Tree-ring response of jack pine and scots pine to budworm defoliation in central Canada

Forest Ecology and Management 347 (2015) 83–95

Contents lists available at ScienceDirect

Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco

Tree-ring response of jack pine and scots pine to budworm defoliation in central Canada J.R.M. Robson, F. Conciatori, J.C. Tardif ⇑, K. Knowles Centre for Forest Interdisciplinary Research (C-FIR), University of Winnipeg, Environmental Studies and Sciences, 515 Portage Avenue, Winnipeg, Manitoba R3B 2E9, Canada

a r t i c l e

i n f o

Article history: Received 31 August 2014 Received in revised form 20 February 2015 Accepted 9 March 2015 Available online 26 March 2015 Keywords: Dendrochronology Climate change Insect outbreak Plantation forestry Exotic tree species Pine sawfly

a b s t r a c t Insect outbreaks constitute major disturbances and global climate changes are expected to increase their frequency and severity. In Canada, an increase in outbreak severity of the jack pine budworm is expected as a consequence of more frequent droughts associated with climate changes. In this study, the impact of jack pine budworm defoliation on radial growth was assessed on two host species: jack pine (Pinus banksiana Lamb.) and scots pine (Pinus sylvestris L.). Standard tree-ring chronologies were developed for each host species in thirteen plantations established in the early 20th century and located in Spruce Woods Provincial Forest (central Canada). Radial growth suppressions caused by jack pine budworm defoliation were identified using a host and non-host comparison and calibrated against historical outbreak records. Five periods of major growth suppression were identified (1956–1958, 1966–1968, 1974–1977, 1979– 1980 and 1984–1986) that matched historical jack pine budworm outbreaks. An annual tree-ring signature made up of a tree ring with thin latewood followed by a narrow ring most often characterized these growth suppressions. The occurrence of missing rings also increased during outbreaks. Based on the timing of suppression, jack pine was the initial host with scots pine often showing a one year lag in suppression. However, scots pine may be more sensitive to jack pine budworm defoliation as indicated by the abundance of missing rings during outbreak years. In the study area, jack pine budworm outbreaks were generally associated with the occurrence of dry summers and cool May temperatures. No outbreak occurred in the study area since the mid-1980s. The occurrence of droughts that were not synchronized with cool May temperatures suggests that warmer springs associated with climate changes could alter the phenological synchrony between the jack pine budworm and its host trees species. Future research should attempt to (i) relate the results of this study to natural forest stands where management practices (and non-native tree species) have not influenced the natural jack pine budworm population dynamics, (ii) assess the spatial dynamics of these outbreaks using a network of tree-ring chronologies and (iii) reconstruct outbreaks prior to historical surveys. Such research would help develop a better understanding of insect population dynamics and subsequent impacts on both European and North American forests under future climate changes. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Insect outbreaks constitute major forest disturbances causing important volume loss and tree mortality in forests around the world (Haack and Byler, 1993; Fleming, 2000; Netherer and Schopf, 2010). In the scots pine (Pinus sylvestris L.) forests of northern Europe the most widespread outbreaks are from the large pine sawfly (Diprion pini L.) with growth losses and mortality rates recorded as high as 94% and 30% respectively (LyytikäinenSaarenmaa and Tomppo, 2002; Lyytikäinen-Saarenmaa et al., 2003). In North America’s boreal forests, insect outbreaks cause ⇑ Corresponding author. Tel.: +1 204 786 9475. E-mail address: [email protected] (J.C. Tardif). http://dx.doi.org/10.1016/j.foreco.2015.03.018 0378-1127/Ó 2015 Elsevier B.V. All rights reserved.

the greatest loss in forest volume compared to any other forest disturbance including fire (Haack and Byler, 1993; Fleming, 2000; Bogdanski, 2008). Among these insects, the jack pine budworm (JPBW, Choristoneura pinus pinus Free., Order: Lepidoptera, Family: Tortricidae), a major defoliator of jack pine (Pinus banksiana Lamb.), causes repeated damage in commercially valuable stands. During the most widespread and severe JPBW outbreak recorded in central Canada (1982–1987) the forested area impaired by severe damage from JPBW defoliation reached over two million hectares in the province of Manitoba by 1985 (Volney, 1988; Grandmaison, 1991, 1993). The reduced volume and mortality associated with severe defoliation during periodic JPBW outbreaks rapidly lowers timber quality, reduces available harvest volume by as much as one third and can double the time

84

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

for a stand to reach a merchantable harvest size (Clancy et al., 1980; Moody and Amirault, 1992). Furthermore, both forest fire suppression and the predicted increase in drought stress under future climate change scenarios may lead to older stands with higher staminate cone abundance thus leading to more frequent and severe JPBW outbreaks (Volney, 1988; McCullough, 2000; Volney and Fleming, 2000).

