Research scientists in the USFS Northern Region (Region 9) – Maine to Minnesota, south to West Virginia and Missouri – continue to be concerned about regeneration patterns of the forest and the future productivity of northern hardwood forests.
The most recent study of which I am aware is that by Stern et al. (2023) [full citation at the end of this blog]. They sought to determine how four species often dominant in the Northeast (or at least in New England) might cope with climate change. Those four species are red maple (Acer rubrum), sugar maple (Acer saccharum), American beech (Fagus grandifolia), and yellow birch (Betula alleghaniensis). The study involved considerable effort: they examined tree ring data from 690 dominant and co-dominant trees on 45 plots at varying elevations across Vermont. The tree ring data allowed them to analyze each species’ response to several stressors over the 70-year period of 1945 to 2014.
In large part their findings agreed with those of studies carried out earlier, or at other locations. As expected, all four species grew robustly during the early decades, then plateaued – indicative of a maturing forest. All species responded positively to summer and winter moisture and negatively to higher summer temperatures. Stern et al. described the importance of moisture availability in non-growing seasons – i.e., winter – as more notable.
snow in Vermont; Putnypix via Flickr
The American Northeast and adjacent areas in Canada have already experienced an unprecedented increase of precipitation over the last several decades. This pattern is expected to continue or even increase under climate change projections. However, Stern et al. say this development is not as promising for tree growth as it first appears. The first caveat is that winter snow will increasingly be replaced by rain. The authors discuss the importance of the insulation of trees’ roots provided by snow cover. They suggest that this insulation might be particularly necessary for sugar maple.
The second caveat is that precipitation is not expected to increase in the summer; it might even decrease. Their data indicate that summer rainfall – during both the current and preceding years – has a significant impact on tree growth rates.
Stern et al. also found that the rapid rise in winter minimum temperatures was associated with slower growth by sugar maple, beech, and yellow birch, as well as red maple at lower elevations. Still, temperature had less influence than moisture metrics.
Stern et al. discuss specific responses of each species to changes in temperatures, moisture availability, and pollutant deposition. Of course, pollutant levels are decreasing in New England due to implementation of provisions of the Clean Air Act of 1990.
They conclude that red maple will probably continue to outcompete the other species.
In their paper, Stern et al. fill in some missing pieces about forests’ adaptation to the changing climate. I am disappointed, however, that these authors did not discuss the role of biotic stressors, i.e., “pests”.
They do report that growth rates of American beech increased in recent years despite the prevalence of beech bark disease. They note that others’ studies have also found an increase in radial growth for mature beech trees in neighboring New Hampshire, where beech bark disease is also rampant.
For more specific information on pests, we can turn to Ducey at al. – also published in 2023. These authors expected American beech to dominate the Bartlett Experimental Forest (in New Hampshire) despite two considerations that we might expect to suppress this growth. First, 70-90% of beech trees were diseased by 1950. Second, managers have made considerable efforts to suppress beech.
Stern et al. say specifically that their study design did not allow analysis of the impact of beech bark disease. I wonder at that decision since American beech is one of four species studied. More understandable, perhaps, is the absence of any mention of beech leaf disease. In 2014, the cutoff date for their growth analysis, beech leaf disease was known only in northeastern Ohio and perhaps a few counties in far western New York and Pennsylvania. Still, by the date of publication of their study, beech leaf disease was recognized as a serious disease established in southern New England.
counties where beech leaf disease has been confirmed
Eastern hemlock (Tsuga canadensis) and northern red oak (Quercus rubra) are described as common co-occurring dominant species in the plots analyzed by Stern et al. Although hemlock woolly adelgid has been killing trees in southern Vermont for years, Stern et al. did not discuss the possible impact of that pest on the forest’s regeneration trajectory. Nor did they assess the possible effects of oak wilt, which admittedly is farther away (in New York (& here) and in Ontario, Canada, west of Lake Erie).
In contrast, Ducey at al. (2023) did discuss link to blog 344 the probable impact of several non-native insects and diseases. In addition to beech bark disease, they addressed hemlock woolly adelgid, emerald ash borer, and beech leaf disease.
Non-native insects and pathogens are of increasing importance in our forests. To them, we can add overbrowsing by deer, proliferation of non-native plants, and spread of non-native earthworms. There is every reason to think the situation will only become more complex. I hope forest researchers will make a creative leap – incorporate the full range of factors affecting the future of US forests.
I understand that such a more integrated, holistic analysis might be beyond any individual scientist’s expertise or time, funding, and constraints of data availability and analysis. I hope, though, that teams of collaborators will compile an overview based on combining their research approaches. Such an overview would include human management actions, climate variables, established and looming pest infestations, etc. I hope, too, that these experts will extrapolate from their individual, site-specific findings to project region-wide effects.
Some studies have taken a more integrative approach. Reed, Bronson, et al. (2022) studied interactions of earthworm biomass and density with deer. Spicer et al. (2023) examined interactions of deer browsing and various vegetation management actions. Hoven et al. (2022) considered interactions of non-native shrubs, tree basal area, and forest moisture regimes.
