EditorialResearch Article
Open Access

Soil Seed Banks in Stock-piled Topsoils in the Western Rio Grande Plains, Texas

Mylea C. Lovell, Sandra Rideout-Hanzak, David E. Ruppert, Veronica Acosta-Martinez, Forrest S. Smith, Paula Maywald Stumberg, Keith A. Pawelek, Anthony D. Falk and David B. Wester
Ecological Restoration, September 2018, 36 (3) 226-237; DOI: https://doi.org/10.3368/er.36.3.226

Abstract

Topsoils often are removed from energy production sites and stock-piled for use later in restoration activities. Effects of this practice on soil seed banks are unknown. We examined seed bank size, species richness, and species composition of stock-piled topsoils as affected by sampling depth and sampling date at two study sites in the western Rio Grande Plains, TX, USA. Stock-piled topsoil and adjacent non-disturbed topsoil samples were collected at 0–10, 10–20, 20–30 and 30–40 cm depths on five dates over an 18-month period following stock-pile construction. Seed banks were assessed with the seedling emergence method. Sampling date had little effect on seed bank characteristics. We detected differences among depths on the stock-pile, and between stock-piles and undisturbed soil. Seed bank size and species richness generally decreased with increased stock-pile sampling depth at both sites. Differences between stock-piles and undisturbed soil varied between sites: at one site, stock-piling effects were common and were expressed in lower seed bank size and richness in stock-piles compared to undisturbed soils; at the other site, stock-piling had fewer effects on richness or seed bank size. Prevalence of exotic species varied between sites and likely reflected differences in surrounding vegetation. Therefore, site-to-site variability precludes strong generalizations. However, density of emerged native seedlings ranged from < 1 to 3.8 seedlings m−2 at both sites. Assuming acceptable species composition, stock-piles supported an adequate seed bank size at time of sampling for restoration without need for additional seed input.

Restoration Recap

  • Topsoil from areas disturbed by oil and gas extraction often is stock-piled for future restoration use.

  • As a reservoir of native seeds, stock-piled topsoil is a valuable resource that can influence future restoration success. Important seed bank properties include seed bank size, species composition, and species richness.

  • We show that seed bank size and species richness generally decrease with increasing depth in soil stock-piles; differences between stock-piles and nearby undisturbed soil, however, were site-specific and likely reflect differences in surrounding vegetation.

  • Factors affecting seed bank size were more closely related to the season when the stock-pile was constructed and then sampled, as well as immediately-preceding climatic conditions, than the age (i.e., residence time) of the stock-pile per se.

Stock-piling topsoils from areas disturbed by oil and gas extraction is a common practice for future site restoration. Topsoils are the major reservoir of seed banks, which reflect past and present vegetation and can influence future community dynamics and structure (Iverson and Wali 1982, Coffin and Lauenroth 1989, Yan et al. 2010, Gioria and Pysek 2016). Seed banks also affect the sustainability of a plant species population (Zhang and Chu 2013) and have important implications for invasive species (Gioria et al. 2014, Gioria and Pysek 2016).

Soil seed banks can be valuable for ecological restoration (e.g., Henderson et al. 1988, Bakker et al. 1996, Vécrin et al. 2007, Bossuyt and Honnay 2008, Gioria and Osborne 2009, Klaus et al. 2017) when they contain viable seeds of desired species—species which may or may not occur in the current aboveground plant community as well as species that have dispersed from surrounding areas. This is particularly true of persistent seed banks because they can be an effective means to “bridge temporally-unsuitable habitat conditions” (Klaus et al. 2017, 2) that result from disturbance associated with energy extraction. Additionally, because replacement of plant species via reseeding may be expensive or impossible (Augusto et al. 2001, Falk et al. 2017), effective use of soil seed banks can be economically attractive.

Seed bank species composition, however, often is different from the aboveground composition (Thompson and Grime 1979, Johnson and Anderson 1986, Dickie et al. 1988, Kinucan and Smeins 1992, Rosef 2008, Erenler et al. 2010, Pekas and Schupp 2013, Gioria and Pysek 2016). Additionally, soil seed banks can be affected by management (Auestad et al. 2013). The value of the soil seed bank, therefore, depends on overall restoration goals (Milberg 1995, Gioria and Obsorne 2009) as well as past management of the vegetation that produced the seed bank. For example, if composition of a seed bank approximates the desired composition of future established vegetation, then adding seed to the seed bank may be unnecessary. When the seed bank is dominated by undesirable species, adding seed may be needed to achieve long-term restoration goals. Seed ecological and plant reproductive components, soil and climatic conditions, and seed predation and decomposition contribute to the dynamics of soil seed bank characteristics (Iverson and Wali 1982). Although seed banks may not result in immediate revegetation of desired late-successional species (Chambers and MacMahon 1994, D’Souza and Barnes 2008), in the case of drastic disturbance, protection of the soil from wind and water erosion, extreme temperatures and water loss, and spills associated with petroleum drilling and production may be more important than immediate restoration of climax plant communities.

Recognizing the potential value of stock-piled topsoils in restoration processes, our goal was to characterize seed banks of stock-piled topsoils at two study sites in the western Rio Grande Plains of Texas, USA. We hypothesized that both sampling depth and sampling date would be important factors affecting seed bank characteristics— sampling depth because the combined effects of seed rain and fluctuating environmental conditions would impact surface depths of the stock-pile more than deeper locations in the stock-pile; and sampling date because of potential residence-time effects as a stock-pile aged as well as seasonal effects on seed rain. Thus, we hypothesized that, over time, surface-soil seed banks would be more variable than deeper-soil seed banks, an effect that would be detected as a sampling depth-by-sampling date interaction.

Additionally, because different layers of a stock-pile can be segregated during restoration activities, it is critical to understand differences in seed bank characteristics as function of depth in a stock-pile. Therefore, two specific objectives of this research were to: 1) evaluate emergent seed bank richness, size (number of seedlings), and species composition on stock-piles as affected by sampling depth and sampling date, and 2) compare seed banks of stock-piles to seed banks of undisturbed soils to assess the viability of stock-piling topsoil for future site restoration.

Methods

Study Areas

Study sites were located on two private (Hixon and San Ysidro) ranches in two counties in the western Rio Grande Plains of southern Texas. These ranches were selected because soil stock-piles were less than 45 days old. The source of the stock-pile material was the top 20 cm of intact soil at each study site (i.e., the A horizon of Dilley soil series, Hixon study site; and the A horizon of the Brundage soil series, San Ysidro study site). The Hixon Ranch study site is an oil and gas drilling pad located near the center of La Salle County near Cotulla, Texas (annual precipitation and temperature averages 53 to 66 cm and 21 to 22°C, respectively [NOAA 2015]). The stock-pile was approximately 15 m long, 6 m wide and 3 m tall. The stock-pile was formed from a site characterized by the Dilley soil series (loamy, mixed, active, hyperthermic, shallow Ustic Haplargids [NRCS 2015]). Midgrass-dominated savannah would have been the historic climax community for this area of La Salle County, characterized by Trichloris crinita (false rhodegrass), Bouteloua curtipendula (sideoats grama), T. pluriflora (multiflowered false rhodegrass), Digitaria californica (Arizona cottontop), and Pappophorum bicolor (pink pappusgrass). Prosopisglandulosa (mesquite), Aloysia gratissima (whitebrush), Condalia sp. (condalia), and Lycium carolinianum (wolfberry), along with diverse forbs are found throughout this area (NRCS 2015). Nomenclature follows Hatch et al. (1990).

