Work in Progress

image of banded elm bark beetle

Banded elm bark beetle found in Fridley

Banded elm bark beetle, an exotic pest from Asian, was first identified in Colorado last year. This year, Minnesota was among five states to find this aggressive pest in residence. Pheromone traps were set out by MDA and APHIS in the Metro< area and in several counties along the Iowa border. In June, three banded elm bark beetles (BEBB) were caught in Fridley, a northern suburb of Minneapolis. Interestingly, the three beetles were caught in three separate traps, each baited with different pheromones.

The map shows locations where BEEB has been found and where it hasn't been found when people have specifically looked for it. The broad distribution probably reflects multiple introductions over a long period of time. It is suspected that banded elm bark beetles have made their way into the US in pallets made of elm wood with the bark still attached.

Exotic insects and diseases all have the potential to alter our natural forest ecosystems and/ or damage ornamental vegetation. We'll have to wait to see what the effects of banded elm bark beetles are and where they become established.

Minnesota participates in a national SOD survey
No Phytophthora ramorum was found

This summer, Minnesota participated in a national sudden oak death (SOD) survey. Phytophthora ramorum is a quarantined pathogen that may have been inadvertently introduced to other states outside of large regulated areas in California and Oregon on infected nursery stock in 2003 or 2004. Given this unfortunate probability, a national detection survey was designed and implemented this year. DNR Forestry participated by gathering samples and information on the possible presence and distribution of the pathogen within the state. Forestry crews, working out of the Resource Assessment office, were trained in data collection and were dispatched to 30 pre-selected sites in 18 counties. See map. The majority of these sites were forested areas adjacent to major nurseries that have the potential to import nursery stock from California nurseries. At each plot location, transects were run in four cardinal directions. Along those transects hundreds of suspect leaf disease samples were collected and later examined. Ultimately, 89 specimens made it to laboratory testing. They were sent to labs in Minnesota (MN. Dept. of Ag.), and Louisiana (Univ. of LA at Baton Rouge). Samples were sent to the lab in Louisiana as a quality control check. All samples were negative for the pathogen Phytophthora ramorum. Neither the disease nor the pathogen was found.

map of counties with SOD surveys

Sudden Oak Death is a forest disease caused by the fungal pathogen Phytophthora ramorum. This pathogen has caused widespread dieback of tanoak and several oak species coast live oak, California black oak, Shreve's oak, and canyon live oak.

The fungus has also been found to infect the leaves and twigs of numerous other plants species. While many of these foliar hosts, such as California bay laurel and Rhododendron species, do not die from the disease, they do play a key role in the spread of P. ramorum. They can act as breeding grounds for the fungus which may then be spread through wind-driven rain, water, plant material, or human activity.

In addition to natural infection in woodland and forested areas, Phytophthora ramorum has also been found in the nursery system on host and associated plants. It is because of these detections that the nursery industry in California is undergoing extensive surveying for the pathogen in high-risk areas as well as falling under heavy regulation. Nurseries in a naturally infested area, or that have been found with previous pathogen infestations, must have a compliance agreement in place before being allowed to ship host and associated plant material. Additionally, any California nursery that has host and associated plants and ships interstate (between states) was placed on hold in April of 2004 and had to be inspected and found free from Phytophthora ramorum before shipping could resume. Just recently, the California Department of Food and Agriculture (CDFA) announced the confirmation of two Phytophthora ramorum-infected coast live oak trees in Golden Gate Park, making San Francisco County the 14th California county regulated for the pathogen.

Map image of outbreak area in Sand Dunes State Forest

Jack pine budworm in red pine

Due to an outbreak of jack pine budworm among red pine plantings in the Sand Dunes State Forest, 128 acres were treated this year with two applications of Bacillus thuringiensis kurstaki (Btk). While follow-up eggmass surveys have not yet been completed this fall, it is looking unlikely that additional treatments will be needed in 2005.

The outbreak was first discovered in July 2003. At that time, the damage had already been done and the adult moths were flying. In a 30Ac portion of the stand, the defoliation was severe enough to cause top kill of the trees. Because of the potential for a bark beetle infestation in the area, the area most severely damaged was cut the winter of '03-04. The rest of the stand was sprayed with Btk in June of 2004.

Spray boundaries were established based on an egg mass survey conducted in November, 2003 with the help of local forestry staff. Defoliation predictions are based on the average number of egg masses found on 15" branch samples (four branches per tree and four trees per plot). Of the 21 plots sampled, egg mass counts indicated 8 plots were at risk of minor defoliation, 7 were at risk of moderate defoliation, and 6 plots were at risk of severe defoliation. Because these sites had already been defoliated once, the concern was that a second defoliation would greatly increase top kill among the residual trees.

