Seed yield response of Lomatium dissectum to water applied in 2009–15.
Seed yield response of Lomatium dissectum to water applied plus spring precipitation in 2009–15.
Relative seed yield response of Lomatium dissectum to water applied plus spring precipitation in 2009–15 and the 7-year average relative seed yield response.
Seed yield response of Lomatium grayi to water applied in 2007–15.
Seed yield response of Lomatium grayi to water applied plus fall, winter, and spring precipitation in 2007–15.
Relative seed yield response of Lomatium grayi to water applied plus fall, winter, and spring precipitation in 2007–15 and the 9-year average relative seed yield response.
Seed yield response of Lomatium triternatum to water applied in 2007–15.
Seed yield response of Lomatium triternatum to water applied plus spring precipitation in 2007–15.
Relative seed yield response of Lomatium triternatum to water applied plus spring precipitation in 2007–15 and the 9-year average relative seed yield response.
Click on author name to view affiliation information
Seeds of native plants are needed for rangeland restoration in the Intermountain West. Many of these plants are rarely cultivated and relatively little is known about the cultural practices required for their seed production. Irrigation trials were conducted for five perennial Lomatium species over multiple years. Lomatium species grown at the Oregon State University Malheur Experiment Station, Ontario, OR received 0, 100, or 200 mm of irrigation per year. Seed yield responses to irrigation were evaluated by linear and quadratic regression. In general, seed yields from the three species grown for 10 years responded linearly or quadratically to irrigation. To improve the accuracy of estimated irrigation water requirements, regressions were also run on seed yield responses to irrigation plus precipitation during the previous spring; spring and winter; and spring, winter, and fall. Over multiple years, Lomatium dissectum (Nutt.) Mathias & Constance and L. triternatum (Pursh) J.M. Coult. & Rose seed yields were best estimated by a quadratic response to irrigation plus spring precipitation with highest yields at 243 and 255 mm, respectively. Lomatium grayi (J.M. Coult. & Rose) J.M. Coult. & Rose seed yields were best estimated by a quadratic response to irrigation plus precipitation during the fall, winter, and spring with highest yields at 358 mm. Two of the Lomatium species were grown for the last 6 years. The seed yields of L. nudicaule (Pursh) J.M. Coult. & Rose did not respond to irrigation. Seed yields of Lomatium suksdorfii (S. Watson) J.M. Coult. & Rose responded linearly to irrigation in 2015.
Seeds of native plants are needed for rangeland restoration in the Intermountain West due to increasing rates of disturbance caused by non-native plant invasions and more frequent wildfires (Balch et al., 2013; Liu and Wimberly, 2015). Postfire restoration can stabilize soils and reduce re-establishment of invasive annuals, although restoration effectiveness depends on techniques and seed mixes (Robichaud et al., 2000). A recent synthesis showed that restoration of forbs (nonwoody perennials) is critical for rehabilitating habitat for pollinators and wildlife, such as the greater sage grouse (Dumroese et al., 2015). A number of promising native forb species have been identified for rangeland restoration purposes, but relatively little is known about the cultural practices required for seed production of these plants (Shaw and Jensen, 2014).
Several members of the genus Lomatium, including L. dissectum (fernleaf biscuitroot), L grayi (Gray’s biscuitroot), L. nudicaule (barestem biscuitroot), and L. triternatum (nineleaf biscuitroot), are recognized as important for use in wildland restoration in the Intermountain West (Dumroese et al., 2015). Lomatium suksdorfii is of interest since it is both rare and a potential source of pharmaceutical compounds. These species are native to the western United States and are long-lived perennials with yellow, purple, or green/brown flowers. The shoot develops from the crown of a large taproot in early spring using the natural moisture from snow melt and spring rain. Vegetative growth, flowering, and seed set are complete by late spring to early summer. The leaves dieback after seed set, usually in midsummer. Lomatium species are dormant from late summer through winter and do not resume growth with fall rains.
Many Lomatium species were used by Native Americans for their food and medicinal properties (Shock et al., 2012a, 2012b; Tilley et al., 2010a, 2010b), and evaluation and use of such species as L. suksdorfii (Suksdorf’s desertparsley) continue today (Lee et al., 1994). Except for L. suksdorfii, which is found only in south-central Washington and north-central Oregon, these species are widespread in the Intermountain West. Lomatium species grow in scattered to occasionally dense stands on well-drained coarse- to fine-textured soils and on scablands, rocky slopes, drainages, and open ridges.
