Isatis tinctoria
(dyer's woad)

Isatis tinctoria is a herbaceous dye plant which has spread, often by human intervention, widely from its area of origin in central Asia or southeast Russia. Long valued for the dye it produces, it is of pest status in alkaline soils of some western states of the United States. It is the main biotic threat to the endangered Siskiyou Mariposa Lily (Calochortus persistens) found on the California-Oregon border and also threatens the endangered Yreka phlox, Phlox hirsuta. It responds well to high inputs of fertilizer and water, but also competes well in disturbed, dry and nutrient poor conditions. Its fruits have germination-inhibiting factors and the plant produces glucosinates which may alter soil microbiological properties. It causes considerable production losses of desired plants in rangelands. Control methods are effective, but with logistical challenges, especially in more remote areas. A native rust, Micropuccinia thlaspeos, limits, but does not prevent spread. Arthropod biological control agents are being researched in Europe, but to date none are specific enough.

The centre of origin of I. tinctoria is thought to be Central Asia (Spataro et al., 2007). It is also reported as a native of southeastern Russia (Hurry, 1930; McConnell et al., 1999), to south-west Asia and perhaps to some parts of south-eastern Europe (Ball and Akeroyd, 2010) and to Afghanistan, western Tibet and China (Anon, 2015a). Hegi (1986) reports it to be native to the steppe area around the Caucasus, Central Asia to eastern Siberia and the Middle East. It is recorded as present in all European countries other than Ireland and Iceland and islands such as the Balearics and Crete (Ball and Akeroyd, 2010, Marhold, 2011), although Reynolds (2002) also reports it, as rare, in Ireland. It has been widely introduced globally.

It is probable that I. tinctoria has already established itself in appropriate habitats through many centuries of trade, travel and colonization. Renewed interest in the use of woad may lead to wider distribution through trade routes, but at a low level compared with previous times. Where it is a serious weed (largely some states in the western USA) legislation should prevent this movement. 

It has been suggested that I. tinctoria is well adapted to the western USA, because the plant is adapted to alkaline conditions and an arid climate (McConnell et al., 1999). The species thrives on rocky soil with limited water-holding capacity (McConnell et al., 1999), but it can grow well under a wide variety of habitats. This includes special habitats where rare plants are found – the endangered Siskiyou Mariposa Lily (Calochortus persistens) was thought to exist in one site, in Siskiyou County, California, where it is found on open ridgetops and north-facing slopes, with bedrock of gray phylite and sandstones high in potassium and low in calcium, iron and magnesium (Diggles et al., 2003). An Oregon population also exists (Anon, 2012). Equally I. tinctoria can dominate plant communities in pastures, forests and rangelands when it occurs in pest status in various western states of the United States. Reeves (2010) who recorded its growth on dry, coarse textured rocky soils also noted it capable of invading undisturbed and disturbed sites such as roadsides, railroad right of ways, pastures, grain and alfalfa fields, forests and rangelands, especially big sagebrush (Artemisia tridentata) communities. In Utah, I. tinctoria prefers loose, alkaline soils (Varga and Evans, 1978). It also grows on mining spoil heaps (Durães et al, 2015), lava (Anon, 2015c) and under cultivation with the addition of high levels of fertilizer (Howard, 2010). Hurry (1930) noted that it responds well to intensive cultivation.

The wide geographic range of I. tinctoria in Europe and Asia may be an important feature of its success in the western USA. It has been suggested that widely distributed species have developed the ability to tolerate varied biotic and resource conditions and to resist many potential enemies (Pyšek et al., 2015). I. tinctoria does show a high degree of genetic diversity (Gilbert et al., 2002).

I. tinctoria competes with plants of pastures, forests and rangeland and with cereal crops and alfalfa. It is most troublesome on rangeland, in disturbed non-crop sites and in undisturbed natural areas in the intermountain west (DiTomaso et al., 2013). Callihan et al. (1984) list 23 plants of rangeland, 13 of pastures and cropland and 12 of disturbed areas that were observed, collected or associated with I. tinctoria. The rangeland habitats were dominated by Artemisia tridentata (big sagebrush). I. tinctoria is reported to quickly suppress annual grasses in rangeland, but perennial grasses appear to co-exist moderately well (Callihan et al., 1984). Perennial grasses were the main native grass species in western rangeland before being replaced by introduced annual grasses (DiTomaso, 2000) which were aided by overgrazing by introduced domestic animals. These livestock tend to avoid grazing I. tinctoria (DiTomaso, 2000).