Larval surveys were conducted in May 2004 to determine winter survival rates and to monitor the potential spray schedule. Six shoots on each of five trees were surveyed on fifteen plots to determine the presence and location of actively feeding larvae. Based on the literature, selected shoots were evenly divided among those with pollen cones and those without. The number of shoots with actively feeding larvae determines the estimated population size, where a sample of 20 infested shoots out of 30 shoots sampled suggests severe defoliation is likely. Survey results this spring indicated 4 plots were likely to experience severe defoliation, while another 4 were likely to see light to moderate defoliation. Because the cone crop was large this year and almost no larvae were found on shoots without pollen cones, there was some later concern that selecting shoots without cones may have significantly under sampled the population.

Informal larval surveys were also conducted through out the summer at a number of sites at up to 3 miles outside the proposed spray area. In all but one site, larvae were found, but in very small numbers.

Based on larval development, the first application would have ideally taken place the wk of June 7th. Because of heavy rains the first application was actually made on June 15 and the second application was made on June 2nd. Because of the delayed applications, light defoliation was noted across much of the spray area by the time the larvae stopped feeding. However, the treatment was deemed a success based on informal larval surveys completely on June 29th following the final application.

However, due to larval dispersal prior to the Btk applications, the stand immediately to the north of the 2004 spray project (see map) saw light to moderate defoliation. Unlike the stand that was treated this year, there had been no shoot moth in the northern stand and it seemed to be less stressed by the previous drought. The stand also seemed to be handling the defoliation this year fairly well. By early fall, secondary growth had largely covered the previous damage so that it could not be seen from the air.

Prior to the 2003 outbreak, the entire area had been under severe drought stress. That plus additional damage by pine shoot moths contributed to the level of top kill seen in 2003 and the increased risk of pine bark beetle attack. With the heavier than normal rains in 2004, much of that stress was relieved. The rains combined with the increased vigor of the younger stands currently infested means that treatments are not likely to be necessary in 2005. A smaller scale egg mass survey is planned for later this fall to confirm that assumption.

image of damage to a tree stump from TLCB insects

Oak decline, birch decline and firewood

Fewer oaks died this year from stress and two-lined chestnut borers than was expected given the droughty spring weather. In light of that and with the fall rains most areas have received, the end of the TLCB outbreak is anticipated. Even so, it would still be a good idea to be mindful of these borers in oak firewood because they could cause localized outbreaks to flare up.

It is not a good idea to pile firewood infested by two-lined chestnut borers in your yard if you or your neighbor have living oak trees. When the beetles emerge next spring and summer, they are going to find to the nearest oak tree and attack it, especially if the tree is stressed. Most yard trees tend to be stressed due to soil compaction, construction damage, root loss due to trenching for septic drain fields, buried power lines, telephone lines, black top over roots, etc. So even if we get adequate rains many of these trees will still be stressed from other factors and attractive to the borers.

If you have infested firewood in your yard, you should keep a tarp over the branches and logs during May, June and July of next year to prevent the infestation of nearby trees. Dig a shallow trench around the firewood pile, put the edge of the tarp into it and cover with dirt. This helps hold the tarp in place so the wind doesn't blow it off.

image of tree 
half-dead from infestation of BBB

An easy way to determine if you have infested firewood is to find out when the tree died.

  • Any oak logs or branches that died in 2003 no longer have any two lined chestnut borers in them and are safe to pile in your yard without tarping them. You might be able to find the characteristic D-shaped emergence holes through the bark on these trees.
  • Any oak that died from two-lined chestnut borer in 2004 will have hundreds and perhaps even thousands of adult beetles emerging from them next May, June and July. (They will not have any exit holes, yet. These you need to tarp.)

Birch is very similar. Stressed birch trees are usually infested by bronze birch borers (BBB). You can find evidence of these borers in most dead birch trees. Bronze birch borer and two-lined chestnut borer are both in the genus, Agrilus, and have the same type of one-year life cycle. Any birch tree that was infested by bronze birch borers and died in 2004 will produce adult beetles in May, June and July of 2005. Even if that tree is cut up and split into firewood, the bronze birch borer beetles will survive and emerge next summer. If you have other birches in your yard, it is not a good idea to pile infested firewood near them unless you tarp the pile.