Stern, R.L., P.G. Schaberg, S.A. Rayback, C.F. Hansen, P.F. Murakami, G.J. Hawley. 2023. Growth trends and environmental drivers of major tree species of the northern hardwood forest of eastern North America. J. For. Res. (2023) 34:37–50 https://doi.org/10.1007/s11676-022-01553-7
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at https://treeimprovement.tennessee.edu/
white flowered trillium (Trillium grandiflorum); via PICRYL One of the “charismatic wildflowers” mentioned by Blossey and colleagues
Bernd Blossey, Darragh Hare, and Don Waller have published a plea that America’s federal government take the lead in formulating a new national program on managing deer. Otherwise, they fear that deer populations will not be reduced to ecologically sustainable levels. I find their argument convincing and well-sourced. I agree that Americans need to figure out how to address this threat. (The full citation is at the end of this blog).
First, Blossey and colleagues describe the damage caused by overabundant deer:
severe declines in populations of many native forest herbs and shrubs, probably including disappearing wildflowers;
their replacement by non-native species that are less palatable;
poor regeneration of many canopy hardwood species;
decreased forest resilience, lowering forests’ ability to adapt to stressors, especially climate change;
decreased ability of forests to deliver benefits that are of increasing value to many people;
increased prevalence of wildlife and human diseases associated with the spread and size of growing tick populations; and
people – and deer — killed vehicle accidents on roads.
The widespread impacts of white-tailed deer (Odocoileus virginianus) in forests of the East are well-documented (see my previous blogs for a few examples; scroll below the “Archives” to find “Categories”). Blossey and colleagues note examples of similar impacts in the West, attributed to elk (Cervus elaphus) and black-tailed and mule deer (Odocoileus hemionus).
The authors review the decimation of deer populations in earlier centuries and the efforts of state wildlife agencies to rebuild their populations during the 20th Century. The problem, in their view, is that federal and — especially — state wildlife agencies have retained their traditional focus on managing wildlife for recreational hunters. However, recreational hunters make up a small and shrinking proportion of all Americans. Many more people now engage in “non-consumptive” enjoyment of wildlife.
lack of regeneration in Rock Creek Park, Washington D.C.; photo by Sam Sheline, NatureServe, via Flickr
State agencies’ narrow focus might partly arise from fragmented authorities. Agencies other than wildlife departments are responsible for addressing some repercussions of overabundant deer. These include threats to human health, loss of agricultural crops.
For several reasons, Blossey and colleagues call for federal leadership. They think that only a national strategy can address, in a holistic way, the interrelated deer, human health, forest, and biodiversity crises. The strategy’s goal should be to protect species that are in decline because of over-browsing by deer and to avoid further declines in environmental and human health.
The authors reason that states are tied to traditional constituencies. Also, they have difficulty acting across jurisdictional boundaries. Second, few state wildlife agencies have authority to protect plant and invertebrate species. Yet these are the taxa most directly affected by overabundant deer. Blossey and colleagues point out that, of the ~1,300 species listed under the federal Endangered Species Act, 942 are plants and 287 are invertebrates.
They point out that deer also suffer the effects of overpopulation. Chronic wasting disease is spreading. It causes a slow, agonizing death of affected animals. Another 2.1 million deer are killed each year in vehicle crashes. [According to the World Animal Foundation, the current number of deer killed in traffic crashes is 1.8 million — 300,000 fewer.) Again, these deaths are often gruesome. Finally, the principal population “control” now is death by starvation in winter. This, too, is cruel.
Blossey and colleagues say that return of large predators, even where feasible, will not result in sufficient reduction in deer populations. Nor will encouragement of greater hunting pressure on does.
They note that the federal government owns nearly 30% of the United States’ terrestrial surface area. Management is divided among many agencies – National Park Service, Fish and Wildlife Service, Bureau of Land Management, USDA Forest Service, Department of Defense, and many smaller agencies. Management approaches vary. However, it would be possible to bring them into agreement – although, in some cases, this would require new legislation.
Another issue requires resolution: federal agencies’ authority to manage wildlife on federal land.. The states have repeatedly claimed constitutional and legal authority to manage (vertebrate) wildlife on the federal lands within their borders. This assertion was countered years ago by Nie et al. (2017):
‘Federal land management agencies have an obligation, not just the discretion, to manage and conserve fish and wildlife on federal lands. … [M]ost states have not addressed the conservation obligations inherent in trust management; rather, states wish to use the notion of sovereign ownership as … a source of unilateral power but not of public responsibility. Furthermore, the states’ trust responsibilities for wildlife are subordinate to the federal government’s statutory and trust obligations over federal lands and their integral resources.’
Blossey and colleagues assert that managing wildlife (typically defined as mammals, birds, and fish) is much broader than establishing hunting seasons or methods. Furthermore, the concept of “public trust resources” means resources should be managed for all citizens, not just the fewer than 10% of US residents who hunt. A growing proportion of society expects this management to support healthy and diverse environments.
The authors stress that reducing deer overpopulations is necessary to meet numerous policy goals. These include fulfilling obligations under international treaties related to climate change, invasive species, and threatened species; restoring and conserving the nation’s forests to provide habitat; and adopting “nature-based” climate adaptations, such as carbon sequestration. They express the hope that recent presidential mandates to better quantify and value natural assets will increase awareness of the harm caused by deer overpopulation. Their proposed national strategy would develop goals and metrics to match specific environmental and human health outcomes.
