Research ArticleRefereed Research

Seedling emergence and seed production of curlycup gumweed

Derek Tilley, Mary Wolf, Derek Jolley and Gordon Hirning
Native Plants Journal, September 2022, 23 (3) 299-308; DOI: https://doi.org/10.3368/npj.23.3.299

Abstract

Native rangeland forbs are in demand among Intermountain restorationists, but they are underrepresented in revegetation projects because of high seed cost, limited availability, and poorly understood establishment requirements. Curlycup gumweed (Grindelia squarrosa (Pursh) Dunal [Asteraceae]) is a short-lived forb, native to semi-arid sites in the Intermountain West, with potential for broadscale use in restoration activities. We conducted a glasshouse experiment examining seedling emergence of curlycup gumweed planted at depths of 0, 1, and 2 cm (0, 0.4, and 0.8 in) in 3 different soil textures (sand, loam, and clay loam). We also compared 3 seed harvest methods of curlycup gumweed in a replicated field trial and evaluated seed yield, purity, and viability. In a third experiment, we tracked viability of seed stored under dry, cool conditions over 3 y using tetrazolium (TZ) staining. Our seedling emergence experiment indicated that maximum emergence can be achieved by planting seed on the soil surface (0 cm) regardless of soil texture. Soil texture had no effect on seedling emergence at any depth. Seed yields from harvesting once via mechanical swathing, simulated using a hedge clipper, were comparable to yields from multiple hand harvests using a racquet-and-hopper, while harvesting with a Flail-Vac resulted in poor yields. Average seed purity and viability did not differ among harvesting methods. Finally, seedling vigor changed little under storage conditions and stayed above 95% over 3 y. Our results suggest curlycup gumweed has excellent potential for larger scale seed production and marketability.

KEY WORDSNOMENCLATURE

Native forbs have been shown to be a critical component of Intermountain Western habitats, and diverse assemblages of native forbs are often desired for restoration and reclamation plantings. Forbs provide a food source for wildlife and pollinator species (Dumroese and others 2016), add ecological resilience against invasion (Pokorny and others 2005; Leger and others 2014), and influence soil chemical properties and nutrient availability (McLendon and Redente 1992). Additionally, successional management, the practice of including species from multiple successional stages in restoration plans, is bringing new and poorly understood native forbs into consideration (Uselman and others 2015; de Queiroz and others 2021). Historically, however, native forb use in seed mixes has been limited in revegetation projects because of high seed cost, limited availability, and poor establishment (Shaw and others 2005). Information regarding forb germination and seed production characteristics is needed to increase establishment rates and seed availability among species of this functional group.

Seed depth, or seed placement, can be critical to achieve desirable establishment, especially with native forbs (Sanderson and Elwinger 2004; Rawlins and others 2009; Bushman and others 2015). Generally, seed placed deeper in the soil is better protected against herbivory, desiccation, and temperature fluctuations; however, seed placed too deeply may germinate but be unable to break through the soil or soil crust. A rule of thumb used by horticulturalists and gardeners is to place the seed to a depth of 2 to 3 times the width of the seed (Vanderlinden 2019). For many small-seeded native forbs and grasses, that may be very close to the soil surface. Further complicating the issue is variation in soil texture, that is, the percentage of sand, silt, and clay in a soil. Soil texture determines differences such as bulk density, porosity, water infiltration, and susceptibility to compaction or crusting. Thus, seed of a certain species may have variable optimum seeding depths for various soil textures (Rawlins and others 2009; Bushman and others 2015). Understanding appropriate seeding depth is crucial to restoration and seed production efforts.

Optimal seed production methodology for a given species is an important determinant of seed availability and cost. Efficient harvesting increases seed availability and reduces cost for the producer and end user. Native forb seed is often quite expensive compared to grasses, and high seed cost can be a limiting factor in forb seed use in restoration efforts (Richards and others 1998). Many native forbs require specialized seed production techniques and are produced only in small quantities (Bartow 2014; de Queiroz and others 2021). For example, seed of some species, such as shy gilia (Gilia inconspicua (Sm.) Sweet [Polemoniaceae]), may be most effectively harvested by hand or by uprooting the entire plant and provides yields of less than 100 kg/ha (89 lb/ac) (de Queiroz and others 2021). Seed of other species may be collected by allowing the seed to shatter in place and picking up the seed from the ground with brushes or vacuums (Bartow 2014). Many native forbs, however, possess attributes that are better suited to large-scale agronomic production. Western yarrow (Achillea millefolium L. [Asteraceae]) and firecracker penstemon (Penstemon eatonii A. Gray [Scrophulariaceae]) seed yields, for example, can range in the hundreds of kg/ha (Cornforth and others 2001).