Genetics

A tetraploid outbreeding plant (Spataro and Negri, 2008a) (2n = 4x =28). The limits of the genus Isatis are debatable and it is likely that the number of species will be greatly reduced (Al-Shehbaz et al., 2006). Spataro et al. (2007) carried out a study of fifteen Eurasian populations of I. tinctoria and demonstrated wide variation within and between populations. This was reinforced with studies showing wide adaptability to different mountain conditions and populations with high morphologic and genetic variability (Spataro and Negri, 2008b). Polyploid plants (with three or more complete chromosome sets) have been obtained and subsequent lines were characterized by high fertility and yields (Qiao et al., 1989).

Reproductive Biology

This paragraph is taken from Zouhar (2009 and references therein) unless stated.

I. tinctoria requires vernalization to achieve flowering, with crown rosettes (second year) and seedling rosettes responding differently to temperature regimes. Seed production is high, at 350-500 per plant, although there are reports of 10,000 from selected plants (McConnell et al., 1999). Kizil (2006) reported a much lower average of 88 seeds per plant from Turkey. They are still contained within fruits (silicles) when these are shed, usually at maturity although some remain on the plants until winter. Fruits generally drop close to the plant, mostly within 50cm, with longer distances being achieved through ant dispersal, human and animal activity, and water. Much dispersal has been through contaminated seed or baled alfalfa hay. The fruits contain water soluble germination inhibitors which may reduce inter- and intra-specific competition; seeds outside the fruits germinate easily (Young and Evans, 1971). Seed viability within the soil appears limited, with most seeds non-viable within 12 months, but this may be related to levels of soil moisture. Seedling establishment seems to be promoted by close proximity to sagebush plants, but in the field 97% of seeds fail to achieve successful establishment. Final numbers of established plants appear to be similar regardless of initial seed levels. Typically, germination occurs in the spring or autumn, the plant remains as a rosette of basal leaves the following summer and winter, and flowers in the second year with seeds ripening in June and July (McConnell et al., 1999). Plants die after producing seed according to Jacobs and Pokorny (2007), but Callihan (1990) states that the crown buds often produce new growth for additional years.

Physiology and Phenology

The combination of the tap root and lateral roots enables moisture to be obtained from an extensive soil profile, increasing its competitive fitness relative to annual and perennial plants in semi-arid areas (Jacobs and Pokorny, 2007). This two layered rooting capacity has been postulated to provide I. tinctoria with a strong competitive advantage in sagebrush (Artemisia spp.) in the Intermountain West (Sturges, 1977). Woad fruits contain water-soluble germination-inhibiting substances (Young and Evans, 1971); these have an effect on a restricted group of competing seeds, both of I. tinctoria and other species, but the reliance on rainfall to leach away these substances presumably enhances the chances of germinating with suitable soil moisture.

Brassica crops often appear to have an inhibitory effect on subsequent crops and this has been linked to levels of indole glucosinates or their breakdown products; I. tinctoria specifically has been implicated in reducing soil fertility (Iqbal, 2012). Related brassicas are known to be heavy users of nitrogen and phosphorus, as well as water (Varga and Evans, 1978). I. tinctoria seeds contain glucobrassicin and shoots contain this and also two other indole glucosinates (Elliott and Stowe, 1971b) which, the authors suggested, may explain the deleterious effect of the species on subsequent plantings (and possibly on existing competing plants). Some products formed by glucobrassicin decomposition have high goiter-inducing activity for grazing animals (Elliott and Stowe, 1971a) and the chemical composition may inhibit extensive grazing. Consequently, it may be postulated that these compounds suppress competing plants and reduce levels of herbivory. Isothiocyanates, derived from glucosinolates, inhibit nitrification processes by direct effects on the size of communities of nitrifying bacteria and by reducing their nitrifying activities (Bending and Lincoln, 2000), further modifying the original environment. The plant also shows competitive physiological adaptations in relation to nitrogen, light and water.