Of course, management of deer must extend beyond federal property lines. This will require cooperation among states, Tribes, and private landowners.
The paper proposes the North American Waterfowl Management Plan as a model. Under this scheme the US Fish and Wildlife Service works with states, tribal governments, Mexico, and Canada to ensure accurate information on waterfowl populations a to calculate harvest levels. States implement their assigned quotas through their own regulations. Waterfowl hunters purchase Duck Stamps to fund the monitoring efforts. This program has worked well for most species covered by the program. Waterfowl are one of the few bird groups that have not declined dramatically.
Reducing deer populations will probably require lethal control. Studies indicate that at least 60% of does must be removed from a population to reduce herd sizes over time. Other means have been attempted at regional or larger landscape levels, such as sterilization, fertility control. These methods have failed even when paired with recreational hunting. Lethal approaches will probably distress many people. However, Blossey, Hare, and Waller believe the program would be supported if it is understood to be undertaken with the goal of improving the health of both humans and also the environment.
In the end, Blossey, Hare, and Waller say they are not willing to leave the killing to cars, disease, and starvation. They emphasize our moral responsibility to protect humans and the many other species that rely on diverse ecosystems. Our policies and choices created the problem, so we must try to correct it.
SOURCES
Blossey. B., D. Hare, and D.M. Waller, 2024. Where have all the flowers gone? A call for federal leadership in deer management in the US. Front. Conserv. Sci. 5:1382132. doi: 10.3389/fcosc.2024.1382132
Nie, M., C. Barns, J. Haber, J. Joly, K. Pitt and S. Zellmer. 2017. Fish and Wildlife Management on Federal Lands: Debunking State Supremacy. Environmental Law, Vol. 47, no. 4 (2017).
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
I have recently recent blogged several times about threats to regeneration of eastern forests. Most of the underlying studies stress the role of deer browsing as a major driver of suppression of native plant species and proliferation of non-native ones. Most studies discussed at a recent Northern Hardwood research forum (USDA, FS 2023b Proceedings) found that deer browsing overwhelms other disturbances, such as fire and canopy gaps that typically promote seedling diversity. Miller et al. also stressed the importance of the deer-invasive plant complex in interrupting regeneration in National parks. Reed et al. found that, on the Allegheny Plateau of western Pennsylvania, high deer densities at the time stands formed reduced tree species diversity, density, and basal area – changes that were still detectable decades later.
On the other hand, Hovena et al. found that at their study sites in Ohio, interaction between non-native shrubs and soil wetness overshadowed even the impact of deer herbivory on the species richness and abundance of seedlings.
Unlike the others, Ducey et al. don’t mention deer as a factor in their analysis of regeneration in a forest in the northern half of New Hampshire. They focus on the minimal impact of silvicultural management. Its effect is secondary to overall forest development as the forest ages. Is it possible that overabundant deer are not a factor in the Bartlett Experimental forest.
American elm (Ulmus americana); photo by F.T. Campbell
Some of the studies acknowledge the impacts of non-native insects and pathogens. The most thorough discussion is in Payne and Peet. They note that specialist pathogens have caused the loss of important tree species, specifically elms and dogwoods plus the impending widespread mortality of ash. Such mortality is resulting in drastic and long-lasting shifts in community dynamics.
Ducey et al. anticipate pest-driven reversals of increases over the decades of eastern hemlock (Tsuga canadensis) and American beech (Fagus grandifolia). Also, they expect that white ash (Fraxinus americana), which has a minor presence, will disappear.
Miller et al. also stressed the importance of emerald ash borer-induced suppression of ash regeneration in some National parks . The authors also noted the threat to beech trees from beech leaf disease in other parks. Hovena et al. state that the interaction between non-native shrubs and soil wetness was more influential than ash mortality in shaping woody seedling communities.
Reed et al. considered the role of non-native earthworm biomass on plant species’ growth.
But too many of the studies, in my view, make no mention of the probably significant role of non-native insects and pathogens.
It is perhaps understandable that they don’t address emerging pests that either have not yet or have barely reached their study sites. For example, Hovena et al. and Yacucci et al. [see below] noted growth of one native shrub, Lindera benzoin, in the face of the challenges presented by deer and invading plants. Neither acknowledges the approach of laurel wilt disease, which has not yet become established in Ohio (it has been detected on the Kentucky-Indiana border). Similarly, neither mentions beech leaf disease, although some of the plots studied by Hovena et al. are just east of Cleveland – where BLD was first detected. The location of the Yacucci et al. study is less than 50 miles away. The North Carolina forests studied by Payne and Peet are apparently too far east to be affected by beech bark disease and beech leaf disease is not yet established nearby.
Less understandable is the failure to mention loss of elms – which were abundant in riparian areas until killed off by Dutch elm disease – which was first detected in Cleveland!); or to discuss the impact of dogwood anthracnose. Their focus on the deciduous forest might explain why they don’t mention hemlock woolly adelgid – which is just now invading the area discussed by Reed et al. I suppose the demise of American chestnut was so many decades ago that it is truly irrelevant to current forest dynamics.
