; adjusted r2 = −0.01) as canopy cover, total plant density, or (US Geological Survey 2015). The eleva- were not related to canopy cover (Table reproductive plant density in the preceding tion of each point for each plot was then 1, Figure 2). The number of seedling and year were related to recolonization. Re- extracted from the DEM using the “Extract juvenile plants was negatively related to productive plant density in the preceding Values to Points” tool in the Spatial Analyst canopy cover (slope = −0.01; adjusted r2 year could be important if soil disturbance toolbox in ArcGIS 10.2. = 0.28), but seedling and juvenile cover facilitated germination and establishment had no relationship to canopy cover (slope in harvest pits. Analyses = −0.00; adjusted r2 = 0.05) (Table 1, Figure 2). Aboveground recolonization data was To understand the relationships between collected for all tagged pits in August canopy cover and oshá populations, we per- 2013, the year following harvest. For each formed univariate linear regressions on all Root Weights in Harvested Plots tagged pit we recorded a binary (yes or no) response variables at the plot level. We used response to whether or not recolonization the ‘lm’ function in R version 3.0.2, with Average root weight per plant per plot was of juveniles and vegetative plants occurred, slope position as a covariate, canopy cover negatively related to canopy cover (F1,57 = whether recolonizing plants flowered, and for each plot as the predictor variable, and 4.1, P = 0.05, adjusted r2 = 0.37) (Figure whether any seedlings were present. Re- count and cover for each stage class (adding 2). In total, there were 1225 mature plants colonization by vegetative or reproductive juvenile and seedlings together because (both vegetative and reproductive) present plants could occur if any root fragment there were too few plots with seedlings; across all plots in the study area (Table 2). remaining from the harvested plant rees- only 16 of 80 total plots had seedlings, Of those, we harvested 593 mature plants tablished, or from rhizomes originating and there were no seedlings in any forest total, which equated to 43.66 kg of dried outside of the harvested pit, though these plots), as well as number of flowering root (Table 2). We harvested three times two sources of recolonization could not stems, and average fresh root weight per more dried root mass in the meadow habitat be differentiated without further digging. plant in each plot, as response variables (R than in the forest habitat (Table 2). Plants were considered to be recolonizing if they were inside the pit or within 2 cm Core Team 2013). We transformed all count of the pit edge. If we were unable to see response variables and average root weight Recolonization the pit, but the tag was visible, we assessed per plot using ln+1, and logit-transformed plants rooted within a 30-cm radius of the all percent cover response variables prior Juvenile and vegetative plants recolonized tag. Seedlings could also cause recoloniza- to analyses (Warton 2011). over 65% of the pits in both meadow and tion, but they were so infrequent that they forest habitats (Table 3, Figure 1). How- are discussed separately below. We were To examine the probability that recoloni- ever, the probability of pit recolonization unable to locate five tags and pits, so we zation of pits (a binary, yes or no response overall (i.e., of all stage classes combined) collected data on 245 pits. for each of the stage classes of seedling, was not significantly related to any of the juvenile/vegetative combined, and repro- variables (canopy cover, total plant density, ductive) was influenced by canopy cover, Canopy Cover or flowering plant density in the preceding total plant density, or reproductive plant year) that we tested (all P ≥ 0.44). density in the preceding year, we used To examine the effect of light availability logistic regression. We combined all on oshá population density, flowering DISCUSSION stem density, and average root weight, we data to indicate one yes or no response per pit because there were few seedlings measured canopy cover in each plot with Our results illustrate oshá’s affinity for open and flowering plants overall. We used the a convex spherical densiometer. Canopy environments with lower canopy cover. In ‘glm’ function in R version 3.0.2 (R Core cover was calculated as the average of particular, oshá appeared to allocate more Team 2013). readings taken from the plot center facing energy to aboveground reproduction and in each of the four cardinal directions root biomass in open environments. Oshá (Lemmon 1956). RESULTS is remarkably productive as compared to other medicinal plant roots, such as Echi- Slope Position Oshá population structure was related to nacea angustifolia (Price and Kindscher canopy cover. The number of reproductive 2007). Our forest populations averaged 20 We calculated slope position for all plots plants (slope = −0.01; adjusted r2 = 0.33), g dried root per 1-m2 area, while meadow
182 Natural Areas Journal Volume 37 (2), 2017 Table 1. Statistics (F, P) for regressions of oshá variables on canopy cover. Slope position was included as a covariate in the model.