Seed viability under storage conditions is a further factor in seed availability and cost. Seed with a short shelf life must be sold quickly to avoid losses in viability, and seed producers may hesitate to risk producing more seed than can be sold within the seed’s life span. Some species used in the Intermountain West, such as sagebrush (Artemisia tridentata Nutt. [Asteraceae]) and forage kochia (Bassia prostrata (L.) A.J. Scott [Chenopodiaceae]), have very short storage life spans (Kitchen and Monsen 2001; Karrfalt and Shaw 2013). Others, like Indian ricegrass (Achnatherum hymenoides (Roem. & Schult.) Barkworth [Poaceae]), have very long shelf lives, losing little viability over many years (Jones and Nielson 1992). Seed that maintains high levels of viability in long-term storage ameliorates the effects of “boom or bust” seed sales common in the wildfire-driven seed market of the western US. Understanding a species’ viability under storage conditions can assist producers and storage facilities in optimizing stocking rates and limiting waste.

Curlycup gumweed is a native forb under investigation for use in various western rangeland applications including sage-grouse and pollinator habitat enhancement, vegetative fuel breaks (Tilley and Wolf 2020), oil and gas reclamation, and general restoration (Tilley and Pickett 2019). It often functions as an early seral or ruderal species, which may offer greater ability to compete with weeds and increase site resilience to invasion when used in the context of succession management (Tilley and others 2020). Curlycup gumweed has also been shown to have potential as a source of biofuel (Neupane and others 2016). This species seemingly displays many characteristics beneficial to agronomic production, namely upright growth, limited seed shatter, and high seed yield potential. To date it has been collected only in limited quantities, however, and larger-scale production methods have yet to be explored.

Gumweed is described in the taxonomic literature as a biennial to short-lived perennial that forms a basal rosette in the first growing season and produces flowers and fruit in its second year of growth (Harrington 1954; Welsh and others 2003). This species rarely produces seed of significant amounts during the first (rosette) year and is unlikely to persist for a third season. It will, in effect, produce only one seed crop within its 2-y life span before a seed producer will have to plant a new field. In comparison, annual seed crops produce seed in the same year planted, while perennial crops (most restoration grasses and forbs) are longer-lived and produce crops for several years before needing to be re-planted. Biennials have a distinct disadvantage for producers in that they will produce seed in only half of the growing years, so maximizing harvest yields and efficiency is critical.

Additional research is needed to improve establishment of native forbs in seed production fields and on rangeland sites. Research is also needed to evaluate production methods to increase seed availability and to reduce the cost to the end user. We established a glasshouse experiment to evaluate curlycup gumweed seedling emergence from various depths and soil compositions. We also conducted a field trial comparing 3 seed-harvesting methods. Finally, in a third experiment we tracked seed viability of curlycup gumweed seedlots for 3 y to observe any decreases in viability under storage conditions.

MATERIALS AND METHODS

Seeding Depth and Soil Type

The purpose of the first experiment was to evaluate seedling emergence from 3 seeding depths in soils of various compositions. Soils of 3 texture classes were collected from sites in southern Idaho including a Quincy fine sand (S), Declo loam (L), and a Fingal clay loam (CL). Soil was sieved to remove any large aggregates, rocks, or plant matter, and the sieved soil was placed into 50 × 25 × 5 cm (20 × 10 × 2 in) plastic trays. Seed of curlycup gumweed from 3 native populations in Idaho, Nevada, and Utah was threshed and cleaned at the USDA NRCS Plant Materials Center at Aberdeen, Idaho (IDPMC). These sites represented 3 Level III Ecoregions (Omernik 1987)—possible surrogates for seed transfer zones (Miller and others 2011; Bower and others 2014) (Table 1; Figure 1). Seeds were tested for viability at the Idaho State Seed Laboratory (ISSL) using tetrazolium (TZ) staining on 200 seeds from each accession prior to experimentation. All 3 accessions were blended in equal amounts and then pre-treated by soaking the seed in an oxygenated water bath for 24 h following the protocol outlined in Tilley and Pickett (2021) to overcome seed dormancy and initiate germination. We placed seeds in a fine mesh bag and submerged them into a 0.95 l (0.25 gal) Mason jar filled with 500 ml (17 oz) distilled water. Air was added using a Profile 1500 aquarium air pump fitted with a 2.5 cm (1 in) bubbling air stone. After soaking the seed, we immediately planted seed in each of the soils and buried them to a depth of 0, 1 or 2 cm (0, 0.4, or 0.8 in). Each treatment consisted of 36 seeds planted in 3, 15 cm (6 in) rows at approximately 1 cm (0.4 in) spacing. We set up the experiment in a randomized complete block design with 6 replications in a glasshouse in Aberdeen, Idaho. Trays were watered every other day for 20 min. We evaluated total seedling emergence after 5 wk. Seedlings were counted as “emerged” if the cotyledon, seedcoat, or hypocotyl were visible.