A new study raises anew these questions about whether inattention to the role of non-native pests has skewed past studies’ results. Yacucci et al. compared regeneration in a military installation (Camp Garfield), to the results in the surrounding second-growth forest. This choice allowed them to overcome one drawback of other studies: using deer exclosures that are small and of short durations. The military facility covers 88 km2. Inside it, deer populations have been controlled for 67 years at a density of 6.6 – 7.5 deer/km2. Outside, deer have been overabundant for decades. Populations have grown to densities estimated (but not measured) to be at least 30 deer/km2 – more than four times as high.
These authors established 21 experimental gaps in the low-deer-density area and 20 gaps outside the installation where deer densities are high. Some of the gaps in both low- and high-deer-density environs were located on wetter, seasonally flooded soils, some on drier sites. None of the forest sites had experience fire in recent decades.
Their findings support the importance of deer browsing as driver of changes to forest regeneration.
northern spicebush (Lindera benzoin); photo by R.A. Nonemacher via Wikimedia
They found that at low deer densities, gaps develop a vigorous and diverse native sapling layer, including oaks. Total stem density of red and pin oaks was 13 times higher in these gaps than in gaps in high-deer-density locations. Oak saplings were growing into the subcanopy – that is, above deer browse heights. Saplings of other species – i.e., tuliptree (Liriodendron tulipifera), red maple, and ash (Fraxinus spp.) were also flourishing. Also present were dogwood (Cornus florida) and two native shrubs — Lindera benzoin and Rubus allegheniensis. One non-native shrub, buckthorn (Rhamnus frangula), also thrived at low deer densities. Other non-native plant species were far fewer; their cover was 80% lower. Overall, abundance, richness, and diversity of native herbaceous and woody species were 37–65% higher at the low-deer-density study sites. On average tree species were more than twice as tall as in high-deer-density plots.
In high-deer-density plots, non-native species were six times more abundant while native species richness was 39% lower. Diversity was 27% lower. Most native tree species were short in stature and in low abundance. The one exception was black cherry (Prunus serotina), which deer avoid feeding on. The cherry was 95% more abundant in these high-deer-density plots.
There were several surprising results. In most cases, neither years since gap formation nor habitat type (wet vs. dry) had a significant impact on plant diversity, richness, or abundance. The exception was that non-native plant species were more abundant in older gaps where deer densities were high. Yacucci et al. warn that this phenomenon is a potential threat to biodiversity since high deer density is now the norm across eastern forests.
The authors also note that fire has probably never been a factor in these forests, which are primarily beech-maple forests. Certainly there have been no fires over the past 70 years, either inside or outside the military installation.
Yacucci et al. did not discuss past or possible future impacts of non-native insects or pathogens. They did not mention emerald ash borer or dogwood anthracnose – both of which had been present in Ohio for at least two decades when they completed their study. Although they said their study forest was a beech-maple forest, they did not discuss whether beech are present and – if so – the impact of beech bark disease or beech leaf disease. Both of these are spreading in Ohio. The latter was originally detected in 2012 near Cleveland, just 50 miles from the location of Camp Garfield (between Youngstown and Cleveland, Ohio). As noted above, they also did they mention that Lindera benzoin is susceptible to laurel wilt disease.
beech seedlings in Virginia; photo by F.T. Campbell
Proposed solutions to deer over-browsing
Given the combined threat from widespread deer overpopulation and invasions by non-native plants, Yacucci et al. propose enlisting those military posts that regularly cull deer into efforts to conserve and regenerate native plants. Otherwise, they say, the prognosis for regeneration is poor.
Bernd Blossey and colleagues propose a more sweeping solution: implementation of a national policy to reduce deer populations on all land ownerships. They point out that overabundant deer:
disrupt the plant communities of affected forests – from spring ephemerals to tree regeneration;
promote disease in wildlife and people; and
lead to miserable deaths of deer on our highways, through winter starvation, and disease.
They call for federal leadership of coordinated deer-reduction programs. I discuss their proposal in detail in a separate blog.
SOURCES
Ducey, M.J, O.L. Fraser, M. Yamasaki, E.P. Belair, W.B. Leak. 2023. Eight decades of compositional change in a managed northern hardwood landscape. Forest Ecosystems 10 (2023) 100121
Hovena, B.M., K.S. Knight, V.E. Peters, and D.L Gorchov. 2022. Woody seedling community responses to deer herbivory, intro shrubs, and ash mortality depend on canopy competition and site wetness. Forest Ecology and Management. 523 (2022) 120488
Payne, C.J. and R.K. Peet. 2023. Revisiting the model system for forest succession: Eighty years of resampling Piedmont forests reveals need for an improved suite of indicators of successional change. Ecological Indicators 154 (2023) 110679
Miller, K.M., S.J. Perles, J.P. Schmit, E.R. Matthews, and M.R. Marshall. 2023. Overabundant deer and invasive plants drive widespread regeneration debt in eastern United States national parks. Ecological Applications. 2023;33:e2837. https://onlinelibrary.wiley.com/r/eap Open Access
Reed, S.P., D.R. Bronson, J.A. Forrester, L.M. Prudent, A.M. Yang, A.M. Yantes, P.B. Reich, and L.E. Frelich. 2023. Linked disturbance in the temperate forest: Earthworms, deer, and canopy gaps. Ecology. 2023;104:e4040. https://onlinelibrary.wiley.com/r/ecy
United States Department of Agriculture, Forest Service. 2023a. Proceedings of the First Biennial Northern Hardwood Conference 2021: Bridging Science and Management for the Future. Northern Research Station General Technical Report NRS-P-211 May 2023
Yacucci, A.C., W.P. Carson, J.C. Martineau, C.D. Burns, B.P. Riley, A.A. Royo, T.P. Diggins, I.J. Renne. 2023. Native tree species prosper while exotics falter during gap-phase regeneration, but only where deer densities are near historical levels New Forests https://doi.org/10.1007/s11056-023-10022-w
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
Gettysburg battlefield; now under attack by emerald ash borer (see below)
Kathryn Miller and colleagues (full citation at end of blog) have published a study that examined the status and trends of forest regeneration in 39 National parks from Virginia to Maine. Four-fifths of the forest plots in the study are classified as mature or late successional – so at first glance the forests look healthy. However, the researchers made an alarming finding: in 27 of 39 parks, forest regeneration is failing – either imminently or probably. Acadia National Park is an exception; it is the only park in the study experiencing healthy regeneration. They warn that without intense, sustained – and expensive! – intervention, these forests are likely to be converted to other types of ecosystems. [I blogged recently about findings regarding regeneration in eastern forests: here and here and here and here.