There is an additional difference between TLCB and BBB. Birch trees infested with bronze birch borer often take three years or longer to entirely die. Infestation starts in the smaller branches in the top of the crown and, over a number of years, succeeding generations work their way down the tree. So even if you are unable to find evidence of the borer in the lower part of the birch stem, borers may be present higher up on the stem or in the branches. Finding the D-shaped emergence hole tells you that the borers have already emerged and are no longer in that part of the tree, but the next generation of BBB is likely to exist in a lower part of the tree. Logs from the lower part of the tree should be tarped.

Finally, you don't have to worry about uninfested branches or logs becoming newly infested by either two-lined chestnut borers or bronze birch borers. If you cut down a live, uninfested oak or birch tree, these borers are not able to attack and produce offspring in this material.

Dogma run over by karma

Until the late 1990's, the prevailing pathological dogma was that if there were no shoot blight infections in the nursery seedbeds, then there were no Sphaeropsis infections of the seedlings. That belief was turned upside down by the work of Glen Stanosz's lab at the University of Wisconsin. Stanosz and co-workers found that there could be Sphaeropsis infections without any symptoms. Sphaeropsis sapinea is a "latent pathogen". A latent pathogen has three attributes:

  • The pathogen can persist in the host plant in the absence of symptoms.
    Sphaeropsis creates structures called "hyphal bundles" in needle stomates or in the bark surface without any disease reaction or symptoms being expressed.
  • It has a prolonged period of inactivity that is induced by the host.
    As long as the seedling is vigorous, latent Sphaeropsis infections do not occur. Even after outplanting, latent infections in red pine seedlings can persist for as long as seven years.
  • The pathogen can be released to cause disease symptoms as a result of a physiological change in the host.
    Sphaeropsis infections are released by internal water deficits in the red pine seedlings. In fact, the incidence and severity of symptoms are directly related to soil moisture deficits.

As it fate would have it, in the spring of 2002, Dr. Stanosz decided to sample all the Mid-western nurseries to document the incidence of latent Sphaeropsis infections in red pine seedlings. We were certain that Minnesota's nurseries would be fairly clean. See dogma above.

Was it fate, destiny or karma that set the dominoes tumbling?

  • As usual, the Nursery rigorously culled out diseased and deformed seedlings during lifting so that only "clean" and "green" seedlings were shipped.
  • Northern Minnesota experienced a significant drought in 2002, especially during the planting season and early summer.
  • By mid-July, losses in newly-established red pine plantations averaged an unprecedented 55 percent . Only part was due to drought. Lab examinations showed that the remainder of these losses were due to Sphaeropsis.
  • In late August, we learned that Badoura Nursery had 18 and 88 percent latent infections at the two sample locations.

The inescapable conclusion was that Sphaeropsis infections evident in our plantations came from our Nursery in the form of latent infections. Now, not only did we have Sphaeropsis in our Nursery, but we were also in the midst of an epidemic.

Our best course of action was to eliminate the sources of the infection in the Nursery by removing the overstory red pines. (Sphaeropsis spores from infected windbreak red pines are spread in raindrops down onto the seedlings in the nursery beds.) In the winter of '02-03, Badoura Nursery removed 1250 cords of red pine windbreaks. These removals would have an impact, but not immediately. Seedlings that were alive in 2002, prior to the removal of windbreaks, had already been exposed to Sphaeropsis and were likely infected. Nursery managers also reinstituted fungicide spray regimes and rogued seedbeds regularly among other actions.

In 2003, the first systematic survey of the entire 3-0 red pine crop was conducted to determine the levels of Sphaeropsis latency. Dr. Stanosz's lab found that latent infections in the red pine fields averaged 40 to 71 percent. As a result, the entire crop of two million seedlings was rejected and destroyed.

This summer, we anticipated a drop in disease levels because this crop of seedlings had emerged after the windbreaks were removed. Another systematic survey of the 3-0 red pine crop was done and latency levels were determined by Dr. Stanosz's lab. We were delighted to hear that less than 3 percent latent infections were found in the nursery beds. These seedlings will be sold and planted next year. Good karma, at last.

image of oak wilt assessment project for 2004 map

2003-2004 Oak wilt assessment project

The program evaluation completed in 2002 and presented in the October 2003 issue of the FH newsletter raised a number of questions about the status of oak wilt in the state. Although a large number of communities have been aggressive in their management of oak wilt, the number of new infection pockets being discovered each year exceeds the number of infection pockets being treated each year. The result has been a gradual increase in the incidence of oak wilt across most of the original control zone and a dramatic increase in the incidence of oak wilt in Sherburne County - in spite of a model control program.