Large-statured and widely distributed Lomatiums are valued as restoration species because they are among the earliest wildflowers to green up in late winter or spring. They provide a source of forage for wildlife (elk, deer, and antelope) and domestic livestock (sheep and cattle) (Ogle and Brazee, 2009). Barnett and Crawford (1994) reported that sage-grouse hens ate Lomatium leaves and stems, while Drut et al. (1994) found that sage-grouse chicks also consumed Lomatium. The Lomatium species were not identified in either study. Lomatiums are also used by birds, small mammals, pollinators, and other insects (Thompson, 1998). Lomatium species compete with non-native annuals such as Bromus tectorum (cheatgrass) because of their early spring growth and deep taproot system (Parkinson et al., 2013; Tilley et al., 2010a, 2010b).
Commercial seed production of Lomatium is needed for use in rangeland restoration plantings in the Intermountain West, but these plants are rarely cultivated, and cultural practices for seed production are largely unknown. A major limitation to economically viable commercial production of Lomatium seed is achieving stable and consistent seed production from year to year. In native rangelands, natural variation in spring rainfall and soil moisture results in highly unpredictable water stress at flowering, seed set, and seed development, which for many seed crops is known to compromise seed yield and quality. The seed yield response of L. dissectum to irrigation has been reported previously without considering the effect of precipitation (Shock et al., 2012b, 2015).
Native forbs are not well adapted to row crop production practices. Supplemental water can be provided by sprinkler or furrow irrigation systems, but these irrigation systems risk encouraging weeds and fungal pathogens. Subsurface drip irrigation can reduce wetting of the soil surface and could reduce weed and disease pressure.
The trials reported here tested the effects of three low rates of subsurface drip irrigation on the seed yield of five Lomatium species. The optimum amount of irrigation for each species was evaluated based not only on the amount of irrigation but also on the amount of seasonal precipitation each year.
An irrigation trial for L. dissectum, L. grayi, and L. triternatum was initiated in 2005 at the Malheur Experiment Station of Oregon State University, Ontario, OR. Ontario is centered at 43°58′ 42.2″ N, 117°1′ 29.8″ at 689 m elevation. Annual precipitation averages 257 mm. The field, a Nyssa silt loam (coarse-silty, mixed, mesic, Xeric Haplodurid), was bedded into 76-cm rows, four rows per bed. Most of the field lacked topsoil due to land leveling in the 1950s. The analysis of a soil sample taken on 22 Nov. 2005 indicated a pH of 8.3, 1.09% organic matter, 12 ppm phosphorus (P), 438 ppm potassium, 27 ppm SO4-sulfur, 4370 ppm calcium, 456 ppm magnesium, 81 ppm sodium, 1.6 ppm zinc (Zn), 0.6 ppm copper, 4 ppm manganese, 3 ppm iron (Fe), and 0.6 ppm boron.
Drip tape (T-Tape TSX 515-16-340; T Systems International Inc., San Diego, CA) was buried at 30-cm depth and spaced 1.52 m apart beneath alternating inter-row spaces. The flow rate for the drip tape was 4.16 L/min/100 m at 55 kPa with emitters spaced 41 cm apart, resulting in a water application rate of 1.7 mm·h−1. Water was filtered through sand media filters. Application durations were controlled automatically.
Seed of L. dissectum, L. grayi, and L. triternatum was planted on 26 Oct. 2005 in four rows 133 m long spaced 76 cm apart using a custom-made small-plot grain drill with disk openers. Seed was planted at 12.5-mm depth with 65–100 seeds/m of row. Seed of each species came from wild collections made by U.S. Forest Service employees (Table 1).
On 25 Nov. 2009, seed of L. nudicaule and L. suksdorfii was planted in the same field as the other Lomatium species in four rows spaced 76 cm apart using the small-plot grain drill. All seed was planted on the soil surface with 65–100 seeds/m of row. After planting, sawdust was applied in a narrow band over the seed row at 26 g·m−1 of row, and the beds were covered with rowcover. Each rowcover (N-sulate Deluxe Plus 1 oz; DeWitt Co., Inc., Sikeston, MO) covered two rows and was applied with a mechanical plastic mulch layer. The field was irrigated for 24 h on 2 Dec. 2009 due to very dry soil conditions.
In April following fall planting, each Lomatium strip was divided into twelve 9.1-m-long plots. Each plot contained four rows spaced 76 cm apart. The experimental design was a randomized complete block with four replicates.
The three irrigation treatments were 0 (control), 100, and 200 mm/year. The 100-mm and 200-mm irrigation treatments received four irrigations, 2 weeks apart, starting at the beginning of flowering. Each irrigation applied 25 (100-mm treatment) or 50 mm (200-mm treatment) of water. In 2007, irrigation treatments were inadvertently continued after the fourth irrigation following seed production, doubling the planned irrigation amounts for each treatment. Flowering dates for each species were recorded and are reported in conjunction with the irrigation dates in Table 2.