1.1. Jack pine budworm (JPBW) The JPBW is found throughout the entire range of its main host, jack pine, which has the largest natural range of any northern pine species in Canada (Freeman, 1953; Rudolph and Laidly, 1990; McCullough, 2000). The range of jack pine (Fig. 1) spreads across Canada from the Northwest Territories to Nova Scotia and also extends south to Minnesota, Wisconsin, Michigan and New York in the United States (Rudolph and Laidly, 1990; Conway et al., 1999b). Within its range the JPBW also defoliates scots pine, red pine (Pinus resinosa Ait.) and white pine (Pinus strobus L.) (Hodson and Zehngraff, 1946; Kulman et al., 1963; Reeks, 1971; McCullough, 2000). Scots pine is a non-native species that was introduced to North America during European settlement in the 17th century (Skilling, 1990). Severe JPBW defoliation has also been observed in non-native lodgepole pine (Pinus contorta Dougl. ex. Loud.) plantations in southwestern Manitoba (Brandt and McDowall, 1968; Walker, 1990). The JPBW completes one lifecycle generation within 12 months over the course of two summers. The JPBW has seven distinct larval instars that cumulate with an adult moth (Nealis, 1995; McCullough, 2000). The adult moth is a short-lived phase that does not damage host trees but functions in reproduction and dispersal (McCullough, 2000). In the first year, the moth will lay eggs on host trees. The first instar larvae will emerge from the egg, spin a hibernaculum for winter diapause, molt and over winter as second instar larvae. Very little feeding occurs in the first year (McCullough, 2000). Between May and June of the second year,

second instar larvae emerge from the hibernaculum and move into staminate cones to feed on pollen (Nealis and Lomic, 1994; Nealis, 1995; McCullough, 2000). By the time the pollen supply is exhausted, bud-break has usually occurred and the larvae will begin to feed on the current year’s foliage. During an outbreak, high populations of the JPBW larvae will also feed on the previous year’s foliage with the sixth and seventh larval instars doing the most damage (Clancy et al., 1980; Nealis et al., 1997). In mid-July the seventh instar larvae will pupate and adult moths will emerge within two weeks to begin the cycle again (McCullough, 2000). Both larval development and survival in JPBW are influenced by climate conditions (Clancy et al., 1980; Ives, 1981; Volney, 1988; Nealis, 1990; Nealis et al., 1997). Extreme temperature fluctuations during the winter diapause were associated with high larvae mortality (Lysyk, 1989; Nealis, 1995). Warm dry conditions throughout the development stages of the larvae and pupae will favour higher survival rates (Clancy et al., 1980; Ives, 1981). However, research on the direct relationship between JPBW population survival and climate is sparse (Nealis, 1995; Volney and Fleming, 2000, 2007).

1.2. JPBW outbreaks Jack pine budworm outbreaks have been recorded throughout the range of jack pine in both Canada and the United States (Kulman et al., 1963; Clancy et al., 1980; Volney, 1988). In Canada, large periodic JPBW outbreaks have mainly been reported for the central region including eastern Saskatchewan, Manitoba and northwestern Ontario (Fig. 1, Ives, 1981; Volney, 1988; Fleming, 2000). In Manitoba and Saskatchewan, populations of JPBW have been detected every year from 1937 to 1986 except 1980 and 1981 (Volney, 1988). Analysis by Volney (1988) indicated six major outbreaks within Manitoba with peaks in severity occurring in 1938, 1944, 1948, 1965, 1979 and 1985. Outbreaks usually last from one to four years with no more than two years of severe defoliation at the local scale (Clancy et al., 1980; Volney, 1988; McCullough et al., 1996; Kouki et al., 1997). In general, the interval

Fig. 1. Map of North America indicating the range of jack pine distribution (top left). Map of central Canada showing location of Manitoba, of Spruce Woods Provincial Forest (SWPF) and of Spruce Woods Provincial Park (SWPP) within SWPF (bottom left). Locations of plantation regions within SWPF (right): Camp Picnic (CP), Camp Hughes (CH), and Camp Shilo (CS).