These increases led to several questions about statewide management strategies and the associated funding. The Forest Health Protection Branch of the USFS, State and Private Forestry provides the current oak wilt suppression funding. Their program targets areas at high risk of forest damage and as such are usually directed at areas with the highest pest incidence. But as seen with the gypsy moth program, not all areas are necessarily given equal consideration because of the likelihood of spread and the adjacent forest acreage at risk of damage. So for instance, the gypsy moth Slow-The-Spread program treats many more acres of forest than does the gypsy moth suppression program, even though the pest incidence is much lower. That is because, in the case of the gypsy moth, control measures have proven to be more cost effective when aimed at the leading edge of gypsy moth expansion or at high value forests under immediate threat of defoliation.

So what are the implications in oak wilt management? Should we target areas along the leading edge of expansion and abandon areas too heavily infested to manage? That might doom adjacent communities who are currently doing their best to limit disease incidence. Or should we continue to target areas with high pest incidence independent of where they occur? That tends to punish those communities that aggressively manage the disease and in the process lower its incidence. Or should we favor those demonstrating a long-term commitment to integrated urban forest management or those struggling to initiate a new program? Any approach we choose favors some communities over others.

Independent of the approach we choose, future federal funding is limited and dependent on demonstrable results. So it is imperative that the state defines where its priorities lie and define management goals accordingly. As such an assessment project was initiated in 2003 and is scheduled for completion this winter.

The hopes are that the assessment project will shed some light on where the incidence of oak wilt is changing and why. The results may help demonstrate which components of a community management program are most effective. The results may also highlight under what conditions disease management is most difficult. How these areas are distributed across the state's oak resource may help determine where we need to focus our efforts or perhaps where we need to redirect federal funds administered under the MN Releaf program.

Photography

Color infrared photography (CIR) was taken of various parts of the study area in 1988, 1989, 1993, 1997, 1999, 2002 & 2003. Photographs taken in 1988, 1989 and 1993 were combined to cover all of Anoka and Ramsey counties, and roughly half of Washington, Chisago, Isanti and Sherburne counties. This set of photos was used to represent time one (T1) in the assessment project. Photographs taken in 2002 and 2003 were combined to cover all of Anoka, Ramsey and Sherburne counties, as well as parts of Chisago, Isanti, Dakota, Hennepin, Scott and Washington. This set of photos was used to represent time two (T2). The intersection of these two photo periods defined the project boundaries (see the CIR map). Photos taken in 1997 and 1999 were not used as part of this study.

Plot Selection and Design

The study area was divided into quarter sections, based roughly on the public land survey. The percent forest cover for each quarter section was determined based on the 1995 GAP data (1). A total of 600 plots were randomly selected from those quarter sections with greater than or equal to 15% forest cover. After exploring different thresholds of forest cover, 15% seemed to represent the best compromise between the labor needed to sample a large number of plots falling into agricultural areas if the threshold of forest cover was set too low, and potentially under sampling highly developed areas if the threshold of forest cover was set too high.

A regular grid of points was laid over the top of each quarter section plot, with an average of 64 points per plot. Each point represented roughly 1/10 acre. Both sets of photographs were rectified to the 2002 FSA photography. The office of DNR Forest Resource Assessment interpreted the photos covering each plot for both time periods. Land usage was defined as outlined below. Polygons representing each land use were visually determined. All points falling within a particular polygon were assigned to that land use category. For each point within a land use, the presence/absence of tree canopies and the presence/absence of oak wilt were recorded.

Field Checks

Five percent of the plots in each county were randomly selected to be field checked. Paper maps were created using the 2002 FSA photography as the base layer. County parcel data was laid over each plot and the landowner determined, where that data was available. Identified landowners were contacted for permission to access their land. Where landowners could not be identified, letters were sent to the current resident. All but one landowner contacted provided access to their property.

Field staff attempted to visit all points that appeared to have trees in the immediate vicinity. Points that were inaccessible were viewed with binoculars from the nearest possible location. Out of a total of (x) possibly treed points, (y) were visited and (z) were viewed from a distance. At each point tree cover was determined to be hardwood, conifer, mixed or none. Disease incidence was recorded as present, absent or unconfirmed, which meant dead oaks were present, but the cause of death could not be determined.

Analysis

The change in incidence of oak wilt will be calculated for each civil division on an area basis. Regression analysis will be performed by the USFS, NC Forest Exp. Sta. to determine the correlation, if any, between the change in disease incidence and the independent variables. At this point in time, all interpretation and fieldwork is complete. The final analysis will be completed and the final results presented in 2005.