How to Develop a Seeding Prescription for Revegetation and Restoration



Introduction

Sowing of native grasses and other herbaceous plant seed is typically undertaken to control erosion on newly exposed soil along forest roads (see photo at right), to de-compact and add soil organic matter to old landings and log sort yards, to enhance grazing opportunities, and to restore the indigenous vegetation of grassland/meadow communities. Local or regionally adapted native plant seed is likely to remain a valuable commodity in most parts of Canada for the foreseeable future, whether collected from the wild or grown in cultivation. Consequently, every effort must be made to optimize its use and to maximize its effectiveness for revegetation and restoration.

       While revegetation procedures will vary with local site conditions and project goals, some generalizations can be made. Foremost among these general rules are to:

  • introduce seed to freshly loosened soil;
  • match the species selected to local soil and site conditions;
  • avoid or minimize the effects of weeds;
  • remember that stand establishment will often take two growing seasons; and
  • employ an adaptive management approach (testing different techniques and monitoring their effectiveness on an operational basis).

logging road
Exposed soil and run-off
along a newly built logging road.





For details on growing native plants in cultivation, see the web page on "How to Produce Seeds of Native Herbaceous Plants in Cultivation."



Site Preparation

It is important to prepare a receptive seedbed that will have good mineral soil exposure, good soil-seed contact, and will support unimpeded root growth. Many sites degraded by industrial activities such as road building, log loading, or overgrazing have very compacted soil. This first must be alleviated by deep cultivation, disking, or ripping. The rooting zone should be loosened (cultivated) to 10 or 15 cm if possible, and heavily compacted sites (such as former roads and landings) may need to be ripped or treated with a winged subsoiler to depths of 40 cm. Simple raking or harrowing of the surface is insufficient to support optimal root growth in most cases, but is better than no treatment at all. The removal or rearrangement of logs, brush, and stones is not necessary or desirable on most revegetation sites, as these features contribute to microsite diversity.

Freshly loosened soil is essential for good seed germination.
 
Simple discing, rototilling, harrowing or raking can help.
There is no point throwing seed on hardened ruts, which should first be loosened and filled in.

       If the remnants of the pre-existing native plant community are already found on the site, one may not want to disrupt the vegetation, no matter how depauperate or degraded it might be. The existing vegetation can be directly “over-seeded,” but provisions for inserting the seed into or onto the soil must be considered, and some degree of site preparation is still desirable. This may consist of mowing the existing vegetation to a short stubble, or raking/harrowing the soil surface to expose some bare soil. Manual removal of weeds or spot-spraying of weeds and other exotic plants with a broad-spectrum herbicide may be desirable. If the weeds are annuals, mowing or clipping them at the time of flowering (i.e., before they go to seed) can help shorten their hold on the site. When digging up perennial weeds such as bull thistle, the filled-in shovel holes left behind then consist of loose, bare soil that is well suited for immediate seeding with native plants.

       When tillage or harrowing is conducted on sloping terrain, it is essential that furrows run parallel to the contours (perpendicular to the direction of slope) in order to minimize erosion. Very steep road cuts are difficult to revegetate unless they are terraced or re-sloped to the natural angle of repose for loose, unconsolidated material. A more common solution, employed for both agronomic and native species, is to use hydroseeding techniques in which commercial “tackifiers” can be added to a slurry of water, seed, fertilizer and mulch that is then applied to the site using a specialized pump and nozzle. The tackifier essentially glues the seed to steep slopes, so some seed will be in place to germinate and establish if weather conditions cooperate.

       Optimal conditions for seeding disturbed lands exist immediately after soil disturbance or site preparation. If disturbed soil is left too long, weeds will have a chance to invade and establish. Weathering results in soil settling over time, and silt and clay particles will fill most soil cracks, resulting in a gradual recompaction of the soil.