The forests’ understories have too few seedlings and – especially – saplings to maintain themselves. Worse, in many cases the seedlings and saplings are not the same species as the mature trees that form the canopy. The saplings are shorter species that never reach the canopy. That is, species like pawpaw (Asimina triloba), American holly (Ilex opaca), American hornbeam (Carpinus caroliniana), and eastern redbud (Cercis canadensis) are regenerating, rather than the oaks (Quercus spp.), hickories (Carya spp.), maples (Acer spp.), and pines (Pinus spp.) that constitute the canopies of mature forests in these parks.
Miller and colleagues call these “regeneration mismatches.” Inabout half of the parks, these native canopy tree species make up less than half of current saplings and seedlings. This situation suggests the forests’ species composition will shift substantially, thereby undermining resilience in the face of other challenges, such as invasive plants and pests and climate change.
In many of these National parks, Miller and colleagues found abundant ash regeneration. For example, ash (Fraxinus spp.) constitute more than half of all seedlings in four parks (Johnstown Flood and Friendship Hill in Pennsylvania; Catoctin Mountain in Maryland; Manassas Battlefield in Virginia). Miller and colleagues consigned ash species to the “subcanopy class” because the emerald ash borer (EAB) has caused such high mortality of mature trees. They think regard it unlikely that current and future seedlings will ever reach full size. The devastating impact is most starkly illustrated in Gettysburg National Battlefield Park. Consistent deer management since 1996 has been rewarded: the Park ranks at the top for regeneration among the 39 parks. However, more than half of the seedlings and a quarter of the saplings are ashes. EAB has shifted the Park’s otherwise secure regeneration status into probable failure.
When regeneration fails: too many deer
Throughout the study region, the overwhelming reason regeneration fails is browsing by overabundant deer. The level of deer browse is considered “acceptable” in only four parks. Deer suppress the number of seedlings and saplings. They also skew species composition of native subcanopy species toward those less palatable. Miller and colleagues found that canopy tree density and cover and past human land use had minimal impacts on seedling and sapling numbers or species composition.
Overabundant deer also promote invasion and spread of non-native plants, which are the second most important factor impeding regeneration. Together, invasive plants and non-native earthworms are ecosystem engineers that negatively impact soil and cause cascades of biotic and abiotic impacts throughout forest ecosystems.
Many of the parks experiencing the most severe impacts of chronic deer browse also have the highest invasions by non-native plants. A natural process of regeneration occurs when the death or collapse of mature trees create gaps in the forest canopy. Where deer and invasive shrubs overlap, this process is often hijacked. Instead of nearby native tree species accelerating their growth toward the canopy, thickets of invasive shrubs crowd the space.
For this reason, Miller and colleagues recommend that park management prioritize treating invasive plants in canopy gaps of disturbed stands to avoid forest loss. They recommend deliberate creation of canopy gaps to promote resilience only for parks, or stands within parks, that have low deer and invasive plant abundance or the capacity to intensively manage invasive plants in gaps.
In most parks, non-native tree species are rare, less than 2% of total regeneration. In seven parks, though, non-native trees exceed ten percent of seedlings and/or saplings. In three parks, saplings of non-native trees are increasing. These are primarily tree-of-heaven (Ailanthus altissima) and Norway maple (Acer platanoides). In Saratoga National Historical Park, seedlings of common buckthorn (Rhamnus cathartica) are increasing.
Beech regeneration in Prince William Forest Park
Role of other pests
Miller and colleagues express fear that beech bark disease andbeech leaf disease might have effects similar to those of EAB, leading to a greater “regeneration debt” in parks where American beech (Fagus grandifolia) is the dominant regeneration component. They cite specifically Prince William Forest Park in northern Virginia, [25 mi2]Rock Creek Park in the District of Columbia, [2.7mi2] and Saratoga National Historical Park. [5.3 mi2] The authors also suggest that thickets of beech root sprouts formed in response to BBD can suppress regeneration of other native canopy species and so might need to be managed.