Number of juvenile Number of Number of Number of Source df + seedling vegetative reproductive flowering stalks df Slope position 1,77 25.8; <0.01 1.0; 0.33 20.8; <0.01 24.6; <0.01 1,57 Canopy cover 1,77 7.1; 0.01 0.6; 0.43 19.4; <0.01 21.8; <0.01 1,57 Juvenile + seedling Vegetative Reproductive Source df percent cover percent cover percent cover Slope position 1,77 5.0; 0.03 0.9; 0.34 6.2; 0.01 Canopy cover 1,77 1.1; 0.30 0.3; 0.59 20.3; <0.01 populations had an average of 60 g dried also influence oshá populations. Grazing oshá seeds prefer moist soil conditions root weight per 1-m2 area. Given that sub- intensity can impact oshá populations, (Panter et al. 2004; Terrell and Fennell stantial differences in aboveground cover with significant declines in cover when 2009), which are promoted by increased can be assessed visually, this suggests over 50% of the population experienced canopy cover. However, lower abundance that harvesters can visually assess prior grazing in other studies (Julander 1968). of reproductive plants and flowering stems to digging, which habitats and plants will Although we did not quantify the effects of suggests a tendency for lower overall yield the most root mass. grazing in our study, the vegetation under seed production with increasing canopy open canopy cover appeared to be more cover, thus decreasing the likelihood of Lower canopy cover, and thus, increased impacted by cattle (dung piles and brows- seedling recruitment relative to the more light availability, led to an increase in the ing) than the vegetation in more closed open habitats. Better understanding factors count and cover of reproductive plants, the canopy cover. The differences in land-use that influence seedling recruitment, and number of flowering stems, and the average history, such as disturbance from logging ultimately development into mature plants, and long-term impacts of burning, could is an important focus for future studies of root weight per plant. The abundance of also result in changes in oshá populations. vegetative plants, however, did not respond oshá population dynamics. Therefore, experiments manipulating mul- to canopy cover, indicating that light is tiple abiotic and biotic factors should be most strongly influencing reproduction. Harvest pits from plants dug in the previous conducted before we can fully understand year were frequently (>65%) recolonized, Specifically, 72% of mature plants flow- the mechanisms driving oshá population ered in our meadow plots, while only 60% presumably either by roots that were dynamics. However, Mooney et al. (2015), inadvertently missed during harvest or flowered in the forest. who studied oshá in meadows and aspen from rhizomes originating outside of the stands, found that a greater proportion of harvested pit. However, both Terry et al. Mooney et al. (2015) found results similar individuals flowered in high light environ- (2011) and Mooney et al. (2015) found to ours when they measured oshá at smaller ments. Our results from different habitat that harvesting of medicinal plants can 2 scales (1-m plots), although they did not types also suggest that canopy cover had result in smaller plants, which could impact find an effect of light on root biomass. a strong influence on oshá populations, populations in the long term. Therefore, it However, Mooney et al. (2015) harvested and these findings are consistent with our will be important to continue monitoring roots in aspen stands only to a depth of personal observations and data collection to assess whether plants that initially 25 cm and recorded data from individual at additional oshá populations in various recolonize following harvest survive to leaves or flowering stalks, whereas our habitat types throughout its range. maturity and contribute to the recovery and study removed all roots of marketable size persistence of harvested oshá populations. from mature plants. These differences in Although the number of seedlings and juve- Although we recorded some seedlings in scale of measurement and harvest likely niles declined with increasing canopy cov- pits, there were very few compared to re- explain our finding of a strong negative er, it is difficult to draw conclusions relating colonizing juveniles and vegetative plants, effect of canopy cover on root biomass, seedling recruitment to canopy cover in this and the seedling life history stage is highly though it is possible that oshá responds study given that the median seedling count vulnerable to environmental stress and differently to seasonal availability in the was zero. This suggests that germination mortality (Leck et al. 2008). Clonal root deciduous canopy of aspen. and recruitment in oshá populations may growth, therefore, appears to be the major be episodic and dependent upon specific source of oshá recolonization following Factors Affecting Oshá Populations climatic and microsite conditions. We did harvest. Since the average root weight not find any seedlings under higher canopy per plant within plots declined linearly Although canopy cover is clearly an import- cover, which is somewhat surprising given with increased canopy cover, it is possible ant factor in this study, other factors may that germination studies have shown that that the influence of canopy cover on root
Volume 37 (2), 2017 Natural Areas Journal 183 Figure 2. Plot-level relationships between canopy cover and juveniles counts (a), juvenile cover (b), vegetative counts (c), vegetative cover (d), reproductive counts (e), reproductive cover (f), flowering stem counts (g), and average root weight per plant in each plot (h) across the three habitat types.