The San Ysidro Ranch study site is adjacent to a fracking pond (a temporary, plastic-lined pond used to store water for hydraulic fracturing) in the south-central area of Dimmit County near Catarina, Texas (annual precipitation and temperature averages 43 to 61 cm and 22 to 24°C, respectively [NOAA 2015]). The stock-pile is approximately 38 m long, 12 m wide and 5 m tall. The Brundage soil series (fine-loamy, mixed, active, hyperthermic Aridic Natrustalfs) is the dominant soil series of this site. Historic conditions of this area in Dimmit County would have been a savannah landscape. Midgrasses would have dominated with some shrub species interspersed throughout. Grasses that would be expected include T. crinita, Heteropogon contortus (tanglehead), Setaria macrostachya (plains bristlegrass), Bothriochloa laguroides (silver blue-stem), D. californica, Chloris cucullata (hooded windmill grass), P. bicolor, Tridens eragrostoides (lovegrass tridens), and Hilaria belangeri (curly mesquite). Scattered among the grasses would have been various forb and woody species. For example, Simsia calva (bush sunflower), Wedelia texana (orange zexmenia), Ambrosiapsilostachya (western ragweed), Acacia berlandieri (guajillo), A. rigidula (blackbrush), and Celtis ehrenbergiana (spiny hackberry) would have been present (NRCS 2015).

Sampling Procedures

Six 400-g (~656 cm3) soil samples (Ball and Miller 1989, Eldridge and Lunt 2010) were collected with a bucket auger at random locations on each topsoil stock-pile on five sampling occasions (17 June (Hixon) and 26 June (San Ysidro) 2013; and on both ranches 20 August 2013; 9 April 2014; 19 May 2014; and 28 July 2014). Soil auger samples were separated into 0-10 cm, 10-20 cm, 20-30 cm, and 30-40 cm depths and placed in marked Nasco® bags. (It was not possible to collect deeper samples because of soil compaction). Six samples of undisturbed soil from the A horizon at each study site (within 50 m of the stock-pile and on the same soil series) also were collected on each sampling date. Each sample from the undisturbed topsoil was mixed into a composite sample, as this better represented the stripped and mixed topsoil in the stock-pile (Mason et al. 2011). Soil samples were transported from study sites to the Texas A&M University-Kingsville, Kingsville, TX, soils laboratory in coolers with ice.

Processing and Greenhouse Procedures

Soil samples were screened in the laboratory within 14 d of collection with a 5.6 mm sieve to smooth soil aggregates and remove rocks and large pieces of vegetative debris. Samples were then spread out on covered tables to air dry at room temperature. Dried soil samples were transported in their same marked bags to a Texas A&M University-Kingsville greenhouse that was maintained at ~30°C during the day and ~20°C at night; there was no artificial lighting.

The emergence method (Espeland et al. 2010) was used to assess seed banks. Each soil sample was evenly spread over 9 cm of clean, coarse sand in a 29.5 cm x 44.5 cm plastic trays with drainage; the soil sample was about 0.5 cm deep across the top of the sand (Gross 1990). The relatively thin layer of soil should have facilitated germination of most viable seeds because only seed near the top layer of soil will typically germinate, especially in fine soils (Ter Heerdt et al. 1996, Traba et al. 2004). Results are expressed in terms of species richness and seed bank size (number of emerged seedlings) m-2.

Six negative control trays and six positive control trays also were prepared. Negative control trays had only 9 cm of sand and no soil (Dougall and Dodd 1997, D’Souza and Barnes 2008, Eldridge and Lunt 2010). Positive control trays had 9 cm of sand and were inoculated with 20 seeds each of Leptochloa dubia (green sprangletop), S. calva, and a mixture of Plantago rhodosperma (redseed plantain) and P. hookeriana (Hooker’s plantain). Positive control trays were implemented to indicate that environmental conditions were appropriate for optimal germination of some common south Texas grass and forb species. Negative control trays were used to detect contamination from the sand that was placed in each tray or from inside the greenhouse.

All trays were positioned on five benches in the greenhouse in completely random fashion. Trays were checked daily for seedling emergence and watered as needed to maintain field capacity. Trays were fertilized after three weeks with Miracle-Gro® Liquafeed® Bloom Booster® liquid fertilizer (12% nitrogen, 9% phosphorus and 6% potash [Kinucan and Smeins 1992]).

Seedlings were identified as soon as possible after emergence. After a plant was identified and recorded, it was removed with a pair of scissors or tweezers at its base to minimize disturbing nearby seeds and seedlings (Thompson and Grime 1979). Each germination trial lasted six weeks, after which unidentified plants were transplanted into pots and observed until positive identification. Each trial took five to eight months to complete.

Statistical Analysis

The soil seed bank of each greenhouse tray was characterized by the germinable seed bank size (total numbers of emerged and identified seedlings) and species composition based on seedling density. Species richness was calculated for each tray. Neither seed bank size nor richness are normally distributed (Fritsch and Hsu 1999, Rogers and Hsu 2001). Therefore, permutation-based analyses (Anderson et al. 2008) were used to test hypotheses addressing effects of sampling date and sampling depth, as well as comparisons between stock-piles and nearby undisturbed soil on seed bank size and richness with Euclidean distance as a resemblance measure. Two analyses were performed to examine: 1) changes over sampling dates and sampling depths on the stock-piles; and 2) comparison between the stock-piles and undisturbed soil. The first analysis used a linear mixed model with sample auger as a random nuisance (block) effect and sampling depth as an effect of interest in a randomized block design. Data from the five sampling dates were analyzed together; in this model, auger sample was nested in date, and the interaction between auger sample and depth was nested with sampling date. In the second analysis, seed bank characteristics were compared between undisturbed soil and stock-piles, with each soil depth analyzed separately in a comparison with undisturbed soil. Therefore, this analysis tested effects of location (undisturbed soil or stock-pile) and sampling date, as well as their interaction using auger sample nested within date and location as an error term. These two analyses were also performed on mean species composition with a permutation-based multivariate analysis of variance based on an unweighted Bray-Curtis similarity measure (Anderson 2001, Anderson et al. 2008). For this analysis, trays which failed to produce seedlings were removed from the data set prior to analysis. When interactions were significant, simple main effects were tested, followed by simple effects tests as needed; otherwise, main effects were analyzed (Kirk 2013). Reported p values are based on permutation; we used Permanova+ software, v. 6 (Primer-e, Auckland New Zealand [Clarke and Gorley 2006]).

