Soil seed banks: a sneak preview into the future

By Sarah Barga

Many arid land plants possess seed dormancy, enabling them to delay germination until receiving environmental cues that stimulate development. Dormant seeds within the soil create seed banks that are a valuable resource for regeneration of native plant communities after disturbance. Seed germination and soil seed bank research is important to better understand soil seed bank dynamics, appropriately select restoration seed mixes, and gauge the restoration potential contained within existing soil seed banks.

My work has focused primarily on the Great Basin of the interior western United States, a cold desert containing vast expanses of sagebrush (Artemisia spp.) steppe shrubland. Native, herbaceous, understory plants add diversity to these shrub communities, but are often lacking in degraded areas dominated by introduced invasive species, such as cheatgrass (Bromus tectorum). Maintaining understory diversity within the sagebrush ecosystem is a growing concern for land managers, as these native plants provide forage and habitat for wildlife and can deter nonnative plant dominance.

Increased frequency and intensity of disturbances, associated with increased abundance of invasive species make maintenance and restoration of functional native sagebrush communities challenging. When disturbances, such as wildfire, occur repeatedly, sagebrush communities, and particularly those communities dominated by Wyoming big sagebrush (Artemisia tridentata ssp. wyomingensis), often transition to monocultures of introduced species (Fig. 1). Under these circumstances, active restoration may be the only way to recover native plant communities. For sites experiencing less frequent disturbance, less is known about how to aid recovery of native communities. More research is needed to identify when and where there are opportunities for passive restoration relying, at least partially, on natural regeneration from the soil seed bank.

  Figure 1.  Diagram of possible shifts in community composition following disturbance of sagebrush steppe plant communities. Repeated and frequent disturbance can shift communities toward monocultures of introduced species, which typically require active restoration; whereas, community changes resulting from less frequent disturbance likely vary with existing soil seed bank potential.

Figure 1. Diagram of possible shifts in community composition following disturbance of sagebrush steppe plant communities. Repeated and frequent disturbance can shift communities toward monocultures of introduced species, which typically require active restoration; whereas, community changes resulting from less frequent disturbance likely vary with existing soil seed bank potential.

Passive restoration, when an option, is the preferred land management approach. Existing soil seed banks may already contain adequate native plant diversity to restore the site’s native understory. Procuring native forb seed and ecologically appropriate collections of any species can be costly, difficult, and may lengthen the project timeframe. Active restoration also has the potential to negatively impact intact native seed banks in disturbed areas.

Passive restoration is not without its challenges. Successional processes can be very slow. While early successional shrub species, such as yellow rabbitbrush (Chrysothamnus viscidiflorus), can rapidly recolonize disturbed areas, they decline in abundance with sagebrush re-establishment, which may require more than 10 years. Perennial grasses can re-establish to pre-disturbance levels in as few as 2 to 5 years, if present in the seed bank; however, pre-disturbance levels can be highly variable. It is important to consider that early successional shrubs may be important for providing safe sites, with higher levels of nutrients and moisture, to encourage regeneration of native grasses and forbs.

Perennial forbs contribute most to the diversity of Great Basin sagebrush communities, but typically account for less than 10% of total cover and biomass. Long-term research on forb regeneration is generally lacking but suggests recovery to pre-disturbance levels is highly variable. Some perennial forbs are fire adapted and sprout after top-kill for rapid post-disturbance regeneration. Annual forbs tend to increase in abundance following disturbance but decline after about 5 years. In general, plant community disturbance responses are highly variable and may depend a great deal on the level of post-fire precipitation, the type and quantity of seeds present in the soil, and the state of the recovering plant community.

Regardless of whether active restoration is necessary or whether passive restoration is feasible, seed germination and soil seed bank research can inform decision-making when developing restoration protocols. If using an active approach, it is important to consider the species composition of the seed mix. The goal is generally to include plants whose seeds germinate readily, in order to produce immediate above-ground biomass. However, including plants whose germination strategy will produce persistent soil seed banks when future disturbances occur should also be considered. For example, Fig. 2 shows results of a trial that included 10 Great Basin native forbs where germination was recorded for different temperature and moisture treatments. It is clear that four species (bigflower agoseris - Agoseris grandiflora, maiden blue-eyed Mary - Collinsia parviflora, wingnut cryptantha - Cryptantha pterocarya, and slender phlox - Microsteris gracilis) germinate readily in response to all treatments, suggesting they would establish in the first year after seeding. In contrast, many of the remaining species achieved only partial germination in response to some or all treatments, indicating a portion of their seed may remain dormant in the soil seed bank, depending on the environmental conditions experienced in a given year, and could remain viable in the seed bank as a seed source after future disturbances. 