Fertilization of the irrigation trial over the 6 years was minimal. On 11 Nov. 2006, nitrogen (N) at 112 kg·ha−1 as urea was broadcast. The following nutrients were applied as soluble nutrients through the drip irrigation: 27 Oct. 2006: P at 56 kg·ha−1 and Zn at 2.2 kg·ha−1; 9 Apr. 2009: N at 56 kg·ha−1 and P at 11 kg·ha−1; 3 May 2011: N at 56 kg·ha−1; 13 Apr. 2012: N at 56 kg·ha−1, P at 11 kg·ha−1, and Fe at 0.012 kg·ha−1; 29 Mar. 2013: N at 22 kg·ha−1, P at 28 kg·ha−1, and Fe at 0.012 kg·ha−1; 2 Apr. 2014: N at 22 kg·ha−1, P at 28 kg·ha−1, and Fe at 0.012 kg·ha−1; 15 Apr. 2015: N at 22 kg·ha−1, P at 28 kg·ha−1, and Fe at 0.016 kg·ha−1. Nitrogen was applied as a urea–ammonium nitrate solution, P was applied as phosphoric acid, and Fe was applied as iron chelate.
During the first 2 years (2005 and 2006), weeds were controlled primarily with cultivation and hand rouging. Herbicides were screened for their effectiveness and plant tolerance in other trials (Shock et al., 2011). These products are not yet registered for commercial use on Lomatium species. Prowl® (pendimethalin; BASF, Research Triangle Park, NC) at 1.1 kg·ha−1 a.i. was broadcast on the soil surface for weed control on 17 Nov. 2006, 9 Nov. 2007, 15 Apr. 2008, 18 Mar. 2009, 4 Dec. 2009, 17 Nov. 2010, 9 Nov. 2011, 7 Nov. 2012, 26 Feb. 2014, and 13 Mar. 2015. Volunteer® (clethodim; Tenkoz, Inc., Alpharetta, GA) was broadcast at 0.07 kg·ha−1 a.i. on 18 Mar. 2009, 3 Apr. 2013, and 26 Feb. 2014. Hand rouging of weeds continued as necessary.
Seed yield was determined by a manual once-over harvest of all mature seed stalks in the middle 7.5 m of the two center rows of the four-row plots. On a few occasions, the maturation and harvest of the nonirrigated plants preceded the harvest of the irrigated plants (Table 2). Seed was cleaned from stalks and chaff and was weighed. Seed yield means were compared by analysis of variance and linear and quadratic regression. Seed yield for each species each year was regressed separately against 1) applied water; 2) applied water plus spring precipitation; 3) applied water plus winter and spring precipitation; and 4) applied water plus fall, winter, and spring precipitation. Winter and spring precipitation occurred in the same year that yield was determined; fall precipitation occurred the prior year.
Seed yield (y, kg·ha−1) in response to irrigation or irrigation plus precipitation (x, mm/season) was estimated by the equation y = a + b·x + c·x2. For the quadratic equations, the amount of irrigation (xʹ) that resulted in maximum yield (yʹ) was calculated using the formula xʹ = −b/2c, where a is the intercept, b is the linear parameter, and c is the quadratic parameter. For the linear regressions, the highest seed yield for a species in a given year was based was on the highest measured average seed yield.
Adding the seasonal precipitation to the irrigation response equation would have the potential to provide a closer estimate of the amount of water required for maximum seed yields for each species. Regressions of seed yield for each species each year were calculated on all the sequential seasonal amounts of precipitation and irrigation, but only some of the regressions are reported below. For each species, the period of precipitation plus applied water that had the lowest standard deviation for irrigation plus precipitation over the years was chosen as the most reliable independent variable for predicting seed yield. To compare yield responses over years, regressions were also made on the relative seed yields compared with irrigation plus precipitation. Relative seed yield for each plot was calculated as the percentage of the yield of the highest yielding treatment for each species for each year.
Lomatium dissectum began flowering and producing seed in 2009, the 4th year after fall planting in 2005 (Tables 2 and 3). Flowering and seed production began for L. grayi and L. triternatum in the 2nd year (2007), L. nudicaule in the 3rd year (2012), and L. suksdorfii in the 4th year (2013) after fall planting (Tables 2 and 4–6). Although there were tremendous variations in seed yields over species and years, general tendencies emerged for the mean seed yields for each species and irrigation treatments (Table 7).