Monaco et al (2005) hypothesized that I. tinctoria would demonstrate low plasticity in growth and physiology with variable soil nitrogen and high morphological or physiological plasticity in response to variable light and water. This was confirmed; the plant increased leaf area when grown under shade compared with greater light and increased overall productivity with additional water. The results supported the theory that early succession of plants on poor soils is based on colonizing, not competitive, ability (Huston and Smith, 1987; Tilman and Wedin, 1991).

Longevity

Although described as a winter annual, biennial or short-lived perennial (McConnell, 1990) I. tinctoria can switch to vegetative reproduction through buds on the root crown (Farah et al., 1988) and from the taproot if the rosette leaves are removed (Roché, 1992). Vorobiov (1960) describes I. tinctoria as a biennial requiring one hibernation to bolt and flower.

Population Size and Structure

I. tinctoria may occur in small patches or as scattered plants or occupy large areas virtually as monocultures (Varga and Evans, 1978), spreading rapidly. The spread of the species was estimated at 14% per year in the Pacific Northwest (Jacobs and Pokorny, 2007). In 1990, I. tinctoria was estimated to occur in 10,000 ha in Idaho (Callihan, 1990). In Montana, I. tinctoria populations were anticipated to cover 39,000 ha, but Montana’s Dyer’s Woad Cooperation Project reduced the area to 2.6 ha (Pokorny and Kreuger-Mangold, 2007).

Nutrition

When grown commercially, the addition of mineral or organic nitrogen fertilizer resulted in increased biomass (Orsini et al., 2012); nitrogen fertilizer requirements may be around 220 kg/hectare (Anon, 2004). Monaco et al (2005, see Physiology above) demonstrated that the plant has low plasticity in relation to soil nitrogen with relatively small differences occurring between different nitrogen levels at any specific water status of the soil. It can thrive on poor rocky soils with limited water availability (McConnell, 1990) including those high in potassium and low in calcium, iron and magnesium (Diggles et al., 2003); equally it can grow well in rangeland, agricultural fields and forests covering a range of fertility types.

Associations

Associations with Artemisia tridentata have been noted (Callihan et al., 1984).

Environmental Requirements

I. tinctoria is associated with loose, alkaline soils in Utah and, more generally, in the western United States (Varga and Evans, 1978; McConnell 1999) and for domestic cultivation a sandy-loamy or gritty-loamy soil of pH 8-10 is recommended (Hortipedia, 2015). When cultivated it can utilize high levels of fertilizer and water, but is capable of competitive growth under arid and nutrient poor conditions. There is no information available for temperature requirements, but its range and habitats, through Eurasian steppes to Morocco and Arizona, shows that it can survive very cold winters and also very hot conditions. I. tinctoria is often found in association with Artemisia tridentata which is capable of surviving -18°C to 38°C extremes in the field (Loik and Redar, 2004) and short term tolerance to 46°C was shown by leaves (Loik and Harte, 1996). It may be conjectured that I. tinctoria has a similar capacity to survive these conditions. 

Many incidental natural enemies occur; Hurry (1930) lists a few including the turnip fly, Phyllotreta nemorum and the cabbage caterpillar Pieris brassicae and fungal pathogens such as Aecidium isatidis, Ophiobolus herbarum, Cicinnobolus cesati [Ampelomyces quisqualis], Monilia glasti and Oidium erysiphoides [Podosphaera fuliginea]. Hinz et al. (2007) list 56 insect species, two nematodes and four fungal pathogens associated with I. tinctoria found from Eurasian literature. In China, Ma (2007) obtained 172 fungal isolates from I. tinctoria (I. indigotica) exhibiting root rot, comprised of Fusarium solani [Haematonectria haematococca], F. oxysporum and Rhizoctonia cerealis [Ceratobasidium cereale]. Alternaria brassicicola causing leafspot on I. tinctoria was reported by Gao et al (2014) who also noted Alternaria napiformis and A. brassicae that have previously been reported as causal agents of leaf spot on the plant. The eriophyid mite Metaculus rapistri was collected from I. tinctoria in Turkey (Monfreda and Lillo, 2012), however this was a misidentification and a description of the Metaculus sp. that was collected from I. tinctoria is currently underway (P. Weyl, CABI Switerland, personal communication, 2017) and Cortat et al. (2008) record nine insect species that caused sufficient impact as to be considered for biological control studies.