View this table:
TABLE 1

Seed source information for 3 seed accessions used in the seedling emergence experiment.

Figure 1.
Figure 1.

Locations of 3 seed collection sites for seeding depth study (red marker) and collection site of seed used for the seed harvest study (yellow marker). Level III Ecoregions of the collection sites are highlighted (US EPA 2013). Satellite imagery courtesy of Google Earth.

Seed Harvest

In the second experiment we evaluated 3 commonly used seed-harvesting techniques on a seed production field of curlycup gumweed. Wildland-collected seed of accession 9106677 originating from the Gardiner Basin of Yellowstone National Park was established in a field at IDPMC during fall 2020. The production field was a single, 146 m (480 ft) long, linear strip of 2 m (6 ft) wide, 116 g (4.1 oz) polypropylene weed-barrier landscape fabric. Weed fabric is used to reduce the abundance of weeds that can interfere with production and reduce seed quality. It can also be effective at reducing soil moisture losses due to evaporation and result in better growth and production (Simonson and others 2006). We burned 10 cm (2 in) holes into the center of the fabric every 46 cm (18 in) and planted 5 to 10 seeds at a depth of 0 to 5 mm (0–2 in) in each hole. The experiment was a randomized complete block design with 4 replications. We divided the fabric strip into 12.2 m (39.6 ft) long plots, each containing approximately 26 plants. For yield analysis, the plots were considered to be 1 m wide × 12.2 m long (12.2 m2) (3.3 ft × 39.6 ft [131 ft2]) based on commonly used 1 m (36 in) row spacing (Cornforth and others 2001; Simonson and Tilley 2016). The field received 5 cm (2 in) of irrigation on 7 May and again on 11 June during the 2020 growing season. No supplemental water was applied during 2021. Plants in our production field were visibly larger than those encountered at the collection site. At the time of harvest, our plants were on average 61 to 69 cm (24–27 in) tall and approximately 120 cm (48 in) wide.

We evaluated 3 harvesting methods: 1) swathing; 2) racquet-and-hopper; and 3) seed stripping with the use of a Flail-Vac. Swathing and combining is a preferred seed-harvesting method, as it allows seed to ripen and cure on the cut stems and reduces seed loss from shattering that may occur when harvesting ripe materials. Racquet-and-hopper harvesting by hand is commonly used for wildland seed collection and is especially useful for gathering seed of indeterminate species. Although effective at small-scale harvesting, it requires high labor inputs (Davison 2003; Jensen 2004; St John and others 2010). Flail-Vac harvesters (Woodward Flail-Vac Seed Stripper, Ag-Renewal Inc, Weatherford, Oklahoma) are commonly used in native forb and grass seed production (Keys 2006; Simonson and Tilley 2016; Tilley and others 2018). Flail-Vacs can be useful tools for harvesting forb seed but can be highly destructive, as the nylon brushes in the header whip the plants at high speeds to remove the seed. This aggressive whipping results in significant damage to mature and unripe flowers, limiting the number of harvests per season of indeterminate flowering species (Simonson and Tilley 2016).

We simulated swathing using a Black & Decker 40 cm (16 in) electric hedge trimmer (Hunt Valley, Maryland) powered by a portable generator (Figure 2). Stems were cut approximately 10 cm (4 in) below the flowers, and the cut material was placed by hand on tarps to dry (Figure 3). To allow for post-harvest ripening, clipping for the swathing treatment was conducted on 12 September 2021, when the first flowers were just beginning to shatter.

Figure 2.
Figure 2.

An electric hedge trimmer was used to simulate a swathing operation. Stems of curlycup gumweed were cut when the first flowers were just beginning to shatter to allow for post-harvest ripening. Photo by Nathaniel Tilley

Figure 3.
Figure 3.

Cut material from the swathing simulation was placed on tarps to dry in lieu of windrowing. Photo by Derek Tilley

The racquet-and-hopper harvest was accomplished using a badminton racquet and a homemade 1 m (39 in) diameter hoop with nylon catch hopper. Flower heads were beaten to capture as much seed as possible as flowers ripened indeterminately (Figure 4). Seed that was ripe and ready to dehisce readily shattered off into the hopper, while unripe seed remained on the seed heads (Figure 5). To maximize seed yields, we conducted racquet-and-hopper harvests on 17, 22, 26, and 30 September 2021.

Figure 4.
Figure 4.

Seed collection using a badminton racquet and a canvas hopper is common for gathering wildland forb seed. Four racquet-and-hopper harvests were conducted in our trial as seed heads ripened to maximize yields. Photo by Mary Wolf

Figure 5.
Figure 5.