Seed yields of L. dissectum exhibited a quadratic response to irrigation amount in 2009–13 (Table 3; Figs. 1–3). The quadratic response to irrigation in 2012 was not statistically significant. Seed yield of L. dissectum exhibited linear responses to irrigation rate in 2014 and 2015. The amount of water applied for maximum seed yield, calculated from the regression equations, ranged from 126 to 242 mm per season. Seed yields were estimated to be highest with 200 to 282 mm of total applied water plus spring precipitation. The highest yields averaged 1097 kg·ha−1 and ranged from 521 to 1573 kg·ha−1. Averaged over 7 years, relative seed yield was highest with 243 mm of applied water plus spring precipitation (Fig. 3).
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Seed yields of L. grayi exhibited quadratic responses to irrigation rate in 2008, 2010, 2011, 2012, 2014, and 2015; linear responses in 2007, 2009, and 2013; and no statistically significant responses in 2010 and 2011 (Table 4; Figs. 4–6). The amount of water applied for the highest yield, calculated from the regression equations, ranged from 51 to 200 mm per season. Seed yields of L. grayi were estimated to be highest with 335 to 425 mm of total applied water plus fall, winter, and spring precipitation, depending on the year. These results imply that if a year were very dry, more than 200 mm of irrigation would be required to maximize seed yields. The highest yields averaged 950 kg·ha−1 and ranged from 148 to 1654 kg·ha−1, depending on the year. Averaged over 9 years, relative seed yield was highest with 358 mm of total applied water plus fall, winter, and spring precipitation.
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Seed yields of L. triternatum exhibited quadratic responses to irrigation rate in 2008–12 (Table 5; Figs. 7–9). Seed yields in 2007, 2013, 2014, and 2015 showed linear responses to irrigation. Seed yields were estimated to be highest using 102 to 210 mm of applied water depending on the year. Seed yields were estimated to be highest using 222 to 307 mm of applied water plus spring precipitation and ranged from 31 to 3580 kg·ha−1, depending on the year and averaged 1529 kg·ha−1. Averaged over 9 years, relative seed yield was highest with 265 mm of total applied water plus spring precipitation.
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Citation: HortScience 51, 10; 10.21273/HORTSCI11150-16
Over 4 years of production, seed yields of L. nudicaule showed no significant response to water applied (Table 6). Maximum yields averaged 505 kg·ha−1 and ranged from 139 to 789 kg·ha−1, depending on the year.
Seed yields of L. suksdorfii did not respond significantly to water applied in 2014, the first year of seed production (Table 6). Seed yields in 2015 showed a linear response to irrigation. Seed yields were estimated to be maximized by 282 mm of total applied water plus spring precipitation in 2015. Maximum yields averaged 1086 kg·ha−1 and were 205 kg·ha−1 in 2014 and 1950 kg·ha−1 in 2015. The higher seed yield in 2015 may be related to advanced plant maturity and greater stature than in 2014.
Seasonal precipitation varied greatly among years, as did the amount of water required for maximum seed production. Adding seasonal precipitation to applied water reduced variation in the optimal water requirement for maximum seed production (Table 8). The regression analyses showed high variability in the optimum amount of water for maximum seed yield among years. When precipitation was added to the amount of water applied in the regressions, the variability in the optimum amount of water decreased.
On average, L. dissectum started flowering on about 9 Apr., and L. triternatum started flowering around 18 Apr. (Table 2). For these two species, the addition of spring precipitation to applied irrigation water resulted in the lowest standard deviation of the amount of water applied for maximum yield according to the quadratic equations for 2007–15 (Table 9). For L. grayi, the addition of fall, winter, and spring precipitation to applied irrigation water resulted in the lowest standard deviation of the amount of water applied for maximum yield according to the regression equations for 2007–15 (Table 9; Figs. 4–6). Lomatium grayi usually started flowering in early to mid-March (Table 2), possibly explaining why it might have responded more to fall, winter, and spring precipitation than L. dissectum and L. triternatum.
Other studies that examined responses of xerophytic plants to irrigation found similar responses within the ranges of irrigation in our study. In Tucson, AZ, with annual precipitation of 293 mm, native grass and native woody plant establishment from seed was optimum with 187 to 210 mm of irrigation plus precipitation (Roundy et al., 2001). In the Chihuahuan Desert in New Mexico, with annual precipitation of 211 mm, only 6 of 15 species increased vegetative growth in response to 338 mm of annual irrigation (Gutierrez and Whitford, 1987).