Mammals have been reported as grazing on the plant; sheep have been deliberately used in grazing trials. Almost certainly a far wider range of pathogens and herbivores than noted at present, incidentally attack I. tinctoria. Further details on promising biological control agents are given in the Prevention and Control section of the datasheet.

Animal feed, fodder, forage

• Fodder/animal feed
• Forage

Environmental

• Erosion control or dune stabilization

General

• Ritual uses
• Sociocultural value

Genetic importance

• Gene source

Human food and beverage

• Vegetable

Materials

• Dye/tanning
• Dyestuffs
• Green manure
• Pesticide

Medicinal, pharmaceutical

• Source of medicine/pharmaceutical
• Traditional/folklore

Largely achieved after growth of the plant, rather than through inspection. Detection in seed, forage or similar is achievable.

Ball and Akeroyd (2010) state that I. tinctoria is very variable, especially in the size, shape and pubescence of the silicula. The genus is difficult and requires revision and a number of the present species are probably subspecies of I. tinctoria (Ball and Akeroyd, 2010); Al-Shehbaz et al (2006) substantiate this view. Callihan et al. (1984) list species that may be confused with I. tinctoria including: Lomatium dissectum (biscuit-root), Descurainia sophia (flixweed), Brassica spp., Sisymbrium altissimum (Jim Hill mustard) and other mustards, Crepis acuminatus (tapertip hawksbeard) Senecio spp, Balsamorhiza sagittata (arrowleaf balsamroot), Wyethia amplexicaulis (mule’s ears), Camelina microcarpa (falsflax), Melitotus officinalis (yellow sweet-clover), Euphorbia esula (leafy spurge), Arnica spp and Eriogonum spp (wild buckwheat).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

SPS Measures

The affected western states of the USA maintain SPS measures with, for example, movement prohibitions, awareness raising and recommendations to prevent spread. The Utah Noxious Weed Act (Utah Department of Administrative Services, 2015) is an example of the procedures in place in the relevant states. The components of early warning, rapid response and public awareness are given by Pokorny and Kreuger-Mangold (2007), who described a control programme in Montana that effectively eliminated the plant as a pest species. Eradication is not feasible as the plant is too well established and widespread, but the plant can be managed.

Containment/Zoning

This is not applicable other than to improve the efficacy of other control measures. Mechanical pulling may be considered containment in the initial years, whereas if done repeatedly it becomes a successful control technique.

Control

A great many control measures are very effective, but logistical issues impact which are used, depending upon the nature of infestations, such as their occurrence in crops, on rangeland or in environmentally sensitive or difficult to reach areas. Consequently, an effective classical biological control would be valuable. As I. tinctoria is often found on or near marginal land, control measures may not be economic or there may be environmental considerations that will favour different control measures. Pokorny and Krueger-Mangold (2007) described a successful control project comprising of four key components –1) early detection, 2) treatment technologies including hand-pulling, digging and spot-spraying with metsulfuron plus nonionic surfactants and the removal of flowering or seeding plants in double-lined plastic bags, 3) repeated site visits with monitoring, and 4) education.

Sanitary Measures

I. tinctoria is a specified noxious weed in 11 western states of the USA and these states have sanitary methods to prevent dissemination within and between states; the requirements from Utah (Utah Department of Administrative Services, 2015) are an example. This requires persons, companies or corporations to clean transport and farm machinery free from all noxious weeds. Screenings for livestock feed, livestock feed itself, hay straw or similar, soil, sod and nursery stock and livestock are also listed with procedures to be followed.

Physical/Mechanical and Cultural Control

DiTomaso et al. (2013) gave recent advice on mechanical control. Hand pulling can be very effective if the crown is removed, which is easier after bolting but needs to be done before seed set. Repeat visits are necessary each season and for several years. Mowing, while it can reduce plant reserves and seed production if repeated, is not effective as there is resprouting from the crown. Clipping can be more effective at preventing resprouting, if close enough, but is ineffective if done too early in the season (West and Farah, 1989). Soil cultivation for field crops can be effective but, in alfalfa may require both spring and late fall/autumn cultivation (Jacobs and Pokorny, 2007). Sheep grazing is not very effective as the plant is not preferred and the required stocking densities would damage the rangeland (West and Farah, 1989). However, goat grazing has been recommended for noxious weeds (Lamming, 2001) and goats have been used effectively on I. tinctoria (L Malmberg, personal communication, 2015).