Beating with a racquet effectively removes ripe seed while leaving the unripened flowers intact for additional harvests. Photo by Derek Tilley

Flail-Vac harvesting occurred on 22 September and again on 6 October 2021. Our apparatus was run with a brush speed of 400 RPM and a tractor speed of 1.9 km/h (1.2 mph). The Flail-Vac head was placed as deep as possible into the flower canopy, sitting approximately 10 to 20 cm (4–8 in) from the ground (Figure 6). We repeatedly adjusted the angle of the Flail-Vac head to maximize the number of flowers making contact with the brush. Despite our efforts, we found it difficult to get the rigid plant stems into contact with the brush, as they were pushed forward by the front of the brush shroud but did not bounce back into the brush. We were also unable to harvest seed on stems that had grown horizontally or those that had fallen due to lodging, the bending of the stem under the weight of the aboveground growth.

Figure 6.
Figure 6.

A Woodward Flail-Vac was used twice to harvest seed. We experienced difficulties in getting the flower heads into the opening to the brush as the plants were pushed over by the front of the apparatus despite adjusting the intake angle. Plant lodging further complicated Flail-Vac harvesting. Photo by Derek Tilley

All harvested material collected by each treatment method was dried on a clean shop floor for 5 to 7 d and then cleaned identically with 2 cycles through a Westrup air screen cleaner (Slagelse, Denmark) with a 2.1 mm top screen and a 1.15 mm bottom screen. We applied light air to remove inert material and empty seed. Because the harvesting treatments were conducted at different stages of maturation and ripeness, and because each method may favor collection of seeds of a certain size or weight, seed samples from each treatment were sent to ISSL for TZ and purity analysis. Random 0.23 kg (0.5 lb) samples from each plot were sent to ISSL and tested for viability as described for the previous experiment. Seed harvests were additionally analyzed for purity using 3 g subsamples.

Viability

In our third experiment, we tested seed viability under specific storage conditions. Curlycup gumweed seed collected in 2018 was held in storage at IDPMC at low temperatures (10–13 °C [50–55.4 °F]) and low relative humidity (10–50%). We analyzed seed viability as a completely randomized design with 5 replications of randomly selected samples sent to ISSL. Seed was tested in 2018 (yr-0) and again in 2019 (yr-1) and 2021 (yr-3).

Experimental Analysis

Data for all 3 experiments were tested for normality using a Shapiro-Wilk test to determine the appropriate test to analyze variance. Analysis of variance for the seedling emergence experiment was assessed using the Factorial AOV procedure of Statistix 10 Analytical Software (Tallahassee, Florida) with depth and soil type selected as factors. Seed harvest yield and purity data were determined to be normally distributed and were subjected to an analysis of variance using a one-way AOV followed by means separation using the Tukey HSD test (P < 0.05). Seed viability data for the harvest experiment and the viability experiment were not normally distributed and were thus analyzed using a Friedman nonparametric two-way AOV. We selected viability as the dependent variable, and replication and treatment were categorical variables.

RESULTS AND DISCUSSION

Seeding Depth and Soil Type

Laboratory testing prior to the seeding depth experiment indicated viability of >96% of all seedlots used. The Factorial AOV indicated no significant effect (P = 0.4783) from soil texture, but we did see a significant effect (P < 0.0001) from soil depth on seedling emergence (Figure 7). We also detected no significant interaction (P = 0.1647) of soil texture and depth on emergence (Table 2).

Figure 7.
Figure 7.

Seeding depth of emergence was evaluated in 3 soil textures (fine sand, loam, and clay loam) at 3 depths (0, 1, and 2 cm [0, 0.4, 0.8 in]). Image taken 4 wk after planting. Photo by Derek Tilley

View this table:
TABLE 2

F and P values testing seedling emergence from 3 depths in 3 soil textures.

In the S soil, the 0 cm (0 in) planted seed had 42% emergence, significantly higher (P = 0.0013) than emergence from the other 2 seeding depths (Figure 8). Percent emergence of seed planted at 1 and 2 cm (0.4 and 0.8 in) in S soil (24 and 26%, respectively) did not differ significantly. The highest percent emergence in the L soil (47%) was likewise achieved by the 0 cm (0 in) planted seed. This was not significantly higher than the percent emergence of the 1 cm (0.4 in) planted seed, which had a percent emergence of 32%. However, percent emergence of seed planted 2 cm (0.8 in) deep in the soil (17%) was significantly lower than the percent emergence at 0 cm (0 in). In the CL soil, we again observed the highest emergence (38%) in the seed planted at 0 cm (0 in), followed by the 1 and 2 cm (0.4 and 0.8 in) planted seed with 26 and 22% emergence, respectively.

Figure 8.
Figure 8.

Percent emergence of curlycup gumweed seed after 5 wk planted at 0, 1, and 2 cm (0, 0.4, 0.8 in) depth in 3 soil textures: sand (S), loam (L), and clay loam (CL). Error bars are ±1 standard error. Different letters indicate significant differences in mean percent emergence at P < 0.05.