There are few published studies examining seed production of native plants in response to irrigation in arid regions. In a west-central Texas, area that receives 530 mm of annual precipitation, Petersen and Ueckert (2005) found that seed production of Atriplex canescens (fourwing saltbush) did not respond to either 400 mm of irrigation in one year or 200 mm of irrigation the next year. In the Owens Valley, CA, with annual precipitation of 113 mm, Sarcobatus vermiculatus (greasewood) seed yields were significantly higher when irrigated (Breen and Richards, 2008). These studies were not designed to determine optimum amounts of irrigation for seed yield.
In this study, all Lomatium species tested, except L. nudicaule, responded to irrigation with increased seed yields. For the three species tested for 7–9 years (L. dissectum, L. grayi, and L. triternatum), seed yields were increased significantly by irrigation. Seed yields of L. suksdorfii also increased by irrigation in 2015. The amount of irrigation and precipitation needed for maximum Lomatium seed yields in this study (200–400 mm/year) was substantially lower than irrigation plus precipitation requirements for row crops in the Treasure Valley of Oregon (500–800 mm/year, AgriMet, 2016). Although Lomatium species are perennials, they complete their cycle from sprouting to seed harvest by mid-June, thus their irrigation requirements would be expected to be lower. However, irrigation requirements for Lomatium are low even when compared with an early season crop such as winter wheat (Triticum aestivum), which has an average water requirement of 633 mm in the Treasure Valley.
Studies of cultural practices for native plants are critical to increasing seed production and availability of plant material used in restoration of ecosystems in the Intermountain West. As restoration needs increase, the economic opportunities for native plant seed growers will also increase. Developing in-depth knowledge on seed production practices will support the native seed industry in the future.
The seed yields of four of the five Lomatium species (L. dissectum, L grayi, L. triternatum, and L. suksdorfii) responded significantly to subsurface drip irrigation. The irrigation requirements for seed production of L. dissectum and L. triternatum were best estimated by a quadratic response to irrigation plus spring precipitation. The irrigation requirements for seed production of L. grayi were best estimated by a quadratic response to irrigation plus precipitation during the fall, winter, and spring. Additional trials are needed for L. suksdorfii to better define its seed yield response to precipitation plus applied water.
Seed yield response of Lomatium dissectum to water applied in 2009–15.
Seed yield response of Lomatium dissectum to water applied plus spring precipitation in 2009–15.
Relative seed yield response of Lomatium dissectum to water applied plus spring precipitation in 2009–15 and the 7-year average relative seed yield response.
Seed yield response of Lomatium grayi to water applied in 2007–15.
Seed yield response of Lomatium grayi to water applied plus fall, winter, and spring precipitation in 2007–15.
Relative seed yield response of Lomatium grayi to water applied plus fall, winter, and spring precipitation in 2007–15 and the 9-year average relative seed yield response.
Seed yield response of Lomatium triternatum to water applied in 2007–15.
Seed yield response of Lomatium triternatum to water applied plus spring precipitation in 2007–15.
Relative seed yield response of Lomatium triternatum to water applied plus spring precipitation in 2007–15 and the 9-year average relative seed yield response.
Contributor Notes
This project was partially funded by the USDA Forest Service, Rocky Mountain Research Station’s Great Basin Native Plant Project, USDI Bureau of Land Management, Oregon State University, Malheur County Education Service District, and by formula grants 2015-31100-06041 and 2015-31200-06041 from the USDA National Institute of Food and Agriculture.
Professor and Director.
Senior Faculty Research Assistant.
Bioscience Research Technician I.
Bioscience Research Technician III.
Research Botanist (Emeritus).
Research Biologist.
Corresponding author. E-mail: clinton.shock@oregonstate.edu.
Seed yield response of Lomatium dissectum to water applied in 2009–15.
Seed yield response of Lomatium dissectum to water applied plus spring precipitation in 2009–15.
Relative seed yield response of Lomatium dissectum to water applied plus spring precipitation in 2009–15 and the 7-year average relative seed yield response.
Seed yield response of Lomatium grayi to water applied in 2007–15.
Seed yield response of Lomatium grayi to water applied plus fall, winter, and spring precipitation in 2007–15.
Relative seed yield response of Lomatium grayi to water applied plus fall, winter, and spring precipitation in 2007–15 and the 9-year average relative seed yield response.
Seed yield response of Lomatium triternatum to water applied in 2007–15.
Seed yield response of Lomatium triternatum to water applied plus spring precipitation in 2007–15.
Relative seed yield response of Lomatium triternatum to water applied plus spring precipitation in 2007–15 and the 9-year average relative seed yield response.