Soil Amendments and Mulches

Sites degraded by industrial activities often consist of compacted soils, stripped of topsoil and plant nutrients. Consequently, some degree of soil improvement is often needed in order to support more than sparse plant growth, and to restore healthy ecosystem functioning. When the area to be revegetated is dominated by subsoil, bare parent material, or shattered lithic material, one challenge is to accelerate soil development. In addition to decompacting this material (as described above), it can also be useful to incorporate organic matter into the substrate to improve soil structure and to provide cation-exchange sites for the retention of nutrients and soil water. Suitable amendments can include wood chips, sawdust, peat moss, straw, manure, or various waste treatment sludges (e.g., from pulp mills, fish farms, or municipal sewage treatment plants). Care must be taken to manage the carbon to nitrogen (C to N) ratio of these amendments. If material with a high cellulose content (e.g., wood chips, sawdust, or straw) is added to the soil, it will be decomposed by soil fungi and bacteria, but those fungi and bacteria will gobble up most of the available nitrogen in the soil, thereby depriving plants of adequate resources for growth. So any time amendments with high C:N ratios are used, supplemental fertilization with nitrogen-rich or low C:N materials must also be done. This fertilizer can be from commercial sources (e.g., ammonium nitrate formulations), or from organic sources (such as manures, or municipal sewage).

       Soil amendments should be well incorporated into the rooting zone of the substrate, typically the top 10 to 20 cm, before seeding. This will usually require the use of farm machinery such as a disker, plow or cultivator, as part of site preparation procedures. If applied as a surface dressing or mulch (see photo below), amendments are not as effective for soil improvement, but can be important for erosion control, especially on sloping sites. Surface mulches are generally applied after seeding, rather than before. Mulch should not be applied too thickly (generally less than 1-2 cm) or densely, (<70% cover), so that seedling emergence is not inhibited. Suitable mulches for revegetation and restoration activities include straw from annual cereal crops (i.e., wheat, oats, barley, rye), or the straw and screenings from native plant seed production fields. Care must be taken not to introduce weed seeds or other contaminants with the mulch. For example, hay bales from fields of domesticated smooth brome, timothy and clover might contain viable seeds of those exotic agronomic species. So the use of hay mulch is not recommended, as the exotic seeds might defeat the purpose and expense of introducing native plants in the seeding process.


Volunteers applying native grass mulch to a restoration site after seeding.

       Fertilizer is often applied when sowing a revegetation site, even if organic matter is not incorporated into the soil. This is because most degraded sites are nutritionally barren, with substrates often consisting of unmodified glacial till. Furthermore, the enhanced plant growth achieved through the use of fertilizer can stretch sparse supplies of native seed by generating more plant cover per plant and greater probability of seedling survival (see our poster on this topic). Where no intact topsoil remains, a balanced, high-concentration fertilizer (e.g., 18-18-18 or 13-16-10 NPK) applied at a rate of 200 to 300 kg/ha is generally sufficient to promote vigorous plant growth. Supplemental fertilization may be beneficial in another three years or so, depending on the chemistry of the site, the effectiveness of plant establishment and the initiation of nutrient cycling above and below ground, and on the success of nitrogen-fixing species. Natural fertilizers such as manure, fish farm waste, or municipal sludge can be beneficial substitutes, since they also include organic matter. If possible, soil testing should be conducted to document the precise nutrient limitations, so that fertilizer prescriptions can be optimized. The heterogeneity of soil conditions on many revegetation and restoration sites often precludes accurate diagnostic soil testing. It is safe to generalize, however, that N will usually be more limiting than P, K, or other nutrients on degraded soils in northern and western forests. Contrary to some recommendations, our research has shown that fertilization even benefits native plant establishment on sites dominated by agricultural weeds.