We saw no significant differences (P = 0.3399) among average percent emergence based on soil texture of 0 cm (0 in) planted seed. Emergence ranged from 38% in the CL soil to 47% in the L soil. At the deeper soil depths, no differences in emergence occurred based on soil texture. Based on these results, we can recommend planting curlycup gumweed via broadcast seeding or very shallow drill seeding near the soil surface. If curlycup gumweed is included in a seeding mix that is drill seeded at greater depths to accommodate other species, lower rates of gumweed establishment can be expected. Our data further suggest that curlycup gumweed can be successfully seeded on a wide range of soil types with no difference in expected early establishment.

Maximum emergence after 5 wk in our trial was 47%, which is much lower than the percent viable seed indicated by laboratory TZ testing. However, these results are similar to findings reported by Tilley and Pickett (2021) and Luna (2008) and suggest that seed dormancy is not fully overcome by soaking in oxygenated water or by 60-d cold-moist stratification. Better germination rates may be achievable in field settings using fall-dormant seedings to achieve natural stratification with numerous temperature and moisture fluctuations.

Seed Harvest Methods

Seed yields from the swathing and the racquet-and-hopper harvest methods were significantly greater (P = 0.0498) than yields obtained using a Flail-Vac harvester. Swathing and racquet-and-hopper harvests averaged yields of 1525 and 1509 kg/ha (1360 and 1346 lb/ac), respectively. These 2 harvest methods collected approximately 6 times more seed than the Flail-Vac harvesting, which yielded an average of 233 kg/ha (207 lb/ac) (Figure 9).

Figure 9.
Figure 9.

Extrapolated seed yield (kg/ha) of curlycup gumweed grown on weed barrier fabric harvested with hedge clippers (swath), racquet-and-hopper, and Flail-Vac. Plot size was 1 × 12.2 m (3.3 × 39.6 ft). Different colors indicate different harvest dates. Different letters indicate significant differences at P < 0.05. Error bars represent ± 1 standard error. (For lb/ac multiply by 0.89.)

We used hedge trimmers to simulate a swathing operation in our experiment. Swathing and wind-rowing curlycup gumweed prior to ripening and allowing the seeds to ripen on the cut stems before combining may be feasible on a farm scale. This method would allow the seed to fully ripen while in windrows and reduce losses from seed shatter. The vegetative and floral material was notably sticky at the time of swathing, while much of the plant was still green. Drying in windrows (on tarps in our experiment) reduced the stickiness significantly, and stickiness was not an issue during seed processing.

The Flail-Vac harvester was ineffective at collecting seed of curlycup gumweed. The brush machine tended to wrap any green and flexible plant material around the brush, yanking some of the plants entirely out of the ground. Most of the seed heads, however, were pushed away from the brush, even with the Flail-Vac lowered down into the canopy. We attempted to improve seed pickup by using chains as demonstrated by Simonson and Tilley (2016) to help push the outer stems upward into the path of the Flail-Vac, but that was also unsuccessful. A modification to the Flail-Vac that elevates the outward-leaning stems toward the center, similar to corn lifters, might yield better results.

Stem lodging (Figure 10) was possibly exacerbated by increased soil moisture under the weed barrier fabric, where increased moisture resulted in above normal biomass. Significant lodging reduced uptake by the Flail-Vac, but even when plants remained upright, the Flail-Vac performed poorly. Lodging did not have as great an effect on the swathing and racquet-and-hopper harvest methods. Swathing was done early before stems had weakened from drying, while the hopper could easily be slid under the fallen stems when capturing seed.

Figure 10.
Figure 10.

Stem lodging was present in many of the harvested plants and was a significant problem for the Flail-Vac harvest. Photo by Derek Tilley

Seed yields will vary widely depending on row spacing and plant density. Our yield data are based on extrapolations of 12.2 m2 (131 ft2) plots, with a single row of plants spaced every 46 cm (18 in). Intermountain forb seed production spacing practices vary widely from 0.6 m (24 in) (Stevens and others 1996; Simonson and Tilley 2016) to 2 m (6 ft) spacing of individual fabric rows. Though not directly comparable, our yields for curlycup gumweed far exceed typical yields for commonly grown native forbs such as western yarrow and firecracker penstemon, which average 170 to 225 kg/ha (150–200 lb/ac), respectively (Cornforth and others 2001).

The use of weed barrier fabric likely affected our seed yields. We saw increased soil moisture retention directly below the weed barrier fabric. This moisture was likely the cause of increased plant size and stem lodging compared to that observed among parent populations, and it possibly resulted in an increase in the number of stems and flowers per plant. Therefore, some addition of water, even for such a drought-tolerant species, may increase seed yields under agronomic conditions, as has shown to be the case with other Intermountain forbs (Shock and others 2015).

