The Case for Sasajiscymnus tsugae : Biological Control has Helped Save Connecticut’s Hemlocks

CAROLE CHEAH Connecticut Agricultural Experiment Station, Valley Laboratory, Windsor, CT

The Facts: Fig. 2 Hemlock Crown Recovery in Monitored 6-11 year S. tsugae Release Sites

•Connecticut’s hemlocks have been under adelgid attack 5. People's State Forest B: 2000 Release 6. Tunxis State Forest: 2001 Release 7. Enders State Forest: 1999 Release 8. Metacomet Trail, W. Suffield: 1999 Release 9. Bigelow Hollow State Park: 1999 Release 10. Mashamoquet Brook S. Park : 1998 Release 100 100 100 100 100 100 since the first discovery of HWA in 1985. From 1985-2000, 80 80 80 80 80 80

60 HWA spread to infest every town in CT (Fig.1); statewide 60 60 60 60 60 40 40 40 40 40 40

20 20 20 MeanFoliageTransparency 20 20 20 Mean FoliageTransparency Mean Mean Foliage Mean Transparency Foliage

decline of hemlocks was noted in 1994 Foliage Mean Transparency Mean Mean FoliageTransparency Mean Foliage Transparency FoliageMean

0 0 0 0 0 0 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 Year Year •There has been no chemical intervention on a forest wide Year Year Year Year 4. People's State Forest A: 1999 Release 11. Natchaug State Forest: 2001 Release scale to combat HWA in CT. Widespread large releases of 100 6 8 100 80 9 80 S. tsugae in the most heavily infested forests over most of 5 60 60 the state has been the main strategy to combat HWA with 40 4 7 40 20 20 Mean FoliageTransparency Mean Mean Foliage Transparency FoliageMean 10 >80% of the total released to date occurring from 1996- 0 3 0 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 Year 1 Year 2001. Over 176,000 S. tsugae have been released in 26 11 3. Burr Pond State Park: 1999 Release 12. Pachaug State Forest: 1999 Release

100 sites statewide from 1995-2007, but releases only began 100 80 80

60 after >90% of the state was infested (Fig. 1) 60

40 2 40 20 Mean FoliageTransparency Mean 20 MeanTransparency Foliage

0 1998 2000 2002 2004 2006 2008 13 0 1998 2000 2002 2004 2006 2008 Fig. 1 Chronology of HWA and S. tsugae in Connecticut Year 12 Year 3 120 200x10 13. Devil's Hopyard State Park: 1999 Release 2. Steep Rock Preserve, Washington: 1998 Release % Towns infested 180x10 3 100 14 100 S. tsugae 100 80 80 160x10 3 60 60 Released 140x10 3 80 40 40 120x10 3 20 20 Mean Foliage Mean Transparency Mean Mean FoliageTransparency 3 60 100x10 tsugae S. 0 0 1998 2000 2002 2004 2006 2008 1998 2000 2002 2004 2006 2008 80x10 3 Year Year

40 3 60x10 1. Nepaug Reservoir, MDC: 1996 Release 14. Salmon River State Forest : 2001 Release 40x10 3 100 100 20 80 80 20x10 3 % townsCT of infestedwith% HWA Baseline sites (Negligible HWA/EHS) (n = 120 trees) 60 60 0 0 Cumulative Total 40 Monitored S. tsugae release sites ( n = 287 trees) 40 1985 1990 1995 2000 2005 20 20 Mean FoliageTransparency Mean

Non-release site matches (n = 210 trees) Transparency Mean Foliage Year 0 0 1994 1996 1998 2000 2002 2004 2006 2008 Other S. tsugae release sites 1998 2000 2002 2004 2006 2008 The Evidence: Year Year Fig. 4. Hemlock Crown Improvement in S. tsugae Fig. 5. Comparisons of Release, Non-release Fig. 6 Pre- and Post-Release Crown Levels of HWA •Substantial crown recovery in many release sites was Release Sites 2003-2006 1 = < 10%; 2 = 11-50%; 3 = 51-75%; 4 = >75% HWA on Crown 100 and Baseline Sites 5 first recorded in 2005 and has continued annually (Fig. 100 Initial HWA Non-release infested sites HWA 06 80 Baseline (uninfested) sites 4 HWA 07 2 ) in spite of much warmer winters in 2006 and 2007 80 S. tsugae release sites