184 Natural Areas Journal Volume 37 (2), 2017 Table 2. Oshá root production based on estimated dry weights. Summaries are presented separately for meadow and forest habitats for descriptive pur- poses only.
Weights summary Meadow Forest Total # of plants dug 338 255 Total dry weight of roots dug (kg) 32.92 10.74 Average dry weight of root per harvested plant (kg) 0.1 0.04 Total # of mature plants present 697 528
Estimated dry weight of all roots present (kg) 67.89 22.23 2 Estimated avg. dry weight of roots present in 30m area (kg) 1.7 0.56 2 Estimated avg. dry weight of roots present in 1m area (kg) 0.06 0.02 recolonization may require more time to CONCLUSIONS ulations, a range of anthropogenic factors, become apparent. Additionally, it is possi- environmental context, and life-history ble that early lateral root recolonization is Environmental influences on population traits must be considered. occurring, but aboveground biomass was structure, combined with harvest pressures, not yet associated with those roots. present unique challenges to land managers ACKNOWLEDGMENTS tasked with the conservation of medicinal plant species. Our results indicate that The work for this project was made Management Implications for Oshá changes in canopy cover have considerable possible by funding from the American and other Medicinal Plant Harvest effects on oshá populations. Open habitats Herbal Products Foundation and the Rio with low canopy cover appear to promote Grande National Forest of the US Forest reproduction and higher root biomass in At least within our study area, open hab- Service. We are grateful for their continued oshá plants. However, other factors related itats with more sunlight benefited oshá commitment to the sustainable harvest of to changes in canopy cover, such as land- stands and root production. Management medicinal plants. A very special thank you use history, could also be important. Initial activities such as timber thinning and wild to Angie Krall, Archeologist, and Lorrie results, one year post-harvest, indicate that Crawford, Environmental Educator, at the fires could increase light and positively many harvest pits were recolonized by benefit oshá populations within their core US Forest Service, for their invaluable oshá. The scarcity of seedlings observed insight, guidance, support, and assistance range, where moisture and soil conditions across the canopy cover gradient suggests are appropriate. Since oshá populations throughout this work. For her invaluable that recruitment is episodic, and that popu- assistance in refining our ideas and set- recolonize from resprouts and adjacent lation recovery following harvest is mostly ting up new plots northeast of Durango, plants, it appears that other factors, rather dependent upon vegetative recolonization. Colorado, we thank Gretchen Fitzgerald, than just harvest, must be considered in Land managers should consider environ- forester for the San Juan National Forest. studying oshá populations that experi- mental variation in conjunction with har- We also thank Andrea Jones, Conejos Dis- ence harvest pressure. Other species, for vest pressure when determining whether, trict Ranger, and Mike Blakeman, Public example Echinacea angustifolia, are also how, and where oshá populations should Affairs Officer, at the US Forest Service capable of regenerating after harvest, yet be permitted for harvest, and also where for their support. Numerous individuals other factors, such as grazing and herbicide, conservation should take priority. When assisted us with field work throughout the also influence populations and harvest sus- concerns are raised about the impact of 2012–2014 field seasons. Students from tainability (Kindscher et al. 2008). Many harvest pressure on medicinal plant pop- Haskell Indian Nation University, Bryn medicinal plant species, such as ginseng, have no ability to resprout or recolonize Table 3. Summary of pit recolonization rates for juvenile and vegetative plants combined, as well as after harvest, but depend upon seed pro- seedlings and flowering plants for the meadow and forest habitats one year after harvest. Summaries duction (Hackney and McGraw 2001). are presented separately for meadow and forest habitats for descriptive purposes only. Multiple factors, ranging from species life-history traits to environmental context, Recolonization type Meadow (%) Forest (%) must be considered when managing lands with populations of species that may be Juvenile and vegetative 79 66 threatened by harvest pressure. Seedlings 13 9 Flowering plants 8 3