Miller and colleagues mention hemlock woolly adelgid (HWA), but provide very little information. They report that Saint-Gaudens National Historical Park in New Hampshire (the home and studio of sculptor Augustus Saint-Gaudens) is at particular risk because of growth of both beech and eastern hemlock (Tsuga canadensis). I know that Delaware Water Gap National Recreation Area [109m2] has experienced major losses of mature hemlocks. [Shenandoah National Park has also, but it was not included in the study.]
Hemlock Ravine, Delaware Water Gap National Recreation Area; photo by Nicholas T via Flickr
Miller and colleagues report that Acadia National Park is seeing recovery of red spruce (Picea rubens) from a major fire in 1947 and possibly also from acid rain. They do not mention the longer-term threat from the brown spruce longhorned beetle. Their focus is on forest dynamics largely unaffected by deer.
In the same way, the authors make no mention of the absence of dogwood trees, presumably because they had been eliminated by dogwood anthracnose decades ago. Nor do they mention vascular streak dieback of redbud; the causal agent still uncertain. [See Annie Self’s presentation to National Plant Board, August 2023.]
dead ash tree in Shenandoah National Park
One omission is large enough that it might affect the study’s findings. At mi2Shenandoah is the largest National Park in the region. It was not included in the study because the Park’s forest monitoring process is not compatible with those in other NPS units. All the other parks – including Acadia (562 mi) – are much smaller, protecting historic sites like Civil War battlefields.
RECOMMENDATIONS
Miller and colleagues recommend that deer management be initiated in parks classified as at imminent or probable regeneration failure, if such programs are not already under way. They warn that effective deer management requires sustained commitment. Studies of deer exclosures show that full forest recovery from chronic deer overabundance can take as long as 40–70 years.
The authors also recommend actions to open the subcanopy to facilitate growth of saplings belonging to desired species. They caution that deer predation must be controlled. Furthermore, either invasive plant cover must be low, or management must ensure that that the park has sufficient resources to sustain an invasive plant control program – especially if invasive plants are combined with abundant deer.
Parks experiencing compositional mismatches and that are dominated by oak–hickory forest types might also benefit from prescribed burning. Again, deer browse pressure must be minimized. In addition, regeneration of oaks and hickories must already be present.
In park forests dominated by species vulnerable to lethal pests, e.g., beech-, ash-, or hemlock-dominated forest stands, Miller and colleagues recommend considering planting alternative native canopy species and protecting those plantings from deer. Park managers should also consider thinning beech thickets formed after beech bark disease kills canopy trees.
Media coverage
The Washington, D.C., public radio station, WAMU, reported on this research on the air (broadcast December 20) and on its website. It is written by Jacob Fenston, with great photographs by Tyrone Turner. The story emphasized the link between deer and invasive plants – since regeneration in eastern deciduous forest happens by saplings taking advantage of gaps formed when mature trees die. The story quotes DC-area people on their efforts to contain vines. The Natural Resource Manager at Catoctin Mountain Park [8 mi2] describes that park’s longstanding deer control program. The story also mentions impacts of EAB and threat of BLD.
News – Funding for these parks to counter the threats!
Lead author Kathryn Miller has informed me that the Bipartisan Infrastructure Law and Inflation Reduction Act has provided the 39 parks involved in this study over $10 million to improve forest resilience largely through reduction of invasive plants and overabundant deer.
Of course, invasive species threats to National parks are not limited to the Northeast – nor are they new. I have raised this problem from the beginning. To see these blogs, on the “nivemnic” website, scroll down below the archives to the “categories”, then click on “national parks”.
SOURCE
Miller, K.M., S.J. Perles, J.P. Schmit, E.R. Matthews, M.R. Marshall. 2023. Overabundant deer and invasive plants drive widespread regeneration debt in eastern United States national parks. Ecological Applications. 2023;33:e2837. https://onlinelibrary.wiley.com/r/eap Open Access
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
hardwood regeneration in northern Virginia forest; photo F.T. Campbell
I have posted several blogs recently about tree species’ regeneration. One blog found poor regeneration of many species throughout forests of the eastern United States. Regeneration is particularly poor in the Great Lakes region, western New York and Pennsylvania, along the Mid-Atlantic and New England coasts, and the coastal plain from southern South Carolina to eastern Texas.
A second blog focused on forest succession in New Hampshire. These findings, by Ducey et al., explicitly recognized the impact of non-native tree-killing insects and pathogens. A third article (Payne and Peet, 2023; full citation at the end of this blog) reports similar findings in North Carolina – and explicitly says that the same conditions are found in forests across the eastern United States.
The locations of neither in-depth study – New Hampshire or North Carolina – include those identified by Potter and Riitters (2022) as suffering particularly poor regeneration.
Payne and Peet find that forest succession in the Piedmont region of North Carolina is not proceeding as expected, based on earlier studies conducted in the same region. The differences are apparent at both the canopy and understory levels. Especially notable is the low recruitment of oaks (Quercus species) and hickories (Carya species) – the genera which previous studies indicated would be the climax taxa. One explanation is the disappearance since early in the 20th Century of fire as a driver of disturbance.