The cleaning protocols described in the Materials and Methods section resulted in excellent viability and high levels of seed purity from all 3 harvesting methods (Figure 11). Seed viability measured using TZ staining ranged from 95.3 to 97.3% and did not differ significantly (P = 0.2177) between harvesting treatments when cleaned in the same manner. Despite that our swathing treatment was conducted a minimum of 5 d before other harvesting treatments, while stems were green and seed had not yet shattered, post-harvest ripening allowed seed to fully mature with no discernable reduction in viability. Seed purity ranged from 94.0 to 96.0% with no statistically significant (P = 0.4454) differences apparent between harvesting methods. Impurities in the seedlots were characterized as broken seed, chaff, plant debris, and soil.

Figure 11.
Figure 11.

Percent viability and purity of processed curlycup gumweed seed harvested via swathing, racquet-and-hopper, or Flail-Vac. No significant differences were detected between means. Error bars represent ± 1 standard error. Note: y axis ranges from 70 to 100%.

We did not track or compare labor hours per kg of seed harvested in our study, but the differences in labor for each method are significant. Racquet-and-hopper harvesting is very effective for small operations; however, it requires multiple harvests for maximum yield. It is also the most labor-intensive harvest method tested here, as it is entirely done by hand. Flail-Vac harvesting is fully mechanized and is thus considerably faster over larger areas, but it is less effective at optimizing seed yields. Swathing is likewise mechanized but requires 2 operations, namely swathing followed by combining of the dried rows. Based on our results, swathing and combining would appear to be the most efficient means of harvesting curlycup gumweed seed under agronomic conditions.

Viability

We compared laboratory TZ results of curlycup gumweed seed over 3 y in cool-dry storage conditions. We saw no discernable changes (P = 0.2303) in seed viability among testing dates. Viability at the year of harvest (2018), after 1 y of storage (2019), and after 3 y of storage (2021), averaged 97.6%, 97.6%, and 98.0%, respectively. Long-term viability of curlycup gumweed seed could relate to its significant seed dormancy and its functioning as an early seral species. Seed of this species can remain in the seedbank for extended periods of time waiting for disturbance before germination. Although these tests do not offer any long-term viability data for curlycup gumweed, short-term storage viability trends are promising. If these trends continue, curlycup gumweed seed may retain acceptable levels of viability for many years under proper storage conditions.

CONCLUSION

Forb seed is in demand by land managers and the public in the Intermountain West for habitat improvement and biodiversification plantings. However, limitations in seed availability and low establishment rates have prevented broadscale use of this species group (Shaw and others 2005). Reliable production methods and optimum seed placement data are critical to increasing the successful use of forb seed. Curlycup gumweed, a native biennial forb, is promoted for rangeland restoration seedings (Tilley and Pickett 2019) and is being investigated as a biofuel source (Neupane and others 2016). Understanding optimum seeding and harvesting practices is key to production and therefore adoption in these industries. Our glasshouse experiments indicate that curlycup gumweed should be seeded at surface or very shallow depths for maximum seedling emergence on all soil types. Field-harvesting evaluations show that greatest seed yields can be obtained from multiple racquet-and-hopper harvests or with a single swathing operation followed by combining. The Flail-Vac harvester, however, was ineffective at seed harvesting, mostly because of difficulties in getting the seed heads into the brush area. Short-term seed viability evaluations indicate that curlycup gumweed maintains high levels of viability for at least 3 y under cool-dry storage conditions. The high seed yields we recorded, and the apparent longevity of stored seed, suggest that curlycup gumweed could be a marketable species for native seed producers.

Footnotes

  • This article was prepared by a U.S. government employee as part of the employee’s official duties and is in the public domain in the United States.