The Seeding Prescription

Designing the mixture of species and the density at which they should be applied to a site constitute the fundamental elements of a revegetation prescription. To optimize the use of scarce seed, it is strongly recommended that sites be roughly mapped and categorized according to soil texture, topographic position, and/or vegetation goals, so that customized seed mixtures can be assigned to each site class. Gravelly soils in low-lying sites should receive a different mixture of species than gravelly sites on upper slopes and ridge crests, while fine-textured or loamy soils require yet a different suite of species. Site preferences and tolerances, where known, are provided as part of the individual species accounts in our manual.

       In formulating a seed mix, it is useful to employ a combination of 5 to 10 species, consisting of some graminoids (grasses, sedges, and rushes) of low and high stature, some species with rhizomes, some species (usually legumes) with nitrogen-fixing ability, and some fast germinators. Among the species with which we have worked, the fastest germinators were Achillea millefolium and Festuca occidentalis. Fast germinators and fast growers are desired to for erosions control and to quickly green up a site, but you will also want long-lived, though perhaps slow to establish, species such as Carex aenea too. Care should be taken to ensure that the mix will not be dominated by a single aggressive species, and the mature stature of each species should be considered when deriving the ratio of seeds to use. That is, only a few individuals of large-statured species (e.g., Lupinus polyphyllus) are needed, while many individuals of smaller plants (e.g., Festuca occidentalis) are required to achieve the same proportional crown cover per unit area. Other considerations, such as the decision to include or exclude species highly palatable to livestock or wildlife, or the decision to include species resistant to trampling, will enter into the design of a mixture according to the land use goals for the land being revegetated. The soil texture, site drainage, and soil fertility of the site to be revegetated must also be considered and matched to the most suitable species. For example, many of the species with which we have worked can establish and grow well on gravelly soils, but Dryas drummondii and Epilobium latifolium require both coarse (sandy, gravelly) soils and good access to soil water, so these species are suitable for the base of old gravel pits or borrow pits, but not their walls. Geum macrophyllum, in contrast, is better suited to loamy soils with high nutrient levels, generally where the original topsoil is still intact.

       It is worth considering the inclusion of a fast-germinating and fast-growing “cover crop” on steep bare sites where heavy precipitation or spring runoff is expected to generate a risk of soil erosion. Most native plants (with some exceptions) are relatively slow to establish and achieve full stature, so some non-native species would need to be used. The challenge is to find species that will quickly generate cover, but will not persist and compete with the native plants as they establish and grow. Suitable cover crops include fall rye (Secale cereale), Italian ryegrass (Lolium multiflorum), and sterile hybrids of slender wheatgrass and wheat (Elymus trachycaulus x Triticum aestivum, sometimes marketed under the trade name of RegreenTM). When added to a native plant seed mixture, such cover crops should be added on top of the desired native plant seeding rate, and should not constitute more than 10 to 20% of the total seed mixture applied to the site.

       All seeding ratios and densities should be formulated on the basis of the amount of pure live seed (PLS) per unit area, not the weight of seed or seed stocks. This is because plant species vary greatly in seed size and consequently in the number of seeds per kg, and individual seed lots also tend to vary in the number of viable seeds per unit weight (relative to other contaminants like dust and plant debris). Adjustments for PLS are required in order to refine the amount of seed to be weighed out and applied to a site. For example, if your seedlot is 90% pure mature seed (by weight), and those seeds exhibit 90% viability, then every 25 kg bag of seed only contains 0.9 x 0.9 x 25 = 20.25 kg of pure live seed. Combined with knowledge of the number of seeds per gram for a particular seedlot or species (as provided for each species in our manual), this determination allows the inter-conversion of seeding densities in PLS/m2 and in kg/ha. Prescriptions should be developed in PLS/m2, but those prescriptions are usually implemented in the field in terms of kg/ha. Individual species treatments in the manual include the mean and range of the number of seeds per gram, and the mean and range in germination capacity encountered in the Symbios research program; seed lot purity varies with every crop and every seed cleaning operation, however. The following table provides a sample worksheet for the preparation of a seeding prescription. Such a table can be easily transcribed to a computer spreadsheet program, in which the amount of seed of each species to be weighed out and the total application rate can then be calculated.