Past seed bank studies (Thompson and Grime 1979, Reichman 1984, Henderson et al. 1988, Coffin and Lauenroth 1989, Shaukat and Siddiqui 2004, Gioria and Osbourne 2009) have shown high variability in the distribution of seeds in the soil; thus, a = 0.10 was used to detect statistical differences. Given high spatial variability in soil characteristics between study sites and the confounding effects of different machine operators, it is unlikely that two stock-piles could reasonably be considered as two replications. Therefore, in this research, each stock-pile is defined as the population of inference (Wester 1992), and soil samples from randomly-selected locations on each stock-pile are samples from these populations.

Results

Results from negative and positive controls indicated that greenhouse conditions were suitable for germination and that sand used in trays was not a source of seed contamination.

Hixon Study Site—Changes on the Stock-pile

Sampling depth and sampling date interacted in their effects on species richness (F12,75 = 1.71, p = 0.0842) and seed bank size (F12,75 = 1.68, p = 0.089; Table 1). Both species richness and seed bank size were highest in August 2013 and lower and generally similar at other sampling dates, a pattern that was strongest at shallower depths and weakest at 30-40 cm. Richness decreased with increasing sampling depth in August 2013 (F3,15 = 3.17, p = 0.0589) and May 2014 (F3,15 = 2.70, p = 0.0856) but was unaffected (p > 0.588) by sampling depth at other sampling dates. Seed bank size decreased (F3,15 = 4.07, p = 0.0274) with increasing sampling depth only during May 2014.

View this table:
Table 1.

Mean (SE) seed bank species richness and seed bank size (m–2) in the stock-pile at the Hixon study area at four sampling depths and five sampling dates. 1Date: F4,25 = 36.63, p = 0.0001. Depth: F3,75 = 3.47, p = 0.0196. Interaction: F12,75 =1.71, p = 0.0842. 2Sampling date means within a depth followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD test). 3Sampling depth means at a sampling date followed by the same uppercase letter are not significantly different (p > 0.10, protected LSD test). 4Date: F4,25 = 16.98, p = 0.0001. Depth: F3,75 = 3.30, p = 0.0232. Interaction: F12,75 = 1.68, p = 0.089.

Sampling depth and sampling date also interacted (F12,29 = 1.29, p = 0.0985) in their effects on mean species composition (Supplementary Tables S1 and S2). Mean species composition differed across sampling dates at 0-10 (F4,16 = 3.52, p = 0.0007) and at 10-20 cm (F4,16 = 3.08, p = 0.0137) depths but was stable (p > 0.4064) across sampling dates at deeper depths. Mean species composition also was affected (F3,3 = 8.29, p = 0.0238) by sampling depth only during the last sampling period.

In addition to seed bank size, richness and species composition, dynamics of native and non-native species on stock-piles are of interest in restoration projects. Mean number of native seedlings was affected by sampling date (F4,25 = 18.90, p < 0.0001) and depth (F3,75 = 2.44, p = 0.0694); these effects, however, acted independently of each other (F12,75 = 1.27, p = 0.2596; Table 2). Native seedling numbers were higher in August 2013 than at other sampling dates; additionally, native seedlings numbers decreased with sampling depth. In contrast, effects of sampling date on exotic seed bank size depended on sampling depth (F12,75 = 1.66, p = 0.0629). In particular, exotic seedling numbers generally were higher in August 2013 than at other sampling dates at all but the deepest sampling depth. Sampling depth had no effect (p > 0.1315) on seed bank size except in May 2014 when they decreased with increasing depth. Native seedling numbers from undisturbed soil were also highest in August 2013 and exotic seedling numbers were unaffected by sampling date.

View this table:
Table 2.

Mean (SE) total number of native1 and exotic2 seedlings (m-2) in the stock-pile at the Hixon study site over four sampling depths and five sampling dates, and in undisturbed soil at five sampling dates. 1Date: F4,25 = 18.90, p = 0.0001. Depth: F3,75 = 2.44, p = 0.0694. Interaction: F12,75 = 1.27, p = 0.2596. 2Date: F4,75 = 5.78, p = 0.0018. Depth: F3,75 = 2.61, p = 0.0581. Interaction: F12,75 = 1.66, p = 0.0629. 3Sampling date means for native species followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD test). 4Sampling depth means for native species followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD test). 5Sampling date means at a depth for exotic species followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD test). 6Sampling depth means at a sampling date for exotic species followed by the same uppercase letter are not significantly different (p > 0.10, protected LSD test). 7Undisturbed date means for an origin of species followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD test).

Hixon Study Site—Comparisons between Stock-pile and Undisturbed Soil

Seed bank species richness differed (p < 0.0001) across sampling dates at all stock-pile sampling depths down to 30 cm (Table 3). Richness was highest in August 2013 at each of these depths. However, richness was similar (p > 0.2422) between undisturbed soil and the stock-pile at all depths and effects were consistent (p > 0.1607) across sampling dates. Sampling date and depth interacted (F4,50 = 5.51, p = 0.009) in their effects on richness only at the 30-40 cm depth, where richness was greater in undisturbed soil than in the stock-pile and also higher in August 2013 than on other sampling dates.

Table 3.
Table 3.

Mean (SE) seed bank species richness and seed bank size (m-2) in the stock-pile and in undisturbed soils at the Hixon study area at four sampling depths and five sampling dates. 1Date: F4,50 = 24.42, p = 0.0001. Location: F1,50 = 0.11, p = 0.7331. Interaction: F4,50 = 0.24, p = 0.9126. 2Date: F4,50 = 25.26, p = 0.0001. Location: F1,50 = 3.42, p = 0.0706. Interaction: F4,50 = 3.15, p = 0.0201. 3Date: F4,50 = 22.08, p = 0.0001. Location: F1,50 = 0.58, p = 0.4514. Interaction: F4,50 = 7.71, p = 0.1607. 4Date: F4,50 = 21.38, p = 0.0001. Location: F1,50 = 3.79, p = 0.0575. Interaction: F4,50 = 2.82, p = 0.0357. 5Date: F4,50 = 29.14, p = 0.0001. Location: F1,50 = 1.38 p = 0.2422. Interaction: F4,50 = 1.08, p = 0.3058. 6Date: F4,50 = 26.92, p = 0.0001. Location: F1,50 = 13.29, p = 0.0010. Interaction: F4,50 = 7.32, p = 0.0002. 7Date: F4,50 = 17.29, p = 0.0001. Location: F1,50 = 7.41, p = 0.0098. Interaction: F4,50 = 5.51, p = 0.0009. 8Date: F4,50 = 19.65, p = 0.0001. Location: F1,50 = 20.34, p = 0.0020. Interaction: F4,50 = 13.76, p = 0.0001. 9Sampling date means followed by the same letter are not significantly different (p > 0.10, protected LSD). 10Sampling date means within a location followed by the same uppercase letter are not significantly different (p > 0.10, protected LSD). 11Location means within a sampling date followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD).

