Can cultivars of invasive be safe? Examples from

Kayri Havens and Glen Madeja Chicago Botanic Garden and Tiffany Knight Washington University The Invasive Problem . Invasive are the 2nd largest threat to biodiversity . We depend on biodiversity • Food, Medicines • Shelter • Ecosystem services . Invasive species cost the US $138 billion/year, invasive costs are conservatively $35 billion/year The “Cultivar Issue”

• Cultivars can vary significantly in their fecundity • When a wild type and/or one or more cultivars are invasive, should all be banned? • Are “sterile” cultivars always sterile? Modeling effect of fecundity on population growth rate

. We took a demographic approach to ask, “When reducing fecundity of cultivars, how much is enough to create a plant that will not be invasive?” . Modeling was done by Tiffany Knight (Washington University) Japanese Barberry (Berberis thunbergii) A popular horticultural that has become a natural area invader

“The reasons? Sheer growability… hardy to zone 4, can take sun or shade, wet or dry, and will come back strong after a beating.”

“one of the most destructive invasive plants in Connecticut” --Connecticut Botanical Society Why the resistance? Japanese Barberry sales in Connecticut Alone --$15-20 million Annually-- Why the resistance? 35 Cultivars of Japanese Barberry • Aurea • Bonanza Gold • Angel Wings • Crimson Giant • Rosy Glow

Individuals within these range from producing less than 100 to over 3000 seeds per year

“I know that some Barberries have become invasive but mine doesn't really set any seed and my plants are only 2 feet high after almost 5 years in the ground.” A demographic approach

Cultivar #1 Cultivar #2

0.05 0.95 0.50 0.95

0.80 0.05 0.05 0.35 Seedling Juvenile Adult Seedling Juvenile Adult

1000 * 0.10 50 * 0.10

λ=1.50 λ=1.00 Plant invasions and matrix models

Plant Life form Invasive Range Selected Reference Dipsacus sylvestris Biennial USA Werner and Caswell 1978 Alliaria petiolata Biennial USA Davis et al. 2006 Carduus nutans Monocarpic herb New Zealand Shea and Kelly 1998 Heracleum mantegazzianum Monocarpic herb Europe Nehrbass et al. 2006 Cirsium vulgare Monocarpic herb USA Tenhumberg et al. 2007 Molinia caerulea Perennial grass Western Europe Jacquemyn et al. 2005 Agropyron cristatum Perennial grass USA Hansen and Wilson 2006 Perennial herb USA Schutzenhofer and Knight 2007 Centaurea maculosa Perennial herb USA Emery and Gross 2005 Clidemia hirta Shrub USA DeWalt 2006 Ardisia elliptica Shrub USA Koop and Horvitz 2005 Cytisus scoparius Shrub USA Parker 2000 Pinus nigra Tree New Zealand Buckley et al. 2005 Alliaria petiolata (Garlic mustard) A typical short-lived invasive plant

Seed Rosette Adult

Seed Rosette Adults Bank

Davis et al. 2006 Alliaria petiolata (Garlic mustard) A typical short-lived invasive plant

Seed Rosette Adult

0.27 .026 0.02 0.36 λ=1.42 0.03 0.62 Seed Rosette Adults Bank 0.09 0.45 .27 0.09 Davis et al. 2006 Reducing seed output reduces population growth of garlic mustard

2.5

) 2.0 λ

1.5 a a a a a a 1.0 a a a

0.5 a Population Population growth rate ( a

0 0 20 40 60 80 100

Percent reduction in seed production or viability Short-lived invasive species

2.5 d Species e d d

e a Alliaria petiolata ) 2.0 d λ e d ca Carduus nutans d ca e ci Cirsium vulgare ca d e cip ca d d Dipsacus sylvestris 1.5 s ci a p ci ca e e Echium plantagineum as p ci d a p caci e a p caci p Polygonum perfoliatum a d s pa ci e s Setaria faberi 1.0 cap ci a p e caa pci s a pe ca ci p a e

0.5 s ca Population Population growth rate ( da s ci s 0 s s s ca 0 20 40 60 80 100

Percent reduction in seed production or viability So, reduced seed output is likely to reduce population growth of short-lived species….