       The amount of PLS applied to a site will not result in an equivalent density of plants. Though all PLS should theoretically germinate, our experience indicates that only 10% to 20% of those seeds will successfully germinate, emerge, and establish as seedlings on degraded sites. This low success rate is due to a variety of suboptimal practices (e.g., surface sowing instead of drill seeding), harsh site conditions (e.g., infertile or compacted soil), and accidents of weather and herbivory (e.g., frost, drought, grazing by insects or small mammals). Recent research with plants native to northern British Columbia indicates that adequate amounts of cover can be achieved across a broad range of sowing densities (375 to 6000 PLS/m2). With fertilization, full cover is attained faster, and can be achileved at densities ranging from 750 to 1500 PLS/m2 (see presentation on this topic). A broadly applicable and generally acceptable recommendation is for 1500 PLS/m2 with 300 kg/ha of 18-18-18 N-P-K when broadcast-sown as a dry seed mix, with higher rates recommended on erosion-prone sites, where rapid green-up is desired, or where weeds populations are high. Lower densities are acceptable if seed is in short supply, if weeds are not a threat, and if establishing cover quickly is not a priority.




Sowing

Seed can be introduced to a site by drilling it into the soil using standard or modified agricultural machinery, broadcast using cyclone spreaders (mounted on a helicopter sling, on the back of a tractor, an ATV, or operated by hand), spread directly by hand, or as part of a slurry by hydroseeder. Drill seeding is most efficient in terms of ensuring that a large proportion of the seeds will have good contact with the soil and will successfully germinate and emerge; unfortunately, it is not suitable for rough or steep terrain, and is not currently the norm for most roadside seeding. Next efficient is dry broadcast seeding (as in the photo below), so long as it is conducted quickly after soil disturbance or after soil loosening by raking or harrowing, and then is followed up by more raking or harrowing. Small areas can be sown by hand if care is taken to distribute seed evenly. Hydroseeding is least efficient in terms of the use of seed, but the presence of tackifier (an organic soil binder) can greatly enhance its effectiveness on steep slopes. Typically mounted on the back of a large truck along with a big mixing tank, hydroseeders can also be small enough to fit on a pickup truck. Simple substitutes can be built using some water pumps, so long as the pumping mechanism does not damage seeds and nozzle apertures are big enough for the seeds being applied. Hydroseeding is not currently recommended for most native plant seed application purposes (except for steep sites), because it uses approximately three times the amount of seed that is used in dry broadcast seeding applications. A large amount of hydroseeded seed ends up being applied to unsuitable microsites and obstacles, and much of the seed is found in the upper layers of the mulch where it dries out, rather than under the mulch where it is protected. So a recommendation of 1500 PLS/m2 for dry seeding should be adjusted to 4500 PLS/m2 when hydroseeding is used.

       Domesticated legumes are routinely treated with bacterial inoculum before being sown. This may not be necessary for native legumes on forest soils, for which natural inoculum seems to be widespread in the forest environment. However, for very sterile or isolated locations such as large landings and mines, it may be prudent to coat legume seeds with commercial inoculum before sowing. Different strains of Rhizobium are needed for each legume genus, with each vendor using different names or labels for the strains applicable to Lathyrus, Lupinus, or Vicia. Inoculum can be mixed in a powdered milk slurry so it lightly coats the legume seeds before they are added to the seed mix.