Effects of sampling date and location (stock-pile vs. undisturbed soil) on seed bank size were different from their effects on species richness (Table 3). In particular, both factors interacted (p < 0.0366) at each depth and location effects depended on sampling date. Seed bank size was greater in undisturbed soil than in the stock-pile at 0-10 and 20-30 cm depths in August 2013, and also greater in undisturbed soil than the stock-pile at 10-20 and 30-40 cm depths in May 2014. Additionally, seed bank size was generally larger in August 2013 than at other sampling dates, a pattern that was more pronounced in undisturbed sites than on the stock-pile.

Location (stock-pile vs undisturbed soil) interacted with sampling date in its effect on mean species composition at 0-10 cm (F4,29 = 2.48, p = 0.0031) and 10-20 cm (F4,29 = 1.62, p = 0.0572) sampling depths of the stock-pile (Supplementary Tables S1 and S2). Relative to date effects, mean species composition in undisturbed locations was different on August 13 from all other sampling dates but varied among sampling dates on the stock-pile at these depths. Relative to location effects, mean species composition of the stock-pile surface differed from that of undisturbed soil in August 2013 and July 2014; and composition of the 10-20 cm depth of the stock-pile differed from undisturbed soil in August 2013 and May 2014. Different patterns were observed when deeper sampling depths were compared to undisturbed soil: location effects acted independently of sampling date (p > 0.3412) and although composition varied across date (p < 0.0013) it was similar (p > 0.3165) between locations.

San Ysidro Study Site—Changes on the Stock-pile

Sampling depth and date interacted (F12,75 = 1.96, p = 0.0409) in their effects on species richness (Table 4). No seedlings emerged at the 0-10 cm depth in August 2013 but richness was higher during subsequent sampling dates. At deeper sampling depths, richness was not affected (p > 0.5418) by sampling date. Sampling depth did not affect (F3,15 = 0.74, p = 0.5418) species richness in July 2013; richness, however, decreased (p < 0.0320) with increasing sampling depths during subsequent sampling dates. Sampling date and depth also interacted in their effects on seed bank mean species composition (F6,13 = 1.56, p = 0.0742 (analysis based on August 2013, April 2014 and May 2014 sampling dates). In particular: 1) depth effects were not significant (p < 0.24) at any sampling date; and 2) date effects were strong (F2,14 = 5.95, p < 0.001) at 0-10 cm and non-detectable (p > 0.114) at deeper sampling depths (Supplementary Tables S3 and S4).

Table 4.
Table 4.

Mean (SE) seed bank species richness and seed bank size (m-2) in the stock-pile at the San Ysidro study area at four sampling depths and five sampling dates. 1Date: F4,75 = 1.95, p = 0.1290. Depth: F3,75 = 11.69, p = 0.0001. Interaction: F12,75 = 1.96, p = 0.0409. 2Date: F4,75 = 2.45, p = 0.0781; Depth: F3,75 = 19.65, p = 0.0001. Interaction: F12,75 = 2.58, p = 0.0061. 3Sampling date means at a depth followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD test). 4Sampling depth means within a sampling date followed by the same uppercase letter are not significantly different (p > 0.10, protected LSD test).

Sampling date and depth also interacted (F4,75 = 2.58, p = 0.0061) in their effects on seed bank size (Table 4). Seed bank size was reduced (F4,25 = 3.17, p = 0.0334) in the 0-10 cm depth at the June 2013 sampling date compared to later sampling dates; however, size was similar across sampling dates at all other depths (10-20 cm: F4,25 = 1.79, p = 0.1613; 20-30 cm: F4,25 = 1.29, p = 0.2995; 30-40 cm: F4,25 = 0.60, p = 0.6625). Relative to depth comparisons, at the June 2013 sampling date, seed bank size was similar across depths (F3,15 = 0.74, p = 0.5418); at all other sampling dates, seed bank size decreased as sampling depth increased (August 2013: F3,15 = 5.97, p = 0.0056; April 2014: F3,15 = 3.77, p = 0.0320; May 2014: F3,15 = 6.91, p = 0.0036; July 2014: F3,15 = 9.71, p = 0.0008).

Only five exotic species seedlings emerged from San Ysidro samples, which were too few to analyze. Therefore, dynamics of the native species seedlings are sufficiently described by the total number of seedlings analysis.

San Ysidro Study Site—Comparisons between Stock-pile and Undisturbed Soil

Sampling location and date interacted in their effects on mean species richness at the 0-10 cm (F4,50 = 2.24, p = 0.0672) and 10-20 cm (F4,50 = 2.69, p = 0.0380) sampling depths (Table 5). Richness of the stock-piled soil seed bank at the 0-10 cm depth was lower than the undisturbed soil at the June 2013 sampling date and higher than undisturbed soil in April 2014; species richness was similar between undisturbed soil and the stock-pile at this depth at other sampling dates. Species richness was higher in undisturbed soil than stock-piled soil at the 10-20 cm depth at each sampling date except April 2014.

View this table:
Table 5.

Mean (SE) seed bank species richness and seed bank size (m-2) in the stock-pile and in undisturbed soils at the San Ysidro study area at four sampling depths and five sampling dates. 1Date: F4,50 = 1.61, p = 0.1864. Location: F1,50 = 2.24, p = 0.1372. Interaction: F4,50 = 2.24, p = 0.0672. 2Date: F4,50 = 1.51, p = 0.2154. Location: F1,50 = 0, p = 1.000. Interaction: F4,50 = 1.47, p = 0.2251. 3Date: F4,50 = 0.31, p = 0.8741. Location: F1,50 = 18.15, p = 0.0043. Interaction: F4,50 = 2.69, p = 0.0380. 4Date: F4,50 = 0.39, p = 0.8209. Location: F1,50 = 5.00, p = 0.0318. Interaction: F4,50 = 1.37, p = 0.2575. 5Date F4,50 = 1.41, p = 0.2448. Location: F1,50 = 18.83, p = 0.0001. Interaction: F4,50 = 1.57, p = 0.2025. 6Date: F4,50 = 0.58, p = 0.6776. Location: F1,50 = 14.78, p = 0.0007. Interaction: F4,50 = 0.55, p = 0.7037. 7Date: F4,50 = 1.25, p = 0.2971. Location: F1,50 = 25.30, p = 0.0001. Interaction: F4,50 = 1.75, p = 0.1569. 8Date: F4,50 = 0.49, p = 0.7479. Location: F1,50 = 17.55, p = 0.0001. Interaction: F4,50 = 0.58, p = 0.6801. 9Sampling date means within a location followed by the same lowercase letter are not significantly different (p > 0.10, protected LSD). 10Location means within a sampling date followed by the same uppercase letter are not significantly different (p > 0.10, protected LSD). 11Location means followed by the same letter are not significantly different (p > 0.10, protected LSD). 12 Sampling date means followed by the same letter are not significantly different (p > 0.10, protected LSD).