REFERENCES

  1. ↵
    1. Bartow A.
    2014. Propagation protocol for early blue violet (Viola adunca Sm. [Violaceae]). Native Plants Journal 15(2):125–128.
    OpenUrl
  2. ↵
    1. Bower AD,
    2. St Clair JB,
    3. Erickson V.
    2014. Generalized provisional seed zones for native plants. Ecological Applications 24(5):913–919.
    OpenUrlPubMed
  3. ↵
    1. Bushman BS,
    2. Johnson DA,
    3. Connors KJ,
    4. Jones TA.
    2015. Germination and seedling emergence of three semiarid western North American legumes. Rangeland Ecology and Management 68(6):501–506.
    OpenUrl
  4. ↵
    1. Cornforth B,
    2. St John L,
    3. Ogle D.
    2001. Seed production standards for conservation plants in the Intermountain West. Technical Note No. 14. Aberdeen (ID): USDA Natural Resources Conservation Service. 13 p.
  5. ↵
    1. Davison J.
    2003. A field guide for collecting native seeds in Nevada. University of Nevada, Reno Cooperative Extension.
  6. ↵
    1. de Queiroz T,
    2. Swim S,
    3. Turner PL,
    4. Leger EA.
    2021. Creating a Great Basin native annual forb seed increase program: lessons learned. Native Plants Journal 22(1):90–102. doi:10.3368/npj.22.1.90
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Dumroese RK,
    2. Luna T,
    3. Pinto JR,
    4. Landis TD.
    2016. Forbs: foundation for restoration of monarch butterflies, other pollinators, and greater sage-grouse in the western United States. Natural Areas Journal 36(4):499–511.
    OpenUrl
  8. ↵
    1. Harrington HD.
    1954. Manual of the plants of Colorado. Denver (CO): Sage Books. 666 p.
  9. ↵
    1. Jensen S.
    2004. Racquets, hoppers, and felt boards: low-tech devices for processing seeds. Native Plants Journal 5(1):50–51.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Jones TA,
    2. Nielson DC.
    1992. Germination of prechilled mechanically scarified and unscarified Indian ricegrass seed. Rangeland Ecology and Management 45:175–179.
    OpenUrl
  11. ↵
    1. Karrfalt RP,
    2. Shaw N.
    2013. Banking Wyoming big sagebrush seeds. Native Plants Journal 14(1):60–70.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Keys G.
    2006. Modifications improve seed harvest with the Woodward Flail-Vac Seed Stripper. Native Plants Journal 7(2):149–150.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Kitchen SG,
    2. Monsen SB.
    2001. Forage kochia seed germination response to storage time and temperature. Journal of Range Management 54(3):299–306.
    OpenUrl
  14. ↵
    1. Leger EA,
    2. Goergen EM,
    3. Forbis de Queiroz T.
    2014. Can native annual forbs reduce Bromus tectorum biomass and indirectly facilitate establishment of a native perennial grass? Journal of Arid Environments 102:9–16. doi:10.1016/j.jaridenv.2013.10.015
    OpenUrlCrossRef
  15. ↵
    1. Luna T.
    2008. Propagation protocol for production of container (plug) Grindelia squarrosa (Pursh) Dunal. plants, 160 ml container; USDI NPS – Glacier National Park, West Glacier (MT). In: Native Plant Network. URL: http://NativePlantNetwork.org [currently, http://npn.rngr.net/propagation] (accessed 5 Dec 2017). US Department of Agriculture, Forest Service, National Center for Reforestation, Nurseries, and Genetic Resources.
  16. ↵
    1. McLendon T,
    2. Redente EF.
    1992. Effects of nitrogen limitation on species replacement dynamics during early secondary succession on a semiarid sagebrush site. Oecologia 91(3):312–317. doi:10.1007/BF00317618
    OpenUrlCrossRef
  17. ↵
    1. Miller SA,
    2. Bartow A,
    3. Gisler M,
    4. Ward K,
    5. Young AS,
    6. Kaye TN.
    2011. Can an ecoregion serve as a seed transfer zones? Evidence from a common garden study with five native species. Restoration Ecology 19(201):268–276.
    OpenUrlCrossRef
  18. ↵
    1. Neupane BP,
    2. Shintani D,
    3. Lin H,
    4. Coronella CJ,
    5. Miller GC.
    2016. Grindelia squarrosa: a potential arid lands biofuel plant. Sustainable Chemistry and Engineering 5:995–1001.
    OpenUrl
  19. ↵
    1. Omernik JM.
    1987. Ecoregions of the conterminous United States. Annals of the Association of American Geographers 77:118–125.
    OpenUrlCrossRef
  20. ↵
    1. Pokorny M,
    2. Sheley RL,
    3. Zabinski CA,
    4. Engel R,
    5. Svejcar TJ.
    2005. Plant functional group diversity as a mechanism for invasion resistance. Restoration Ecology 13(3):448–459. doi:10.1111/j.1526-100X.2005.00056.x
    OpenUrlCrossRef
  21. ↵
    1. Rawlins JK,
    2. Anderson VJ,
    3. Johnson R,
    4. Krebs T.
    2009. Optimal seeding depth of five forb species from the Great Basin. Native Plants Journal 10(1):32–42.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Richards RT,
    2. Chambers JC,
    3. Ross C.
    1998. Use of native plants on federal lands: policy and practice. Journal of Range Management 51:625–632.