Volume 37 (2), 2017 Natural Areas Journal 185 Fragua and Allyson Prue, funded by the Quinn Long is a Botanist/Ecologist at the outliers: A neglected center of biodiversi- National Institute of Health Bridge and Center for Conservation and Sustainable ty. Pp. 36–51 in Biodiversity and Manage- Rise Programs, and University of Kansas Development at the Missouri Botanic ment of the Madrean Archipelago: The Sky student Schuyler Kraus, accompanied us, Garden where his work focuses on the Islands of Southwestern United States and Northwestern Mexico. General Technical providing invaluable assistance with field conservation and restoration of rare plant Report RM-264, US Forest Service, Fort work, research, and data entry. For their species and natural communities. Collins, CO. assistance with field work and dedication Fitzgerald, G. 2014. Personal communication. performing strenuous labor in a challenging Rachel Craft is a Ph.D. candidate in the US Forest Service, San Juan National Forest. terrain, we are grateful for the hard work Sociology Department at the University 10 August 2014. of US Forest Service Archeological Tech- of Kansas whose research focuses on the Guernsey, K. 2005. Final Report. USDA-CREES nicians Ken Frye and Marvin Goad; and use of herbal products for human health. SBIR Phase 2 Project: Determining the US Forest Service Interns Aly Albertson, Commercialization Potential for Oshá (Li- Jacob Baker, Brianna Boyd, Dan Delosso, Hillary Loring is a botanical consultant gusticum porteri). Bryce Frank, Erin Hegberg, Andy May- whose work takes her to Kansas, Colorado, Hackney, E.E., and J.B. McGraw. 2001. Exper- day, Dan Nauman, Brianna Roybal, Joel Texas, and other parts of the world. imental demonstration of an Allee effect in Saaranen, Summer Sanford, Jay Seebach, American ginseng. Conservation Biology and Melanie Weber-Sauer. Maged H. M. Sharaf is the Chief Science 15:129-136. Officer for the American Herbal Products Hulme, P.E. 2005. Adapting to climate change: We greatly appreciated our partnership Association, which is the national trade as- Is there scope for ecological management with the Sembrando Semillas under the sociation for the herbal products industry. in the face of a global threat? Journal of Applied Ecology 42:784-794. guidance of Sandra J. Santa Cruz, Agri- cultural Heritage and Youth Leadership Julia Yang is a recent graduate from the Julander, O. 1968. Effect of clipping on herbage and flower stalk production of three summer Development Director at Canto Al Pueblo University of Kansas Biology Department range forbs. Journal of Range Management Cultural Arts; and the hard work and enthu- and has been working as a Botany Intern 21:74-79. siasm of Carlos Santa Cruz and Sembrando at Yosemite National Park. Kauffman, J., J. Chamberlain, and S. Prisleyb. Semillas volunteers Christian Borrego, 2015. Monitoring abundance and distribu- Fidel Gamboa, Tonya Olivas, and Justine tion of non-timber forest products using the Sanchez. We are additionally thankful for LITERATURE CITED FIA database. Journal of Forestry 113:149. the individuals who devoted their time and Kindscher, K., ed. 2016. Echinacea: Herbal energy to guide and assist us with numer- Albrecht, M.A., and B.C. McCarthy. 2006. Medicine with a Wild History. Springer, ous, many strenuous, field work endeavors, Comparative analysis of goldenseal (Hy- New York. including Trish Flaster, Executive Director drastis canadensis L.) population re-growth Kindscher, K., D. Price, and L. Castle. 