Sampling date did not affect species richness at the 20-30 cm (F4,50 = 1.41, p = 0.2448) or 30-40 cm (F4,50 = 125, p = 0.2971) depths (Table 5). Species richness was higher in undisturbed soil than the stock-pile at both the 20-30 cm (F1,50 = 18.83, p = 0.0001) and 30-40 cm (F!,50 = 25.30, p = 0.0001) depths.

At the 0-10 cm depth, mean species composition differed across date (F4,37 = 2.86, p = 0.0001) and undisturbed species composition was different (F1,37 = 2.29, p = 0.0268) from the pile. Sampling date and location interacted at the 10-20 cm (F4,37 = 1.49, p = 0.0578) and 30-40 cm (F3,20 = 1.50, p = 0.067) depths. Mean species composition was not affected by either sampling date (F4,24 = 1.23, p = 0.195) or location (F1,24 = 1.45, p = 0.1604), nor did date and location interact (F4,24 = 1.18, p = 0.2195; Supplementary Tables S3 and S4) at the 20-30 cm depth.

Sampling location (stock-pile vs undisturbed) and date acted independently in their effects on seed bank size at each sampling depth. Seed bank size was not affected by sampling date at any depth (0-10 cm: F4,50 = 1.51, p = 0.2139; 10-20 cm: F4,50 = 0.39, p = 0.8278; 20-30 cm: F4,50 = 0.58, p = 0.6959; and 30-40 cm: F4,50 = 0.49, p = 0.77; Table 5). Seed bank size was identical between undisturbed soil and the 0-10 cm depth of the stock-piled topsoil but was smaller at deeper depths in the stock-pile compared to the undisturbed soil (10-20 cm: F1,50 = 4.00, p = 0.0285; 20-30 cm: F 1,50 = 14.78, p = 0.0002; and 30-40 cm: F1,50 = 17.55, p = 0.0001); seed bank size did not differ across sampling dates (F4,21 = 0.51, p = 0.7579) at these depths.

Discussion

Seed bank size, species composition (including presence/absence of invasive species) and richness can affect outcomes of future restoration (e.g., Gioria and Pysek 2016). These properties were compared at two study sites as a function of sampling depth and sampling date after construction of soil stock-piles. In general, species richness and seed bank size decreased with increasing soil depth at both sites; other seed bank features differed between study sites.

Soil seed banks at Hixon study site were, in general, very different from the San Ysidro study site with respect to size, richness and composition. Differences were expected because of differences in micro-climate, soil order, and past and present management. Study sites differed in other ways as well. For example, stock-piles were made by different machine operators on different dates. Without specific instructions, operators use their own methods for removing and piling soil. These confounding factors are typical of on-the-ground management practices.

Changes Over Time

At the Hixon study site, sampling date strongly influenced nearly every aspect of our results. The August sampling date (with the largest seed bank size and highest richness) was different from all other sampling dates. Seed bank composition also changed over sampling date on the stock-pile as well as between stock-pile depths and undisturbed soil. Differences among sampling dates might be attributed in part of environmental changes associated with woody vegetation removal prior to stock-pile construction. Woody vegetation removal affects air movement patterns important to seed dispersing seeds such as Cenchrus ciliaris (buffelgrass [Olsson et al. 2012] ) as well as light, temperature and moisture conditions at the soil surface which in turn can affect dormancy and germination (e.g., Wester 1995, Baskin and Baskin 2001).

Precipitation events prior to the August sampling date also may have contributed to increases in emerged seedlings (Figure 1). One possible explanation begins with precipitation events prior to sampling in June 2013 that resulted in a flush of germination of a large portion of resident seeds in the seed bank resulting in the depletion of the seed bank (Barton 1962). This could explain the relatively low number of emerged seedlings in the June greenhouse trial. In addition, seedlings which resulted from this flush of germination flowered after being triggered by a period of low precipitation (Grime 1977, Kozlowski and Pallardy 2002) in July and August and consequently contributed a significant and immediate input of seeds into the soil seed bank of both undisturbed soil and the stock-pile which was subsequently recovered in the August greenhouse trial.

Figure 1.
Figure 1.

Total precipitation in 2013 (A) and 2014 (B) and mean total number of seedlings (m–2) at four depths in the stock-pile and in undisturbed soil on five sampling dates (plots on top) at the Hixon study site, Texas, USA. Black dots at the specified dates represent cumulative precipitation three weeks prior to sample collection. Precipitation data for Cotulla Airport, Texas, USA (NOAA 2015).

Although sampling date had little effect on species richness at the San Ysidro study site, both seed bank size and species composition were affected by a sampling date-by-depth interaction. This may be a result of dispersing seeds (“seed rain” [Caughlin et al. 2016]) from nearby vegetation which affect the soil surface more immediately than soil depths below the surface (e.g., Dickie et al. 1988). Precipitation events (Figure 2) may have influenced seed bank dynamics in a similar manner, though to lesser extent than at the Hixon study site.

Figure 2.
Figure 2.

Total precipitation in 2013 (A) and 2014 (B) and mean total number of seedlings (m−2) at four depths in the stock-pile and in undisturbed soil on five sampling dates (plots on top) at the San Ysidro study site, Texas, USA. Black dots at the specified dates represent cumulative precipitation three weeks prior to sample collection. Precipitation data for Carrizo Springs Weather Station, Texas, USA (NOAA 2015).

Duration of residence in a stock-pile caused no discernible effects on seed bank characteristics at our study sites after 18 months of storage. Seed banks of these soils either are evidently resilient to environments typical of stock-piles, or alternatively the resident seed bank of this area may consist of mainly transient seeds which may be missed in seedling emergence trials and would be expected to decline in stock-pile conditions (Milberg 1995, D’Souza and Barnes 2008).

Changes by Depth

Species richness decreased with increasing depth at both study sites, indicating death or depletion of species which are transient or more sensitive to environmental conditions. Seed bank size at the Hixon site decreased with increasing depth only on the May 2014 sampling date. At the San Ysidro site, however, size decreased with increasing sampling death on all but the June 2013 sampling date (consistent with Rivera et al. 2013); this suggests that the duration of storage in the stock-pile is nearing an age at which seed viability and density are reduced. Seed rain likely was involved; additionally, upper stock-pile depths would have been more been more quickly exposed to temperature fluctuations, possibly relieving dormancy of seeds near the surface. Therefore, after eight months of storage, the seed bank of the stock-pile at the San Ysidro study site developed a profile by depth similar to what would be expected in an undisturbed soil profile (Inverson and Wali 1982, Rosef 2008).