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

between outbreaks is 6–12 years (Volney, 1988; Volney and McCullough, 1994; McCullough, 2000). The climate relationship with JPBW population dynamics is modulated through the host trees’ staminate cone production. During an outbreak, the increase and survival of budworm larvae populations is strongly related to the staminate cone production (Hodson and Zehngraff, 1946; Nealis, 1990; Nealis and Lomic, 1994; McCullough et al., 1998). High pollen production promotes survival rates of JPBW larvae (Nealis and Lomic, 1994; McCullough, 2000). Drought conditions stimulate greater staminate cone production in trees and drier sites in the southern limit of the JPBW’s range tend to experience more frequent JPBW outbreaks (Nealis, 1990; Volney and McCullough, 1994; Volney and Fleming, 2000). Following a year of severe JPBW defoliation the host trees’ staminate cone production decreases (Hodson and Zehngraff, 1946; Nealis and Lomic, 1994; Nealis et al., 2003; Rhainds et al., 2012; Hughes et al., 2014). This decrease combined with parasitism by an estimated 16 species of generalist parasites of Choristoneura spp. on late larval instars and pupae usually leads to the decline of JPBW populations, which accounts for the short duration of local outbreaks (Nealis, 1991; Nealis and Lomic, 1994; Volney and McCullough, 1994; Nealis, 1995; McCullough, 2000).

85

used in dendrochronological and dendroclimatological research (Brooks et al., 1998; Hofgaard et al., 1999; Girardin et al., 2009; Hoffer and Tardif, 2009; Pisaric et al., 2009; Tardif et al., 2011; Genries et al., 2012; Huang et al., 2013). Despite its importance, very few studies have documented the host species’ tree-ring response to JPBW defoliation (Kulman et al., 1963; Gross, 1992; Volney and Mallet, 1992). This is the first study that calibrates long-term records of JPBW outbreaks with radial growth suppressions identified in host species. Not only could tree-ring records provide more accurate predictions for long-term JPBW population dynamics (Volney, 1988) but they could also provide a better assessment of the signal to noise ratio in dendroclimatic reconstructions. Furthermore, little is known about the impact of insect defoliators on scots pine, an introduced species to North America. Given the availability of historical documents depicting both JPBW outbreaks and related management practices for plantations in Spruce Wood Provincial Forest (SWPF) four objectives were put forward. The first objective was to identify in both host species the tree-ring signature associated with documented outbreaks. The second objective aimed at comparing the sensitivity of the two host species to JPBW defoliation. The third and fourth objectives were to investigate the impact of management operations and of selected climate variables on JPBW outbreak dynamics.

1.3. JPBW outbreaks and tree growth Jack pine budworm defoliation occurs primarily in the upper crown of host trees where staminate cones are more abundant. Depending on the severity of defoliation, the JPBW may cause significant growth loss, top-kill or mortality of the host tree (Hodson and Zehngraff, 1946; O’Neil, 1962; Kulman et al., 1963; Gross, 1992; McCullough et al., 1996; Volney, 1998). The risk of mortality becomes greater for host trees when they begin producing large staminate cone crops between the ages of 10 and 25 (McCullough, 2000; Nealis et al., 2003). Trees producing the largest staminate cone crops are usually dominant or co-dominant, in older age classes or suppressed (Mallet and Volney, 1990; Nealis and Lomic, 1994; Kouki et al., 1997; Conway et al., 1999b; Hughes et al., 2014). Host tree mortality is known to occur on healthy dominant trees after two successive years of severe JPBW defoliation (Moody and Amirault, 1992). From a growth perspective, the first year of a JPBW outbreak has been associated with the formation of a tree-ring with a thin latewood (Kulman et al., 1963; Gross, 1992). Radial growth suppression follows with narrow rings being produced until host tree recovery (Kulman et al., 1963; Gross, 1992; Moody and Amirault, 1992; Volney and Mallet, 1992; Conway et al., 1999b). In some situations the reduction in radial growth may be delayed by one to two years following defoliation (Kulman et al., 1963; Gross, 1992; Moody and Amirault, 1992) due to stored reserves being used thus compensating for reduced photosynthesis in the year of defoliation (Kulman, 1971; Schweingruber, 2007; Speer, 2010). In surviving host trees, growth suppression may continue until one year following JPBW defoliation (Volney and Mallet, 1992; Gross, 1992; Conway et al., 1999a). However, in trees showing top-kill, suppression may last longer and lead to host-tree decline (O’Neil, 1962; Gross, 1992; Volney and Mallet, 1992; Volney, 1998; Conway et al., 1999b). Furthermore, the production of light rings (e.g. tree rings with fewer latewood tracheids and thin cell walls) were associated with declining jack pine trees following severe JPBW defoliation (Volney and Mallet, 1992).