2a.png
  Figure 2.  Fraction of seeds germinated of Great Basin A) perennial and B) annual native forb species experiencing different factorial combinations of after-ripening temperatures (Cool - 21ᵒC, Hot - 40ᵒC) and moist, cool stratification (2ᵒC for 0, 2, 4, or 6 weeks corresponding to 15C, 2C2, 2C4, 2C6 respectively). Germination was evaluated in both cold stratification (2ᵒC) and 15ᵒC treatments, with some species germinating during cold stratification. Species include: A) AGGR -  Agoseris grandiflora -  bigflower agoseris, CHDO -  Chaenactis douglasii -  Douglas’ dustymaiden, CRIN -  Crepis intermedia -  limestone hawksbeard, PHHA -  Phacelia hastata  - silverleaf phacelia and B) BLSC -  Blepharipappus scaber -  rough eyelashweed, COPA -  Collinsia parviflora -  maiden blue-eyed Mary, CRPT -  Cryptantha pterocarya -  wingnut cryptantha, GIIN -  Gilia inconspicua -  shy gilia, MEAL -  Mentzelia albicaulis -  whitestem blazingstar, MIGR -  Microsteris gracilis -  slender phlox. Reprinted with permission from “Climate variability affects the germination strategies exhibited by arid land plants,” by S Barga, T Dilts, and EA Leger, 2017, Oecologia 185(3): 437-452. Copyright 2017 by Springer-Verlag GmbH Germany.

Figure 2. Fraction of seeds germinated of Great Basin A) perennial and B) annual native forb species experiencing different factorial combinations of after-ripening temperatures (Cool - 21ᵒC, Hot - 40ᵒC) and moist, cool stratification (2ᵒC for 0, 2, 4, or 6 weeks corresponding to 15C, 2C2, 2C4, 2C6 respectively). Germination was evaluated in both cold stratification (2ᵒC) and 15ᵒC treatments, with some species germinating during cold stratification. Species include: A) AGGR - Agoseris grandiflora - bigflower agoseris, CHDO - Chaenactis douglasii - Douglas’ dustymaiden, CRIN - Crepis intermedia - limestone hawksbeard, PHHA - Phacelia hastata - silverleaf phacelia and B) BLSC - Blepharipappus scaber - rough eyelashweed, COPA - Collinsia parviflora - maiden blue-eyed Mary, CRPT - Cryptantha pterocarya - wingnut cryptantha, GIIN - Gilia inconspicua - shy gilia, MEAL - Mentzelia albicaulis - whitestem blazingstar, MIGR - Microsteris gracilis - slender phlox. Reprinted with permission from “Climate variability affects the germination strategies exhibited by arid land plants,” by S Barga, T Dilts, and EA Leger, 2017, Oecologia 185(3): 437-452. Copyright 2017 by Springer-Verlag GmbH Germany.

When considering passive restoration, understanding the composition of existing soil seed banks is critical (e.g. the relative density of seeds from invasive and native species). My research included a study that examined soil seed bank composition at sagebrush sites with different disturbance histories, representing various condition classes. Site characteristics, including: shrub cover, ground cover, climate, grazing use, and fire history, were measured at each soil seed bank collection location. In general, soil seed bank samples contained considerable native forb diversity (Fig. 3). When assessing relationships between site characteristics and seed bank composition, we found cover of the three most dominant shrubs (Wyoming big sagebrush - A. tridentata ssp. wyomingensis, yellow rabbitbrush – C. viscidiflorus, and rubber rabbitbrush - Ericameria nauseosa) was related to seed densities of native annual and introduced species (Fig. 4). In fact, shrub cover was a strong predictor of several aspects of soil seed bank composition, including the presence of rare species and a variety of species diversity measures.

  Figure 3.  Species richness of native and introduced species in soil seed bank collections from sagebrush steppe communities with different disturbance histories in northeastern Nevada (S Barga and EA Leger, unpublished).

Figure 3. Species richness of native and introduced species in soil seed bank collections from sagebrush steppe communities with different disturbance histories in northeastern Nevada (S Barga and EA Leger, unpublished).

  Figure 4.  Relationship between cover of dominant shrub species (Wyoming big sagebrush -  Artemisia tridentata  ssp.  wyomingensis , yellow rabbitbrush -  Chrysothamnus viscidiflorus , and rubber rabbitbrush -  Ericameria nauseosa ) at soil seed bank collection sites and seed density of native annual, native perennial, and introduced species in soil seed bank collections. Reprinted with permission from “Shrub cover and fire history predict seed bank composition in Great Basin shrublands,” by S Barga and EA Leger, 2018, Journal of Arid Environments 154: 40-50. Copyright 2018 by Elsevier.

Figure 4. Relationship between cover of dominant shrub species (Wyoming big sagebrush - Artemisia tridentata ssp. wyomingensis, yellow rabbitbrush - Chrysothamnus viscidiflorus, and rubber rabbitbrush - Ericameria nauseosa) at soil seed bank collection sites and seed density of native annual, native perennial, and introduced species in soil seed bank collections. Reprinted with permission from “Shrub cover and fire history predict seed bank composition in Great Basin shrublands,” by S Barga and EA Leger, 2018, Journal of Arid Environments 154: 40-50. Copyright 2018 by Elsevier.

Restoring native plant communities is challenging and soil seed bank dynamics in natural systems are not well understood. From my perspective, the characteristics that make studying forbs complicated are also what make them interesting. Variation in seed dormancy from species to species, and even from population to population, can be considerable. Little is known about seed movement and presence or abundance of viable seed within soil seed banks. Understanding how variation in these characteristics translates into species and ecosystem persistence could foster successful incorporation of native species into restoration protocols. With additional research, I’m hoping that these characteristics will be seen as less of a burden for restoration practitioners, that we can gain an appreciation for the natural variation in life history characteristics across species, and that we learn how to incorporate this variation into restoration planning as a way to increase long-term restoration success.

 

Contact Information:

Sarah Barga, PhD

Research Geneticist

Rocky Mountain Research Station, U.S. Forest Service

sbarga@fs.fed.us