The understory communities are also novel, due largely to invasive species: dramatic loss of flowering dogwood (Cornus florida) killed by the non-native pathogen dogwood anthracnose (Discula destructiva), plus overcrowding of the shrub level by invasive plant species. Other drivers are probably suppression of growth of woody species caused by excessive deer herbivory, and overall accelerated shifts in successional trajectory due to hurricane damage.
flowering dogwood autumn display; F.T. Campbell
Forests in eastern North America in the 21st Century face several drivers of change that are either novel or greatly heightened. In addition to the disappearance of chronic fire, these are frequency and timing of hurricanes, feeding by herbivore populations, and introduction of non-native tree-killing pests and plants. Payne and Peet say scientists and managers need to consider these additional drivers – and their interactions! – when anticipating successional change.
Like Ducey et al. in New Hampshire, Payne and Peet used 80 years of data from 33 permanent plots established and 55 years of data from another 3 plots. Twenty-eight of the plots are transitioning from loblolly pine (Pinus taeda) to hardwood dominance; eight plots have been mixed-age hardwood stands since before the study plots were established.
In the North Carolina piedmont, the composition of canopy trees in plots evolving from pine compared to hardwood stands continue to be different 90–120 years after succession began. Canopy trees in upland and bottomland hardwood stands also differ. These differences reflect the relative species in the forest at the initiation of succession dynamics. Hurricanes – especially Hurricane Fran in 1996 – apparently accelerated succession in some plots by toppling the oldest pines. Despite the persistent differences, the species compositions of both canopy and subcanopy layers are trending toward increasing similarity.
deer-damaged red maple; photo by Eli Sagor via Flickr
The impact of deer browsing is complicated. Deer populations in the study area quadrupled after measurement began in 1980. Deer herbivory suppressed growth of all plant species when their stems were thin (3 – 10 cm DBH). However, after 1996 rapid growth of plants in openings caused by Hurricane Fran’s passage began to reverse the effects of deer browsing. Also, while deer browsing decreases regeneration, growth, and abundance of oak and hickory seedlings and saplings, it also decreases the abundance of other tree species that have – nevertheless – increased in abundance, e.g., red maple (A. rubrum) and black cherry(Prunus serotina).
Payne and Peet found that soil attributes (wetness, texture, organic matter and chemical components), as well as topographic position were minor factors in determining succession trajectories. Increased light availabilitydue to the new or exacerbated drivers of change (thinning of understory vegetation by disease and deer herbivory and opening of the canopy by hurricanes) overcame the influence of nutrients. At most, aunique soil condition might constraining the impacts of these disturbances. Furthermore, these soil-related conditions and other environmental variables change through time — and as a result so does the vegetation. Specifically, the conditions that once supported establishment of oaks and hickories apparently differ today. Payne and Peet conclude that other drivers might be continuing to impact these species’ maturation.
A partial exception is soil nitrogen, through its influence on mycorrhizal patterns. I review mycorrhizal patterns in the discussion of individual tree species, below.
How are Individual Tree Species Responding?
Oaks and hickories are not expanding as expected – either as canopy-sized trees or as seedlings / saplings in the understory. Payne and Peet agree that century-long suppression of low-intensity ground fires is probably the most significant factor in this compositional shift. This decline has been exacerbated by selective logging and deer herbivory. Hickories have established more widely, possibly because young stems have greater shade tolerance. Only plots located on sandy and acidic soils and plots with the greatest hurricane damage have moderate recruitment of oaks and hickories. Oaks and hickories on the poor soils might be aided by the types of ectomycorrhizal fungi that survive in acidic soils with relatively low nitrogen levels. In addition, these soils’ lower water retention probably impedes competition by more mesic, faster-growing, shade-tolerant species. However, even oaks and hickories that have established as seedlings or saplings only rarely progress to canopy dominance. Payne and Peet conclude that oaks might have lost competitive advantage in many of the undisturbed stands.
More mesophytic hardwoods, especially red maple(Acer rubrum), are becoming more numerous and larger – a trend seen throughout forests of the eastern United States. Damage from Hurricane Fran apparently accelerated this trend. However, red maple growth is significantly inhibited by competition from thicket-forming shrubs, especially in bottomland plots. The invasive non-native species thorny olive or oleasterElaeagnus pungensincreased dramatically following Hurricane Fran in 1996. The situation is likely to worsen: two other invasive species, Amur honeysuckle Lonicera maackii and privet Ligustrum japonicum were first detected in the Duke Forest plots in the 2013 survey.
[In New Hampshire, Ducey et al. detected an unexpected levelling off of red maple increases and decline in sugar maple (Acer saccharum); they were unable to determine a cause.]
beech-dominated understory in northern Virginia; F.T. Campbell
Another mesophytic hardwood – American beech(Fagus grandifolia) – has become very abundant in bottomland hardwood stands, especially in small-stem size classes in the understory. Beech prefers sandy soils and its ectomycorrhizal associations are apparently more tolerant of more acidic soils.
Payne and Peet mention – briefly and vaguely – uncertainty about the future of beech. The reference cited discusses the impact of beech bark disease (BBD) in the northeast. Range maps indicate that BBD is well established in the southern Appalachians along the North Carolina/Tennessee border; it has apparently not spread as far east as the study area. There is no mention of beech leaf disease (BLD), which is the primary threat to seedlings and saplings. BLD is currently known to be in northern Virginia. It is unknown whether the disease has any climatic or other barrier that would prevent its moving farther south.