2. Material and methods 2.1. Study area The study area is located in SWPF (49.8°N, 99.4°W) in the prairie ecozone of southwestern Manitoba (Fig. 1). The origin of SWPF can be dated back to 1895 (MCWS, 2013). In 1964, Spruce Woods Provincial Park (SWPP) was created within SWPF (Henderson et al., 2002; MCWS, 2013). At an elevation of 366 metres above sea level (masl), the SWPF covers an area of 601 square kilometers (Fig. 1; Smith et al., 1998; Henderson et al., 2002). The climate of the region is characterized by a warm growing season with few precipitation followed by long cold winters (Smith et al., 1998). Climate data from the Brandon CDA weather station (49.5°N and 99.6°W, 363 masl) located within 30 km of the SWPF indicate an annual mean temperature and total precipitation of 2.7 °C and 461.7 mm (46% of which falls from June to August) for the reference period 1981–2010 (Environment Canada, 2014). Within the prairie ecozone the SWPF falls into the transition between the boreal forest and the prairie grassland (Smith et al., 1998). Forest islands of native white spruce (Picea glauca L.) are commonly found throughout the region’s sandy dunes (Smith et al., 1998; Henderson et al., 2002; Chhin et al., 2004). Eastern larch (Larix laricina [Du Roi]), intermixed with white spruce and black spruce (Picea mariana Mill. B.S.P.) is found growing in moist lowland areas (Smith et al., 1998). The dominant deciduous tree species include trembling aspen (Populus tremuloides Michx.), balsam poplar (Populus balsamifera L.) and bur oak (Quercus macrocarpa Michx.) (Smith et al., 1998; Henderson et al., 2002). Experimental forest plantations of white spruce, and non-native coniferous species including scots pine, jack pine, lodgepole pine and Norway spruce (Picea abies [L.] Karst) were also established throughout the region between 1904 and 1979 (Walker, 1990; Henderson et al., 2002).

1.4. Research objectives

2.2. Plantation selection

In Canada, jack pine is one of the major commercial softwood tree species (Rudolph and Laidly, 1990; Sims et al., 1990; Gross, 1992; Pines et al., 1995; Conway et al., 1999a) and is also widely

In SWPF, scots pine and jack pine plantations were established in three regions, Camp Picnic (CP), Camp Hughes (CH), and Camp Shilo (CS), each within 20 km of each other (Fig. 1). Within each

86

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

region, historical plantation maps with location, species, and establishment data were used to select specific plantations for sampling. When feasible, each selected jack pine plantation within a region was paired with a scots pine plantation of a similar establishment date. Thirteen plantations (11 monospecific and 2 mixed) with establishment dates from 1916 to 1945 were sampled including remnants of one 1905 scots pine plantation at CH (Table 1). 2.3. Field sampling methods In each of the 13 plantations, sampling was conducted along a line transect (northwest to southeast). Starting at 20 m inside the plantation boundary and every subsequent 10 m, the closest living dominant or co-dominant tree on either side of the transect was sampled for a total of 20 trees. In the mixed plantations (CP and CS) the transect length was extended to 110 m for a total of 22 trees. If the transect intersected large canopy gaps due to past silvicultural treatments (row thinning and salvage logging) it was adjusted to maintain a homogenous pine forest cover type. For each selected tree, two increment cores were extracted approximately 30 cm from the base of the tree and at least 90 degrees apart. In some cases, trees were cored at breast height (1.3 m) because of rot. In addition to sampling host species, between 10 and 15 non-host white spruce trees were similarly sampled at each region in SWPF. In each plantation, the height, diameter at breast height (DBH, 1.3 m) and the location of each sampled host tree were recorded. A point-centred quadrat (PCQ) was established using the selected tree as its centre to calculate (1) Hegyi’s competition index for each sampled tree (Husch et al., 2003) and (2) the plantation tree density and basal area (Table 1; Mitchell, 2007). This data indicated that at similar ages, jack pine trees tended to be smaller in diameter and height compared to scots pine trees in matched plantations (Table 1). One exception was the 1934 scots pine plantation at CP (CP 3-34) that also shows a greater tree density than any other plantations (Table 1). 2.4. Laboratory methods and chronology development All tree ring samples were processed using standard dendrochronological methods (Swetnam et al., 1985; Speer, 2010). After mounting and drying, the cores were sanded so tracheids and annual ring boundaries could be clearly observed under the microscope. Proper sanding procedures are essential for accurate detection of tree-ring boundaries, micro-rings and light rings. In this study, crossdating followed the list method approach (Speer,