Mean seedling densities at both sites generally met the industry standard for restoration and exceeded the one plant per m2 standard considered acceptable in arid and semi-arid environments (Cox and Jordan 1983, Dahl et al. 1988). Potential of the seed bank of these stock-piled soils for restoration efforts would be increased if native species dominated the soil seed bank, thereby reducing the need for seeding native species. In this study, both the mean number of exotic seedlings and mean number of native seedlings were affected by sampling date at the Hixon study site. However, depth had no effect on exotics species, whereas native species declined in the deeper stock-piled soil. Changes in mean numbers of exotic and native species over sampling date may be a function of precipitation or other seasonal factors such as temperature. Changes did not appear to be affected by the age of the stock-pile. Decline of native seedling germination with increasing depth suggests that native species may be less able to withstand these conditions than exotic species. At the San Ysidro study site, very few exotic species emerged which hindered analysis of differences in natives and exotics.

Because emerged exotics were virtually absent in the San Ysidro stock-pile seed bank, comparisons of characteristics can be made between mean total number of seedlings at the San Ysidro and the native seed bank of the Hixon study site. There were similarities between native seed banks in the stock-piles of both study sites. Mean numbers of seedlings at the San Ysidro site decreased as depth increased in all but the June 2013 sampling date; this is similar to the decrease in native species with increasing depth in the Hixon stockpile seed bank. And although mean total number of seedlings declined in the Hixon study site stock-pile at the May 2014 sampling date, depth did not have an effect at other sampling dates. It is possible that the resilience of exotic species at deeper stock-pile depths masked the effects of depth on seed bank size, which were revealed when native species were analyzed separately.

Stock-piled Soil vs. Undisturbed Soil

The effect of stockpiling on seed banks can be directly assessed by comparing seed bank properties of stockpiles to those of undisturbed soil collected at the same date; because of potential depth effects in stock-piles, this comparison is most meaningful when a given depth in a stock-pile is compared to undisturbed soils. Stockpiling had little effect on seed bank species richness at the Hixon study site: richness was higher in undisturbed soil than stock-piled soil on only one sampling date and at one depth. When stockpiling effects were detected on seed bank size (20% of the comparisons), undisturbed soils had more seedlings than stock-piled soils. In contrast, at the San Ysidro site stockpiling reduced seed bank richness and size in 80% and 75%, respectively, of the comparisons between stock-piled and undisturbed soil. These observations are compatible with Iverson and Wali (1982) who found that both numbers and diversity of seeds decreased as depth in the stock-pile increased.

Many studies of seed banks in stored topsoil have found that duration and depth in the stock-pile affected seed bank characteristics (Iverson and Wali 1982, Dickie et al. 1988, Rokich et al. 2000, Scoles-Sciulla and DeFalco 2009, Rivera et al. 2012). Studies of mixed grass prairies in North Dakota (Iverson and Wali 1982), woodlands in Australia (Rockich et al. 2000), the Mojave Desert in Nevada (Scoles-Sciulla and DeFalco 2009) and annual dominated grassland in Spain (Rivera et al. 2012) concluded that seedlings decreased as time in the stock-pile increased. A range of six to 13 months was considered sufficient to negatively impact seed bank dynamics. Our results differ from many of these studies. This is not unexpected because seed bank dynamics are specific to the ecology of the specific seed bank being studied.

Our study addressed the readily germinable portion of the soil seed bank that would be available for immediate germination under ideal climatic conditions after the top-soil is redistributed. The emergence method does not detect dormant seeds or seeds which require specific conditions such as cold stratification. It is therefore likely that some seeds were not accounted for in our results. Additionally, we were only able to conduct seed bank trials to examine the piles up to 18 months of storage. It would be beneficial to continue this process and gather data on the characteristics of these same seed banks at older ages of the pile. This is an important finding from our study because it suggests that stock-piles may not be as effective as many practitioners have assumed because they differ in fundamental ways from nearby intact soils.

Management Implications

Averaged over soil depth, seed bank size at the Hixon study site ranged from 4.6 seedlings m-2 in August to less than 1 seedling m-2 at other sampling dates (3.8 native seedlings m-2 in August to less than 1 native seedling m-2 at other sampling dates). At the San Ysidro study site, emergence ranged from 2.3 native seedlings m-2 to less than 1 seedling m-2. These ranges fall within recommended plant establishment guidelines. Given the preponderance of native species in the seed bank (especially at the San Ysidro study site), there would have been little reason to re-seed these stock-piles had they been used for restoration on the dates sampled. It is also true, however, that both seed bank richness and size often were reduced in stock-piles at depth compared to intact soil, especially at San Ysidro, suggesting that whereas the restoration seed bank may have been “adequate”, it was nevertheless “not equivalent” to that of intact soil. We detected no apparent patterns to changes in seed bank dynamics of the stored soil that could be attributed solely to storage time (up to the 18-month sampling period of this study). Future restoration success likely would be improved if stock-piled soil could be segregated in the restoration process during seasons appropriate for adequate precipitation to enhance chances of success.

Acknowledgments

We would like to thank Mr. and Mrs. A. and H. Beinhorn and Mr. and Mrs. T. and K. Hixon for providing access to study sites; Mr. and Mrs. Beinhorn also provided financial support. Funding was provided by the U.S. Department of Interior, Coastal Impact Assistance Program through the Houston Advanced Research Center (Contract No. CITP08-TAMUK0113B); we are particularly grateful for the support of R. Haut and C. LaFleur of HARC. The South Texas Chapter of the Quail Coalition and René Barrientos also provided funding. We would like to thank J. Grace, J. Hoffman, A. McCloughan and A. Toomey for field assistance. We are grateful to E. Grahmann and J. Grace for reviewing this paper and improving it with their constructive criticism. This is Caesar Kleberg Wildlife Research Institute manuscript number 17-123, Texas A&M University-Kingsville, TX. USDA Disclaimer: Trade names and company names are included for the benefit of the reader and do not infer any endorsement or preferential treatment of the product by USDA ARS. USDA-ARS is an equal opportunity provider and employer.