Movement Control

Within seriously affected states, movement of I. tinctoria is prevented. For example, in Utah it is unlawful to disseminate I. tinctoria and the articles of such dissemination are listed (Utah Department of Administrative Services, 2015,) and other states have similar legislation.

Biological Control

A plant pathogen, described as Puccinia thlaspeos (preferred name now Micropuccinia thlaspeos) was identified from I. tinctoria in Idaho in 1978. Phylogenetic studies of the rust showed that it is most probably native to North America (Kropp et al., 1997) and so was considered as a potential biological control agent. The systemic colonization may allow survival of the rust in root tissues over winter and after plant exposure to range fires (Kropp et al., 1996). Infection greatly reduces vegetative growth, reducing its competitive ability and also seed production; in most cases no seed is produced on infected plants, although those that are produced germinate as well as those from non-infected plants (Kropp et al., 1996).

A product, “Woad Warrior” based on I. tinctoria leaves infected with the rust was submitted for registration to the US Environmental Protection Agency (EPA) (Anon, 2002) who recommended registration. The EPA reviewed the evidence finding that there were no unreasonable adverse effect to humans or the environment and also noted that occasionally wild deer fed selectively on infected I. tinctoria plants, as did mice (Kough, 2002). Woad Warrier was available, in a limited fashion, for a few years but is no longer on the market (RE Whitesides, personal communication, 2015). The product was never commercially available due to a lack of a commercial backer, but after registration the fungus was spread by researchers (Bellgard, 2008). Woad Warrior was specified for use in Utah’s waterfowl management areas in 2009 (Berger, 2009), but this appears to be outdated advice. Although the rust has significant adverse effects on the plant and seed production it does not seem to be able to stop the spread of I. tinctoria (RE Whitesides, personal communication, 2015), although presumably it does slow the spread. Seedling mortality is normally extremely high (Farah et al., 1988) so reduced seed production may result in reduced seedling competition with a less than commensurate reduction in established plants.

A project to look for classical biological control agents in 2004 identified four species for detailed research, and further candidates were identified from extensive literature and field surveys (Cortat et al., 2008, Hinz et al., 2007). Of the original four species, the weevil Aulacobaris fallax and the flea beetle Psylliodes isatidis were found to be insufficiently specific (Cortat et al., 2008; Hinz et al., 2015) for consideration. As of 2018, the seed-feeding weevil Ceutorhynchus peyerimhoffi and the root crown-mining weevil Ceutorhynchus rusticus have been prioritized as promising agents.

The requirements for a successful biological control agent for I. tinctoria must take into account its biology. Farah et al. (1988) considered that an agent effective at the young rosette stage would be most effective as this stage is very vulnerable, whereas mature rosettes generally survive and reproduce. This could well be an argument to actively target the mature rosettes. The majority of seeds fail to establish, and markedly different seed production often results in similar levels of establishment, one reason why Puccinia thlaspeos has not been as successful as anticipated. The level of successful attack of the stems by P. thlaspeos appears low (H Hinz, CABI Switzerland, personal communication, 2015) but plants can be asymptomatic for many months, so infestation levels may be underestimated (Kropp et al., 1996).

Successful macrobial biological control agents are likely to share attributes promoting rapid population build up and phenotypic plasticity that are demonstrated by successful herbivorous invasive species (Nahrung and Swain, 2015). This may be of critical importance with I. tinctoria which also demonstrates marked plasticity and the ability to succeed under stress tolerant conditions.