(Fig. 3). Declining hemlocks can significantly recover 60 3 in one year when precipitation is ample and drought 60 * a * 2 40 a a 40 stress is minimal (Fig. 4) * * b b b b b b b CrownLevelsof HWA •Hemlock recovery has occurred statewide in all types 20 20 1 Mean FoliageTransparency Mean of sites and soils for 14 monitored release sites in 0 0 0 Mean Foliage Release Site Transparency 0 1 2 3 4 5 6 7 8 9 101112131415 2002 2003 2004 2005 2006 2007 2008 2004 2005 2006 2007 2008 riparian to ridge top, ravine to level hemlock habitats S. tsugae Release Sites Year Year •Hemlock recovery has also occurred in southern sites *Means followed by different letters are highly significant at p < 0.0001 while means followed by the same letter are not significantly different where winter mortality of HWA has not recently been a •In 2005, non-release sites were matched climatically, topographically and 15 trees/site rated, as were significant factor in depressing adelgid populations baseline uninfested sites for comparison (all ratings used standard FIA crown health assessments) Fig. 3 HWA Winter Mortality in Connecticut 2000 - 2007 •Results showed that hemlocks in non-release sites had significantly thinner crowns than in release sites (p < 0.001). Crown health in release sites had recovered to similar levels in uninfested sites (p > Northern CT 15 0.05; Fig. 5) by 2005 through 2007. 100 Southern CT F) o 10 •HWA crown levels have also been reduced to the original levels prior to S. tsugae introduction in the 80 5 majority of release sites (Fig. 6) 0

60 -5 •In fact, eastern hemlocks statewide are generally showing healthier crowns; decline is patchy and

Fig. 3 HWA Winter Mortality in Connecticut-10 2000 - 2007 40 correlated with the intensity of concurrent elongate hemlock scale infestations -15 Min. Daily WinterMin. ( Temp. Daily r2 = 0.5104; p < 0.001 •This is extensive hemlock recovery that is post peak S. tsugae introduction. Prior to S. tsugae releases, % HWA Mortality Winter % 20 -20 Fig. 3 Winter Mortality of HWA in Connecticut0 20 40 2000 60 - 80 2007 100 120 no extensive recovery had been observed when there were also very cold winters to mitigate adelgid Mean % HWA Winter Mortality 0 spread (Fig. 7). Connecticut’s eastern hemlocks are currently healthier than in any prior period during 2000 2001 2002 2003 2004 2005 2006 2007 YEAR the establishment and spread of HWA with negligible hemlock mortality since 2001. If indeed, S. tsugae was as ineffective as is claimed, there would be no hemlock recovery to document. Perhaps Occam’s Fig. 7 Winter patterns and drought events in CT Razor holds true in this case…….. “All other things being equal, the simplest explanation is the best. ”

10 Additional Evidence: A sleeve cage experiment was also conducted in CT to 100 Fig. 8. Density-dependent field oviposition by S. tsugae 8 investigate recent reports that S. tsugae minimally oviposits in field bag St ovipositional response to HWA density in field sleeve cages 80 Mean weekly temperatures within sleeve cages 6 60 studies. Reproductive 3 month old laboratory reared S. tsugae were caged 35 60 50 4 in single pairs on hemlock branches with a range of late HWA sistens egg 30

CT Winter Rank Winter Rank CT 40

C 25

mass densities - developing progrediens from 5/23/06-7/14/06, when ambient 40 o 2 20 Reproduction 20

Severe / DroughtDuration Extreme (mo) / Severe temperatures were optimal. Sleeve cages (n = 15) were then harvested and 30 0 0 15 1989 1992 1995 1998 2001 2004 2007 counts of all live and dead adults, larvae and pupae were tallied for each S. tsugae 20 Year CT Winter Rank Temperature 10 Drought Duration (South) Data obtained from Weekly mean temp. Northeast Regional Climate Center female’s reproductive response. Results showed that 100% oviposition Drought Duration (Central) 10 5 Weekly min. temp. Drought duration (Northwest) occurred (mean 27 ± 14 eggs/female) and females showed a density- TotalNet r2 = 0.6175; p < 0.05 Weekly max. temp. 0 0 dependent oviposition response with a maximum of 56 eggs/female (Fig. 8). 0 100 200 300 400 500 600 0 1 2 3 4 5 6 7 8 Initial HWA density per sleeve cage Week ACKNOWLEDGMENTS: USDA FOREST SERVICE NORTHEASTERN AREA STATE & PRIVATE FORESTRY FOR FUNDING S UPPORT; Great Mountain Forest Corporation, Mt. Riga, Inc. , CT DEP Forestry Division, CT DEP State Parks, Metropolitan District Commission (Natural Resources), Camp Sequassen CYC