2010) and allowed identification of pointer years and plantation establishment date. All crossdating was also cross-validated by an experienced dendrochronologist. Following completion of visual crossdating, the tree-rings of individual cores were measured to the nearest 0.001 mm using a Velmex measuring system and program MeasureJ2X (Version 4.2). The earlywood and latewood widths were measured separately in the host species to determine the proportion of latewood within an annual ring. In this study, the earlywood to latewood boundary was defined as the transition where the cell walls begin to thicken, cell lumen shrinks and there is a noticeable qualitative darker colour change in cell walls (Hoffer and Tardif, 2009; Stahle et al., 2009). The earlywood and latewood measurements together also provided the total ring width per year. Crossdating and measurements were further validated using the program, COFECHA (Version 6.00; Holmes, 1983). Potential problems were re-checked and corrected if needed. From the validated raw measurements, the latewood proportion was calculated in the host species by dividing the latewood width by the total ring width for each year in a series. For each of the tree-ring parameters (earlywood width, latewood width, total ring width and latewood proportion), standard chronologies were developed using the program ARSTAN (Version 44h2; Cook and Holmes, 1999). The measurement values were detrended using a cubic smoothing spline of 20 years with a frequency response of 50% to produce radial growth indices. Such a flexible spline retains more than 99% of the variance at a wavelength of six years (Speer, 2010) corresponding to the short duration of JPBW outbreaks (McCullough, 2000). A total of 15 standard chronologies were thus developed; one for each of the eleven monospecific host plantations (five jack pine and six scots pine) and one for each species in the two mixed plantations (two jack pine and two scots pine). A single standard ring-width chronology was also developed for the non-host species (white spruce) by pooling tree-ring measurements collected from the three regions in SWPF (this study) and in SWPP (Fig. 1; Chhin et al., 2004). The non-host standard chronology was created using the same ARSTAN procedures as for the host chronologies. The white spruce non-host chronology was developed from 140 series and had a mean r-bar of 0.532 for all series indicating strong common signal among the standardized series. 2.5. Outbreak identification A review of literature and historical survey records was done to determine the timing and length of JPBW outbreaks. The Canadian

Table 1 Selected plantations in Spruce Woods Provincial Forest. Regions are Camp Picnic (CP), Camp Hughes (CH) and Camp Shilo (CS). The mean and standard deviation (in parenthesis) are indicated. Pba: jack pine, Psy: scots pine.

a

Region

Plantation

Species

Planting Yr.

DBH (cm)

Height (m)

Competition index

Basal area (m2/ha)

Density (trees/ha)

CP CP CP CP CP CP CH CH CH CS CS CS CS CS CS

16-34 03-34 09-44 03-45 05-46a 05-46a 02-05 06-30 08-30 16-17 12-16 05-33 01-32 02-32a 02-32a

Pba Psy Pba Psy Pba Psy Psy Pba Psy Pba Psy Pba Psy Pba Psy

1934 1934 1944 1945 1946 1946 1905 1930 1930 1930 1916 1933 1932 1932 1932

24.8 23.3 25.1 30.1 19.9 25.0 34.7 27.6 34.1 25.9 30.8 25.0 25.3 23.2 29.8

22.6 20.3 23.0 23.0 20.3 20.9 24.9 22.1 25.1 20.8 26.1 21.2 18.1 19.8 20.0

1.3 1.5 1.1 1.1 1.6 1.3 1.2 1.1 0.9 1.1 1.2 1.2 1.2 1.4 1.0

31.2 48.1 29.9 40.7 17.0 25.8 58.5 27.7 34.2 33.6 49.8 31.1 33.8 20.9 16.6

618.4 1056.7 693.0 642.1 665.6 460.8 676.0 466.7 430.0 702.7 891.1 687.1 835.3 601.5 311.1

Mixed species plantations of Pba and Psy.