Chemical Control

Metsulfuron or chlorsulfuron combined with 2,4-D and a nonionic surfactant are recommended as is the combination of imazic and methylated seed oil on rosettes or bolting plants (Jacobs and Pokorny, 2007). Berger (2009) described chemical treatments for Utah’s waterfowl management areas and lists dicamba plus 2,4-D, 2,4-D ester and amine and glyphosate. Other chemicals that can be appropriate for use in natural areas are given by Tu et al. (2001). The scale of I. tinctoria infestations in rangeland and difficulties of application in some areas combined with the cost:benefit aspect means that herbicide use is often impractical and/or uneconomic.

IPM

Control techniques can be used in various combinations for sensitive areas (Tu et al., 2001); Utah’s Waterfowl Management Areas’ plans for eradication involves herbicide treatments or biological control using Puccinia thlaspeos or a combined herbicide, mowing and sheep grazing strategy (Berger, 2009). Metsulfuron and chlorsulfuron did not adversely affect P. thlaspeos teliospore germination, whereas 2,4-D and some surfactants did (Jacobs and Pokorny, 2007).

Pokorny and Krueger-Mangold (2007) described a successful control project comprising of four key components –1) early detection, 2) treatment technologies including hand-pulling, digging and spot-spraying with metsulfuron plus nonionic surfactants and the removal of flowering or seeding plants in double-lined plastic bags, 3) repeated site visits with monitoring and 4) education.

Communication is an important component of IPM. A general strategic plan for managing noxious and invasive weeds in Utah is one example of IPM, covering education and research, mapping and monitoring, prevention, early detection and rapid response, control through integrated weed management and regulation and enforcement (Whitesides, 2004). A Noxious Weed Field Guide for Utah (Belliston et al., 2009) deals exclusively with Utah’s state noxious weeds, listing 27 weeds as noxious, and notes that individual counties may add county-listed weeds to the list. I. tinctoria is a Class B weed: “Control – Declared noxious weeds not native to the state that pose a threat to the state and should be considered a high priority for control”.  The Utah Noxious Weed Act (Utah Department of Administrative Services, 2015) designates the state noxious weeds (as above), groups ten articles as capable of disseminating noxious weeds (e.g. machinery, seed, manure and livestock) and gives prescribed treatments to prevent dissemination by these articles. In addition each county is required to submit an “Annual Progress Report of County Noxious Weed Control Program” to the Commissioner of Agriculture and Food and general and individual notices “pertaining to the control and prevention of noxious weeds” are specified. These are a) a General Notice to be posted in at least three public places within a county, stating the responsibility of every property owner to control and prevent the spread of noxious weeds on their land, b) an Individual Notice requiring a landowner to take (specified) actions to control weeds on the land and c) a Notification of Noxious Weed Lien Assessment whereby the Board of County Commissioners may carry out the control on land and require payment by the owner of that land.

Monitoring and Surveillance

Early detection is important; using Landsat-5 Thematic Mapper spectral data and field survey information Dewey et al (1991) determined specific land cover types that were associated with I. tinctoria infestation sites (highly suited types). Estimating the non-infested areas associated with the highly suited cover types gave an estimate of the potential for plant invasion. In a practical setting, the use of a Global Positioning System, with data subsequently recorded in the GPS data dictionary, assisted the implementation of a successful control programme that was largely achieved by manual removal and spot applications of herbicides (Pokorny and Kreuger-Mangold, 2007). Despite improved technologies for remote sensing (Huang and Asner, 2009; Naupari et al., 2013) this has not been reported as a major tool for I. tinctoria management.

Ecosystem Restoration

Competition from desired plants reduces I. tinctoria spread; reseeding after appropriate seed bed preparation is effective. On rough ground, fire (see Zouhar, 2009 for guidelines) or grazing may be used, followed by establishing competitive plants and maintaining grazing (Jacobs and Pokorny, 2007).

Determining which life stages to target for biological control is important. The merits of agents reducing seed production may be debated if this just reduces seedling competition (Farah et al., 1988), but seed persistence in the soil is limited, so constant seed attrition should have a positive control effect over time. Seed reduction may need to be accompanied by rosette control. Modelling of a related plant, garlic mustard (Allaria petiolata) suggests that reduced rosette survival and reduced seed output would control populations (Davis et al., 2006). 

22/01/18 Updated by:

Philip Weyl, CABI, Delémont, Switzerland 

17/06/2015 Original text by:

Dr. Dave Moore, CABI, Egham, UK