       It is generally recommended that one weigh out and mix the different species in a seed mixture ahead of time (i.e., in the warehouse). When dry seeding, fertilizer and mulch (if used) are generally applied to the site in separate steps from the seed. It helps to weigh out bags or tubs of seed for each discrete stratum (e.g., moist sites, ridge crests, gravel patches, etc.) or fixed areas of uniform land (e.g., in 100 m2, 900 m2, or 2500 m2 units). Then flag out the boundaries or corners of each unit of land on the revegetation site, so that the rate of seeding can be adjusted to make sure it uniformly covers the designated area. Until experience is acquired at judging the rate of application, it is better to seed sparsely at first and then do a supplemental application, rather than to run short of seed because original application rates were too generous. The site should be raked or harrowed to expose fresh, loose soil immediately before seeding (as mentioned above), and then raked, harrowed or dragged to cover the seeds somewhat after sowing too (see photo at right). In northern British Columbia, seeding can be done in the spring (any time before July) or in late summer, but late fall seeding is often best.

quad seeding
Dry broadcast seeding using a cyclone
seeder mounted on an ATV.











raking
Raking seeds in after sowing.












Monitoring and Evaluation

All revegetation and restoration prescriptions and subsequent seeding operations using native species should be considered experimental in nature. That is, they may represent your best effort at designing an appropriate mixture and density of seeds and associated amendments, but there is no a priori reason to expect this design to be the optimal solution to local site management challenges. It is therefore prudent to incorporate assorted modifications to the restoration prescription, apply them at multiple locations, keep careful records of what treatment was employed where and when, and to monitor the results. Monitoring might be as simple as the installation of permanent photo points, with photographs taken of each treatment area at the same time of year over a number of years. Or it might be as quantitative as a rigorous program utilizing line transects or randomly located quadrats to sample plant cover and density (see photo, below). A template for recording plant density and cover observations in three quadrats per treatment stratum is provided as a field monitoring form, below. Note that it is usually worthwhile to record weed cover and evidence of soil erosion and the accumulation of organic matter as well as the abundance of each plant species. A more rapid form of monitoring can consist of recording plant cover in four broad groups: sown species, other native plants, agronomic grasses and legumes, and weeds. Whatever the nature and intensity of monitoring undertaken, it is important to adopt a philosophy of structured adaptive management, constantly improving one’s expertise in restoring native vegetation.


Monitoring, evaluation, and prescription refinements based on local experience
are key to any successful program of revegetation or ecosystem restoration.

Sample data collection form for monitoring plant community development after revegetation.
This form records plant count and cover observations from three quadrats; remember to record quadrat area, e.g., 0.25 m2 or 1.0 m2);
x and y values denote random rectangular coordinates for each sample. Sown species are pre-listed on the form,
along with a number of other species (native invaders, agronomics, and weeds) commonly encountered in sub-boreal B.C.

       Revegetation success ultimately depends on the establishment of an adequate amount and composition of plant cover. Depending on the goals of the project, the vegetation may be expected to be a faithful re-creation of a natural plant community, and to dominate the site indefinitely. In other situations, it will be sufficient for the vegetation to reduce erosion, and to cover the soil for only a few years until tree and shrub species dominate the site. Mowing or weeding of exotic plants may be desired in some cases, and supplemental fertilization in two or three years may be appropriate, depending on the land use goals. But areas revegetated to native plants are generally expected to be self-maintaining.

       The value of creating a plant assemblage that is pleasing to the eye (see photo, below) should not be underestimated, as public support for the wider use of native plants still needs to be nurtured. Coupled with a desire to promote indigenous biodiversity and to mimic the composition of natural meadows, such aesthetic considerations provide another incentive for including non-leguminous wildflowers in native plant seed mixtures.


A successfully established stand dominated by Elymus glaucus, Festuca occidentalis, and Achillea millefolium, with scattered Bromus ciliatus, Lupinus arcticus, Geum macrophyllum, Carex species and Aster species. This mixture was sown along recent ski trail improvements and along a forest access road in the Smithers Community Forest, with seeding done by volunteers from the local cross-country ski club. This photograph was taken two growing season after sowing.



Authorship: Carla M. Burton and Philip J. Burton (copyright) 2004.

[ Symbios Home | Native Plants Project | Growing Native Plant Seed ]