This open access article is distributed under the terms of the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0) and is freely available online at:http://jhr.uwpress.org

References

  1. ↵
    1. Auestad I.,
    2. Rydgren K.,
    3. Spindelböck J.P.
    2013, Management history affects grassland seed bank build-up. Plant Ecology 214:1467–1477.
    OpenUrl
  2. ↵
    1. Anderson M.J.
    2001. A new method for non-parametric multivariate analysis of variance. Austral Ecology 26:32–46.
    OpenUrlCrossRef
  3. ↵
    1. Anderson M.J.
    2008. Gorley R.N., Clarke K. R. 2008. PERMANOVA + for PRIMER: Guide to Software and Statistical Methods. Plymouth, UK: PRIMER-E, Ltd.
  4. ↵
    1. Augusto L.,
    2. Dupouey J-L,
    3. Picard J-F,
    4. Ranger J.
    2001. Potential contribution of the seed bank in coniferous plantations to the restoration of native deciduous forest vegetation. Acta Oecologica 22:87–98.
    OpenUrl
  5. ↵
    1. Bakker J.P.,
    2. Poschlod P.,
    3. Strykstra R.J.,
    4. Bekker R.M.,
    5. Thompson K.
    1996. Seed banks and seed dispersal: important topics in restoration ecology. Acta Botanica Neerlandica 45:461–490.
    OpenUrlCrossRef
  6. ↵
    1. Ball D.A.,
    2. Miller S.D.
    1989. A comparison of techniques for estimation of arable soil seed banks and their relationship to weed flora. Weed Research 29:365–373.
    OpenUrl
  7. ↵
    1. Barton L.V.
    1962. The germination of weed seeds. Weeds 10:174–182.
    OpenUrl
  8. ↵
    1. Baskin C.C.,
    2. Baskin J.M.
    2001. Seeds: Ecology, Biogeography and Evolution of Dormancy and Germination. San Diego, CA, US: Academic Press.
  9. ↵
    1. Bossuyt B.,
    2. Honnay O.
    2008. Can the seed bank be used for ecological restoration? An overview of seed bank characteristics in European communities. Journal of Vegetation Science 19:875–884.
    OpenUrlCrossRef
  10. ↵
    1. Caughlin T.T.,
    2. Elliott S.,
    3. Lichstein J.W.
    2016. When does seed limitation matter for scaling up reforestation from patches to landscapes. Ecological Applications 26:2439–2450.
    OpenUrl
  11. ↵
    1. Chambers J.C.,
    2. MacMahon J.A.
    1994. A day in the life of a seed: Movements and fates of seeds and their implications natural and managed systems. Annual Review of Ecology and Systematics 25:263–292.
    OpenUrlCrossRef
  12. ↵
    1. Clarke K.R.,
    2. Gorley R.N.
    2006. PRIMER v6: User Manual/Tutorial. Plymouth, UK: PRIMER-E.
  13. ↵
    1. Coffin D.P.,
    2. Lauenroth W.K.
    1989. Spatial and temporal variation in the seed bank of a semiarid grassland. American Journal of Botany 76:53–58.
    OpenUrlCrossRef
  14. ↵
    1. Cox J.R.,
    2. Jordan G.L.
    1983. Density and production of seeded range grasses in southeastern Arizona. Journal of Range Management 36:649–652.
    OpenUrl
    1. Dahl B.E.,
    2. Cotter P.F.,
    3. Wester D.B.,
    4. Britton C.M.
    1987. Range plant establishment in the Southern Plains region. Pages 42–46 in Mitchell J.E. (ed). Proceedings of the Impacts of the Conservation Reserve Program in the Great Plains; 16–18 September 1987, Denver, CO, USA.
  16. ↵
    1. Dickie J. B.,
    2. Gajjar K. H.
    1988. The survival of viable seeds in stored topsoil from opencast coal workings and its implications for site restoration. Biological Conservation 43:257–265.
    OpenUrl
  17. ↵
    1. Dougall T.A.G.,
    2. Dodd J.C.
    1997. A study of species richness and diversity in seed banks and its use for the environmental mitigation of a proposed holiday village development in a coniferized woodland in south east England. Biodiversity and Conservation 6:1413–1428.
    OpenUrlCrossRef
  18. ↵
    1. D’Souza L.E.,
    2. Barnes P.W.
    2008. Woody plant effects on soil seed banks in central Texas savanna. The Southwestern Naturalist 53:495–506.
    OpenUrl
  19. ↵
    1. Eldridge D.J.,
    2. Lunt I.D.
    2010. Resilience of soil seed banks to site degradation in intermittently flooded riverine woodlands. Journal of Vegetation Science 21:157–166.
    OpenUrl
  20. ↵
    1. Erenler H.E.,
    2. Ashton P.A.,
    3. Gillman M.P.,
    4. Ollerton J.
    2010. Factors determining species richness of soil seed banks in lowland ancient woodlands. Biodiversity and Conservation 19:1631–1648.
    OpenUrl
  21. ↵
    1. Espeland E.K.,
    2. Perkins L.B.,
    3. Leger E.A.
    2010. Comparison of seed bank estimation techniques using six week species in two soil types. Rangeland Ecology and Management 63:243–247.
    OpenUrl
  22. ↵
    1. Falk A.D.,
    2. Pawelek K.A.,
    3. Smith F.S.,
    4. Cash V.,
    5. Schnupp M.
    2017. Evaluation of locally-adapted native seed sources and impacts of livestock grazing for restoration of historic oil pad sites in South Texas. Ecological Restoration 35:120–126.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Fritsch K.S.,
    2. Hsu J.C.
    1999. Multiple comparison of entropies with application to dinosaur biodiversity. Biometrics 55: 1300–1305.
    OpenUrlPubMed
  24. ↵
    1. Gioria M.,
    2. Osborne B.
    2009. Assessing the impact of plant invasions on soil seed bank communities: use of univariate and multivariate statistical approaches. Journal of Vegetation Science 20:547–556.
    OpenUrlCrossRef
  25. ↵
    1. Gioria M.,
    2. Jarosik V.,
    3. yšek P.
    2014. Impacts of invasions by alien plants on soil seed bank communities: Emerging patterns. Perspectives in Plant Ecology, Evolution and Systematics 16:132–142.
    OpenUrl
  26. ↵
    1. Gioria M.,
    2. Pyšek P.
    2016. The legacy of plant invasions: Changes in the soil seed bank of invaded plant communities. Bioscience 66:40–53.
    OpenUrlCrossRef
  27. ↵
    1. Grime J.P.
    1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111:1169–1194.
    OpenUrlCrossRef
    1. Hatch S.L.,
    2. Gandhi K.N.,
    3. Brown L.E.
    1990. A checklist of the vascular plants of Texas. MP-1655. Texas Agricultural Experiment Station, College Station.
  29. ↵
    1. Henderson C.B.,
    2. Peterson K.E.,
    3. Redak R.A.