…..But, what about long-lived species such as and trees? Scotch Broom (Cytisus scoparius) Cytisus scoparius (Scotch Broom)

422

0.32 0.37 97.5 0.03 0.02 0.23 0.21 0.01 0.46 0.03 0.37 0.60 0.39 0.11 0.05 0.01 0.01 0.35 0.03

0.49

861

0.44 0.11 0.01 0.85 2725

λ=1.40 Parker 2000 Reducing seed output has minor effects on scotch broom

2.5

) 2.0 λ

1.5 cy cy cy cy cy cy cy cy cy cy 1.0 cy

0.5 Population Population growth rate (

0 0 20 40 60 80 100 Percent reduction in seed production or viability Long-lived invasive species

2.5 Species

a Ardesia elliptica

) 2.0 ce Centaurea maculosa λ p p p m m p p cl Clidemia hirta cl cl m m p p cl cl m p cl m m p cy Cytisus scoparius cl cl m p 1.5 cl m m Molina caerulea cy cy cy cy cl ce ce ce cy cy cy m p Pinus nigra ce ce cy cy cl ra ra ar ar ar arce r r r rcy cea a a a r r Rubus discolor 1.0 ce ce maclp ce cy ce

0.5 Population Population growth rate (

0 0 20 40 60 80 100 Percent reduction in seed production or viability Reducing seed output has larger effects on the growth rate of short-lived species

Short-lived species Long-lived species

2.5

2.5

) )

λ d e d λ d e d 2.0 e d 2.0 p p d m m p p p ca e cl cl m m p p ca d cl cl m p e cl m m p cips ca d cl cl m p 1.5 a pci ca e 1.5 cl m as pci ci d cy cy cy cl a p caci e ce ce cy cy cy m a p ci ce ce cy cy cl a cap e d ra ra ar ar arce arce ar r rcy rcy r s a capci ce cea a a p 1.0 a ci e 1.0 ce mcyacl caap ci ce s p e ce caa cip p a e

0.5 s ca 0.5 Population growthrate ( a Population growth rate ( s cid s 0 s s s ca 0 0 20 40 60 80 100 0 20 40 60 80 100

Percent reduction in seed production or viability Conclusions from models

Reductions in seed output will not necessarily result in non-invasive cultivars, particular for long-lived species

Demographic modeling can illuminate the invasive potential of different cultivars if you have the data -seed output, seed viability -life history (time to reproduction)

Knight, Havens and Vitt, 2011, BioScience 61: 816-822 Evaluation of Miscanthus sinensis cultivars

. A long lived perennial grass that forms large clumps (doesn’t run like M. sacchariflorus) . Known to invade in eastern U.S. . Fecundity will need to be zero (or extremely close to zero) in order for the cultivar to be unlikely to invade . Glen Madeja (NU grad student) evaluated 35 cultivars for fecundity in the Chicago region (Madeja et al., 2012, J Env Hort, in press) Miscanthus trials

Glen taking data >

Tiffany with ‘Morning Light’ > Quantifying seed set

. Randomly harvested 3 infloresences per individual . X-rayed inflorescences . 2 people independently counted filled seeds and averaged . Multiplied by number of inflorescences on individual . Averaged filled seed set over 4 individuals per cultivar Cultivar Average Filled Cultivar Average Filled Seed Set Seed Set Kleine Silberspinne 191,202 Ferner Osten 9,190

Rotsilber 179,957 Goliath 7,137 Autumn Light 157,936 Sarabande 3,278 Malepartus 106,172 Gracillimus 3,146 Blutenwunder 91,569 Little Zebra 1,359 Graziella 90,984 Superstripe 1,226 Nippon 81,024 Gold Bar 1,031 Huron Sunrise 77,925 Morning Light 968 Minuett 69,997 Strictus 907 Andante 61,103 Dixieland 785 Silberfeder 49,060 M. sacchariflorus 746, but runs!

Adagio 27,078 Variegatus 211 Puenktchen 19,376 Yaku-jima 138 Silberturm 19,133 Hinjo 0 Zebrinus 16,621 Silberpfeil 0 Purpurascens 13,390 M. x giganteus 0 Autumn Red 12,995 Cabaret 0, but died out Conclusions

. Three ornamental cultivars (‘Silberpfiel’, ‘Hinjo’ and ‘Cabaret’) of Miscanthus sinensis and the hybrid Miscanthus x. giganteus (used ornamentally and as a potential biofuel crop) have not set seed in Chicago and thus appear to have a low invasion risk. . ‘Cabaret’ was not reliably hardy, so was not recommended. . These results should be interpreted with caution because it is possible they may be able to set seed as the plants mature and/or as the climate warms. These cultivars may be able to mature seed in warmer regions of the US. One of our seedless cultivars (‘Hinjo’) did set seed in Zone 5 in another trial. Acknowledgements:

. Thanks to the staff of the Division of Plant Science and Conservation and collaborators on these projects who shared data and photos for this talk: Pati Vitt, Glen Madeja, Tiffany Knight, Lauren Umek.