2008. following human harvest: Implications for of Botanical Liaisons, LLC; Daniel Gagnon Resprouting of Echinacea angustifolia aug- conservation. The American Midland Nat- ments sustainability of wild medicinal plant of Herbs Etc., in Santa Fe, New Mexico; uralist 156:229-236. Tom Grant of the Mountain Studies Insti- populations. Economic Botany 62:139-147. American Herbal Products Association. 2012. Kindscher, K., J. Yang, Q. Long, R. Craft, and H. tute, Silverton, Colorado, and Adrienne Tonnage Survey of Selected North Amer- Antonsen, AmeriCorps VISTA volunteer Loring. 2013. Harvest Sustainability Study ican Wild-Harvested Plants, 2006–2010. of Wild Populations of Oshá, Ligusticum at the Mountain Studies Institute; Meghan American Herbal Products Association, porteri. Open-File Report No. 176. Kansas Ibach, Director of the Alamosa Community Silver Spring, MD. Biological Survey, Lawrence, KS. Gardens; Susan Leopold, Director of the Applequist, W.L. 2005. Root anatomy of li- Krall, A. 2014. Personal communication, US United Plant Savers; Autumn Summers of gusticum species (Apiaceae) sold as oshá Forest Service, Rio Grande National Forest. Herb Pharm; and oshá enthusiasts Cara compared to that of potential contaminants. 10 August 2014. Priem, Linda Boyd, Matt ‘Roots’ Craig, Journal of Herbs, Spices & Medicinal Plants 11:1-11. Leck, M.A., V.T. Parker, and R.L. Simpson. and Maggie Riggs. 2008. Seedling Ecology and Evolution. Castle, L., S. Leopold, R. Craft., and K. Kind- Cambridge University Press, Cambridge, scher. 2014. Ranking tool created for me- UK. Kelly Kindscher, whose research is focused dicinal plants at risk of being overharvested on medicinal plants and conservation in the wild. Ethnobiology Letters 5:77-88. Lemmon, P.E. 1956. A spherical densiometer for biology, is a Professor in Environmental estimating forest overstory density. Forest Cech, R. 2002. Growing At-Risk Medicinal Science 2:314-320. Studies and Senior Scientist at the Kansas Herbs: Cultivation, Conservation, and Biological Survey, both at the University Ecology. Horizon Herbs Publication, Wil- Letchworth, B. 2001. The Industry of Wildcraft- liams, OR. ing, Gathering, and Harvesting of NTFPs: of Kansas. An Insider’s Perspective. Pp. 128–132 in Fahrig, L. 2003. Effects of habitat fragmentation I. Davidson-Hunt, L. Duchesne, and J.C. Leanne M. Martin, Assistant Researcher on biodiversity. Annual Review of Ecology, Zasada, eds. Forest Communities in the at the Kansas Biological Survey, is an Evolution, and Systematics 34:487-515. Third Millennium, 1–4 October 1999, Ke- Ecologist/Botanist who has published Felger, R.S., and M.F. Wilson. 1995. Northern nora, ON, Canada. General Technical Report extensively on prairie restoration. Sierra Madre Occidental and its Apachian NC-217, US Forest Service North Central
186 Natural Areas Journal Volume 37 (2), 2017 Research Station, St. Paul, MN. Chief Science Officer, American Herbal in S. Shackleton, C. Shackleton, and P. McKeon, K. 1999. Making wise choices for Products Association, Shanley, eds. Non-Timber Forest Products L. porteri’s future. United Plant Savers Small, C.J., J.L. Chamberlain, and S.D. in the Global Context. Springer, Berlin Newsletter, Summer. Mathews. 2011. Recovery of black cohosh Heidelberg, Germany. Turi, C., and S.J. Murch. 2010. The genus Ligu- Mooney, E.H., A.A. Martin, and R.P. Blessin. (Actaea racemosa L.) following experi- 2015. Effects of light environment on recov- mental harvests. The American Midland sticum in North America: An ethnobotanical ery from harvest and antibacterial properties Naturalist 166:339-348. review with special emphasis upon species of Oshá Ligusticum porteri (Apiaceae). Soil Survey. 2015. Web Soil Survey 2015. Nat- commercially known as ‘Oshá.’ HerbalGram 89:40-51. Economic Botany 69:72-82. ural Resources Conservation Service, United United Plant Savers. 2012. Species at Risk. Panter, K.L., R.E. Ashley, K.M. Guernsey, and States Department of Agriculture. Accessed Accessed 20 July 2015.
Volume 37 (2), 2017 Natural Areas Journal 187