    1988. Spatial and temporal patterns in the seed bank and vegetation of a desert grassland community. Journal of Ecology 76:717–728.
    OpenUrl
  30. ↵
    1. Iverson L.R.,
    2. Wali M.K.
    1982. Buried, viable seeds and their relation to revegetation after surface mining. Journal of Range Management 35:648–652.
    OpenUrl
  31. ↵
    1. Johnson R.G.,
    2. Anderson R.C.
    1986. The seed bank of a tallgrass prairie in Illinois. American Midland Naturalist 115:123–130.
    OpenUrl
  32. ↵
    1. Kinucan R.J.,
    2. Smeins F.E.
    1992. Soil seed bank of a semi-arid Texas grassland under three long-term (36 years) grazing regimes. American Midland Naturalist 128:11–21.
    OpenUrl
  33. ↵
    1. Kirk R.E.
    2013. Experimental Design, Procedures for the Behavioral Sciences, 4th ed. Thousand Oaks, CA: Sage Publications.
  34. ↵
    1. Klaus V.H.,
    2. Hoever C.J.,
    3. Fischer M.,
    4. Hamer U.,
    5. Kleinebecker T.,
    6. Mertens D.,
    7. et al.
    2017. Contribution of the soil seed bank to the restoration of temperate grasslands by mechanical sward disturbance. Restoration Ecology DOI 10.1111/rec.12626.
    OpenUrlCrossRef
  35. ↵
    1. Kozlowski T.T.,
    2. Pallardy S.G.
    2002. Acclimation and adaptive responses of woody plants to environmental stresses. The Botanical Review 68: 270–334.
    OpenUrlCrossRef
  36. ↵
    1. Mason A.C. Driessen,
    2. Norton J.
    2011. First year soil impacts of well-pad development and reclamation on Wyoming’s sagebrush steppe. Natural Resources and Environmental Issues 17:29–34.
    OpenUrl
  37. ↵
    1. Milberg P.
    1995. Soil seed bank after eighteen years of succession from grassland to forest. Oikos 72:3–13.
    OpenUrlCrossRef
  38. ↵
    1. National Oceanic and Atmospheric Administration (NOAA)
    2015. National Climatic Data Center. www.ncdc.noaa.gov/.
  39. ↵
    1. Natural Resources Conservation Service (NRCS)
    2015. Web Soil Survey. websoilsurvey.sc.egov.usda.gov/App/HomePage.htm.
  40. ↵
    1. Olsson A.D.,
    2. Betancourt J.L.,
    3. Crimmins M.A.,
    4. Marsh S.E.
    2012. Constancy of local spread rates for buffelgrass (Pennisetum ciliare L.). Journal of Arid Environments 87:136–143.
    OpenUrl
  41. ↵
    1. Pekas K.M.,
    2. Schupp E.W.
    2013. Influence of aboveground vegetation on seed bank composition and distribution in a Great Basin Desert sagebrush community. Journal of Arid Environments 88:113–120.
    OpenUrl
  42. ↵
    1. Reichman O.J.
    1984. Spatial and temporal variation of seed distributions in Sonoran Desert soils. Journal of Biogeography 11:1–11.
    OpenUrlCrossRef
  43. ↵
    1. Rivera D.,
    2. Jáuregui B.M.,
    3. Peco B.
    2012. The fate of herbaceous seeds during topsoil stock-piling: Restoration potential of seed banks. Ecological Engineering 44:94–101.
    OpenUrl
  44. ↵
    1. Rogers J.A.,
    2. Hsu J.C.
    2001. Multiple comparisons of biodiversity. Biometrical Journal 43:617–625.
    OpenUrl
  45. ↵
    1. Rokich D.P.,
    2. Dixon K.W.,
    3. Sivasithamparam K.,
    4. Meney K.A.
    2000. Topsoil handling and storage effects on woodland restoration in western Australia. Restoration Ecology 8:196–208.
    OpenUrl
  46. ↵
    1. Rosef L.
    2008. Germinable soil seed banks in abandoned grasslands in central and western Norway and their significance for restoration. Applied Vegetation Science 11:223–230.
    OpenUrl
  47. ↵
    1. Scoles-Sciulla S.J.,
    2. DeFalco L.A.
    2009. Seed reserves diluted during surface soil reclamation in eastern Mojave Desert. Arid Land Research and Management 23:1–13.
    OpenUrl
  48. ↵
    1. Shaukat S.S.,
    2. Siddiqui I.A.
    2004. Spatial pattern analysis of seeds of an arable soil seed bank and its relationship with aboveground vegetation in an arid region. Journal of Arid Environments 57:311–327.
    OpenUrl
  49. ↵
    1. Ter Heerdt G.N. J.,
    2. Verweij G.L.,
    3. Bekker R.M.,
    4. Bakker J.P.
    1996. An improved method for seed bank analysis: seedling emergence after removing the soil by sieving. Functional Ecology 10:144–151.
    OpenUrlCrossRef
  50. ↵
    1. Thompson K.,
    2. Grime J.P.
    1979. Seasonal variation in the seed banks of herbaceous species in ten contrasting habitats. Journal of Ecology 67:893–921.
    OpenUrlCrossRef
  51. ↵
    1. Traba J.,
    2. Azcárate F.M.,
    3. Peco B.
    2004. From what depth do seeds emerge? A soil seed bank experiment with Mediterranean grassland species. Seed Science Research 14:297–303.
    OpenUrlCrossRef
  52. ↵
    1. Vécrin M.P.,
    2. Grévilliot F.,
    3. Muller S.
    2007. The contribution of persistent soil seed banks and flooding to the restoration of alluvial meadows. Journal for Nature Conservation 15:59–69.
    OpenUrl
  53. ↵
    1. Wester D.B.
    1992. Viewpoint: Replication, randomization and statistics in range research. Journal of Range Management 45:285–290.
    OpenUrlCrossRef
  54. ↵
    1. Wester D.B.
    1995. Seed physiology and germination. Pages 168–238 in Bedunah D.J., Sosebee R.E. (eds), Wildland Plants: Physiological Ecology and Developmental Morphology. Denver, CO: Society of Range Management.
  55. ↵
    1. Yan Q.,
    2. Zhu J.,
    3. Zhang J.,
    4. Yu L.,
    5. Hu Z.
    2010. Spatial distribution pattern of soil seed bank in canopy gaps of various sizes in temperate secondary forests, Northeast China. Plant Soil 329:469–480.
    OpenUrl
    1. Zhang H.,
    2. Chu L.M.
    2013. Changes in soil seed bank composition during early succession of rehabilitated quarries. Ecological Engineering 55:43–50.
    OpenUrl
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Soil Seed Banks in Stock-piled Topsoils in the Western Rio Grande Plains, Texas
Mylea C. Lovell, Sandra Rideout-Hanzak, David E. Ruppert, Veronica Acosta-Martinez, Forrest S. Smith, Paula Maywald Stumberg, Keith A. Pawelek, Anthony D. Falk, David B. Wester
Ecological Restoration Sep 2018, 36 (3) 226-237; DOI: 10.3368/er.36.3.226
Twitter logo Facebook logo Mendeley logo
Bookmark this article

Jump to section

Keywords