    OpenUrlCrossRef
  23. ↵
    1. Sanderson MA,
    2. Elwinger GF.
    2004. Emergence and seedling structure of temperature grasses at different planting depths. Journal of Agronomy 96:685–691.
    OpenUrl
  24. ↵
    1. Shaw NL,
    2. Lambert SM,
    3. DeBolt AM,
    4. Pellant M.
    2005. Increasing native forb seed supplies for the Great Basin. In: Dumroese RK, Riley LE, Landis TD, editors and technical coordinators. National proceedings: Forest and Conservation Nursery Associations. 2004: Jul 12–15; Charleston, NC; and 2004: Jul 26–29; Medford, OR. Proceedings RMRS-P-35. Fort Collins (CO): USDA, Forest Service, Rocky Mountain Research Station. p 94–102.
  25. ↵
    1. Shock CC,
    2. Feibert EBG,
    3. Shaw N,
    4. Shock M,
    5. Saunders LD.
    2015. Irrigation to enhance native seed production for Great Basin restoration. Natural Areas Journal 35:74–82.
    OpenUrl
  26. ↵
    1. Simonson DB,
    2. Cornforth B,
    3. Ogle D,
    4. St John L.
    2006. Function and operation of a machine to lay weed barrier fabric. Technical Note No. 25. Aberdeen (ID): USDA Natural Resources Conservation Service. 8 p.
  27. ↵
    1. Simonson DB,
    2. Tilley DJ.
    2016. A low-cost modification to a Flail-Vac harvester for collecting lightweight, wind-dispersed seed. Native Plants Journal 17(2):103–108.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. St John L,
    2. Ogle DG,
    3. Scianna J,
    4. Winslow S,
    5. Tilley D,
    6. Hoag JC.
    2010. Plant materials collection guide. Technical Note No. 1. Boise (ID): USDA Natural Resources Conservation Service. Boise, ID. 17 p.
  29. ↵
    1. Stevens R,
    2. Jorgensen KR,
    3. Young SA,
    4. Monsen SB.
    1996. Forb and shrub seed production guide for Utah. Utah State University Extension. AG 501. 52 p.
    OpenUrl
  30. ↵
    1. Tilley D,
    2. Pickett T.
    2019. Curlycup gumweed initial evaluation planting, 2016–2018. Progress Report. Aberdeen (ID): USDA Natural Resources Conservation Service, Aberdeen Plant Materials Center. 15 p.
  31. ↵
    1. Tilley D,
    2. Pickett T.
    2021. Germination response of curlycup gumweed to oxygenated water treatment. Native Plants Journal 22(1):4–12.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Tilley D,
    2. Wolf M.
    2020. Curlycup gumweed (Grindelia squarrosa (Pursh) Dunal [Asteraceae]): a native forb candidate for inclusion in Great Basin greenstrips. Native Plants Journal 21(2):138–149. doi:10.3368/npj.21.2.138
    OpenUrlAbstract/FREE Full Text
    1. Tilley D,
    2. Taliga C,
    3. Bair C.
    2018. Case study of in situ harvesting in restored grasslands to increase native seed availability. Native Plants Journal 19(2):109–116.
    OpenUrlAbstract/FREE Full Text
    1. Tilley D,
    2. Tilley N,
    3. Fund A,
    4. Wolf M.
    2020. Seedling growth and competition of a late-seral, native perennial grass and 2 early-seral, native forbs in the presence of 2 densities of the invasive annual grass Bromus tectorum L. (Poaceae). Native Plants Journal 21(3):299–311. doi:10.3368/npj.21.3.299
    OpenUrlAbstract/FREE Full Text
    1. [USDA NRCS] USDA Natural Resources Conservation Service
    . 2018. The PLANTS database. URL: http://plants.usda.gov (accessed 9 Oct 2021). Greensboro (NC): National Plant Data Team.
    1. [USDA NRCS] USDA Natural Resources Conservation Service
    . 2021. Web soil survey. URL: http://websoilsurvey.sc.egov.usda.gov/ (accessed 10 Oct 2021).
  37. ↵
    1. [US EPA] U.S. Environmental Protection Agency
    . 2013. Level III ecoregions of the continental United States: Corvallis, Oregon, U.S. EPA – National Health and Environmental Effects Research Laboratory, map scale 1:7,500,000. URL: https://www.epa.gov/eco-research/level-iii-and-iv-ecoregions-continental-united-states
  38. ↵
    1. Uselman SM,
    2. Snyder KA,
    3. Leger EA,
    4. Duke SE.
    2015. Emergence and early survival of early versus late seral species in Great Basin restoration in two different soil types. Applied Vegetation Science 18(4):624–636. doi:10.1111/avsc.12175
    OpenUrlCrossRef
  39. ↵
    1. Vanderlinden D.
    2019. How to determine the proper depth to plant seeds. The Spruce. URL: https://www.thespruce.com/how-deeply-should-seeds-be-planted-2539711 (accessed 10 Oct 2021). New York (NY).
  40. ↵
    1. Welsh SL,
    2. Atwood ND,
    3. Goodrich S,
    4. Higgins LC.
    2003. A Utah flora. 3rd ed., revised. Provo (UT): Brigham Young University. 912 p.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Seedling emergence and seed production of curlycup gumweed
Derek Tilley, Mary Wolf, Derek Jolley, Gordon Hirning
Native Plants Journal Sep 2022, 23 (3) 299-308; DOI: 10.3368/npj.23.3.299
Twitter logo Facebook logo Mendeley logo
Bookmark this article

Jump to section

Keywords