Another bottomland indicator taxon that is also increasing in abundance is ash (Fraxinus species). Along with sweetgum (Liquidambar styraciflua), tulip poplar (Liriodendron tulipifera) and black cherry Prunus serotina, ash density and basal area increased dramatically in plots heavily damaged by Hurricane Fran. Payne and Peet expect most ash trees to be killed by emerald ash borer (Agrilus planipennis) by 2022. The beetle was detected in the study area in 2015.
ash killed by EAB on Potomac lowlands; F.T. Campbell
Flowering dogwood (Cornus florida)was one of the most abundant understory species throughout the study area until the late 1980s. The species has declined by more than 80% since then due to the non-native disease dogwood anthracnose (Discula destructiva). No other species has experienced as precipitous a decline. There is now almost no regeneration in most upland sites.
A second species almost eradicated from the study area by a non-native pathogen is American elm (Ulmus americana). Its basal area in 2013 was 5% of peak levels in the 1950s. Most of this loss occurred by the 1960s, shortly after arrived of Dutch elm disease (DED) in North Carolina. A congeneric species, slippery elm U. alata, is reported to beabundant; it is somewhat resistant to DED. There is no mention of the zig-zag sawfly (Aproceros leucopoda) which has been detected in North Carolina, a few counties away from the study area. The foliage-feeding insect’s long-term impact on elm species is not yet understood.
Payne and Peet note that the study area has twice experienced loss of important components due to specialist non-native pathogens: elms and dogwoods. A third similar event looms: ash [The article does not discuss prospects for biological control.] A fourth is less certain: beech. [This numbering assumes that American chestnut and eastern hemlock were not significant components of forests in the study area.] In their view, these events demonstrate the drastic impacts such non-native organisms can have, especially when the host species is highly abundant or otherwise dominant in a specific community. The resulting shifts in community dynamics and modifications to light and water availability due to such losses, can be dramatic and long-lasting, even resulting in novel successional trajectories.
Members of the 23rd Civil Engineer Squadron/23rd Wing chainsaw a tree lying across a street in the NCO housing area- damage to piedmont North Carolina by Hurricane Fran. Photo courtesy of U.S. National Archives.
Payne and Peet also emphasize the impact of large, episodic disturbances (in their case, hurricanes). These can have widespread and long-lasting impacts on plant community dynamics. Hurricanes’ frequency, intensity, and timing relative to successional stage are key in determining their impacts on successional trajectories. E.g., strong storms that felled the even-aged pine canopy accelerated succession toward more mixed hardwoods. These changes affect biomass, diversity, competitive dynamics, and invasion by invasive plant species, especially in sites with advantageous soil conditions.
Scientists must also evaluate interactions (both reinforcing and antagonistic) between these drivers. For example, in this study deer herbivory and damage from episodic storms had opposite effects on the density of stems in the understory and therefore the future dynamics of forested stands. Hurricane aftereffects frequently accelerated existing or developing trends resulting from various other drivers (e.g., loss of dogwood to anthracnose disease). [While Ducey et al. also detected lasting impacts from hurricane damage in New Hampshire, these effects did not include changes in tree species composition.] Broader regional and global drivers of change, especially those associated with climate change and nitrogen deposition, interact with these many indicators in novel ways based on their own local loadings.
The Nature Conservancy focuses on fire
The Nature Conservancy magazine for Winter 2023 carries an article describing the organization’s experimental efforts to promote oak succession in the Piedmont forests of North Carolina. Greg Cooper, TNC’s forest ecologist in North Carolina, describes retaining dominance by oaks and hickories – rather than maples and poplars – as vital to protecting the region’s faunal diversity and minimizing impacts from climate change. He says this is because oaks use a quarter of the water of maples and poplars.
Cooper links oaks’ failure to reproduce on fire suppression. TNC kills midstory maples and poplars through hack and squirt methods. This allows more light to penetrate the forest and foster oak seedling recruitment. Then they apply controlled fire. “We currently have 700 acres of [controlled-] burn plots, some of which have been burned twice, some of which have been burned once, [and already] we’re getting more light and an immediate flush of herbaceous diversity. We’re getting a lot more berry species, more wildflowers.” TNC is monitoring plots that have been burned, with and without the pre-burn herbicide treatments, and those that have not been burned. They hope to have results in five to ten years that will indicate whether they are achieving the desired improvement in oak regeneration. If so, they also hope is that in future prescribed burns will be sufficient.
Cooper adds that through the Fire Learning Network and a 23-person fire crew they carry out similar work not just on TNC properties, but also federal and state properties.
SOURCES
Ducey, M.J, O.L., Yamasaki, M. Belair, E.P., Leak, W.B. 2023. Eight decades of compositional change in a managed northern hardwood landscape. Forest Ecosystems 10 (2023) 100121
Payne, C.J. and R.K. Peet. 2023. Revisiting the model system for forest succession: Eighty years of resampling Piedmont forests reveals need for an improved suite of indicators of successional change. Ecological Indicators 154 (2023) 110679
Posted by Faith Campbell
We welcome comments that supplement or correct factual information, suggest new approaches, or promote thoughtful consideration. We post comments that disagree with us — but not those we judge to be not civil or inflammatory.
For a detailed discussion of the policies and practices that have allowed these pests to enter and spread – and that do not promote effective restoration strategies – review the Fading Forests report at http://treeimprovement.utk.edu/FadingForests.htm