(3.7) (3.4) (3.2) (2.6) (1.9) (3.6) (5.1) (3.7) (7.4) (3.4) (3.5) (3.8) (3.9) (2.7) (4.7)

(1.6) (2.1) (1.4) (1.4) (0.6) (1.3) (1.6) (1.5) (2.1) (1.2) (1.5) (1.1) (1.2) (0.8) (1.7)

(0.3) (0.4) (0.3) (0.3) (0.4) (0.5) (0.3) (0.4) (0.3) (0.2) (0.3) (0.4) (0.3) (0.2) (0.4)

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

Forest Service has kept detailed historical records of JPBW outbreaks in the Prairie Provinces since 1937. Information was collected from the Forest Insect Survey Reports (1937–1950), the Forest Insect and Disease Reports (1950–1968) and the unpublished Annual District Reports of the Forest Insect and Disease Survey in Winnipeg (1952–1969) as well as the Forest Insect and Disease Condition reports published in Edmonton (1970-present). Herein, these historical survey reports will collectively be referred to as the Forest Insect and Disease Survey (FIDS) records. Records showed that JPBW outbreaks occurred from 1939 to 1942, 1956 to 1958, 1964 to 1967, 1974 to 1976, 1979 and 1984 to 1986 within the SWPF (Brandt and McDowall, 1968; Volney, 1988; Walker, 1990; Rhainds et al., 2012). Given that JPBW and the eastern spruce budworm (SBW, Choristoneura fumiferana [Clem]) were considered the same species prior to 1953 (Brown and MacKay, 1943; Freeman, 1953; Ives, 1981), reported outbreaks before 1953, unless otherwise mentioned, were excluded. In addition to comparison of the host species standard chronologies to the FIDS records, the ring-width indices of the host species from each plantation were also compared with the nonhost chronology using the OUTBREAK program (Version 6.00; Holmes and Swetnam, 1996). Host and non-host comparison is based on two important assumptions: (i) non-host tree growth is not affected by the defoliator and (ii) both the host and non-host species have similar growth responses to climate and other environmental factors (Swetnam et al., 1985). For this approach to be valid, the radial growth of both host and non-host species should be highly correlated in absence of outbreaks. In this study, white spruce constituted a suitable non-host species as it is not defoliated by the JPBW and the non-host standard chronology was significantly correlated to that of the two host species (r > 0.70, p < 0.001). Previous research has also shown that both jack pine and white spruce share similar climate responses (Girardin and Tardif, 2005; Girardin et al., 2009). Radial growth of white spruce in the area is also clearly negatively associated with summer drought (Chhin et al., 2004). To identify JPBW outbreaks, the duration of radial growth suppression was set from one to five years with a maximum growth reduction threshold of 1.28 standard deviations and a rate of increase in growth reduction set to 1.0. These parameters were based on the reported length of JPBW outbreaks (Volney, 1988; Gross, 1992; McCullough, 2000) and the timing of radial growth loss identified in previous studies (Kulman et al., 1963; Gross, 1992). The fractional power to raise the non-host chronology was set to 0.1 to avoid false identification of host suppression when non-host growth indices are not negative (Holmes and Swetnam, 1996). Each period of suppression identified by the OUTBREAK program was further investigated and validated using the FIDS records.

87

MR and TRS in each plantation/species was compared with both the OUTBREAK program results and the FIDS records.

3. Results 3.1. Outbreak identification The standard chronologies of the two host species and the nonhost species showed a high level of concordance (year to year variation) in non-outbreak years (Fig. 2) stressing that they shared similar responses to environmental (climate) forcing in non-outbreak years, an essential condition in this type of analysis. Radial growth suppressions corresponding to the five recorded FIDS outbreaks were also observed in both host species with no correspondence in the non-host white spruce. Suppressions were observed beginning in 1956, 1966, 1974, 1979 and 1984 with variability in start date and severity noted among plantations. The 1966 suppression lagged the 1964 FIDS outbreak by two years (Fig. 2). During the 1974 FIDS outbreak the beginning of suppression among plantations varied between 1974 and 1977. Some plantation chronologies also showed suppression in 1983, one year prior to the 1984 FIDS outbreak. A unique period of radial growth suppression was also observed in scots pine chronologies from 1992 to 1995 with no equivalent in jack pine chronologies or FIDS records (Fig. 2). The OUTBREAK results allowed better capturing of the variability in radial growth suppressions among regions and plantations (Fig. 3). Similar to previous results, the suppression periods identified as potential JPBW outbreaks generally matched the FIDS outbreaks. During the 1956 FIDS outbreak, major suppressions were identified at three plantations: CH 8-30 scots pine, CH 6-30 jack pine and CP 3-45 scots pine with the most severe suppression during any FIDS outbreak occurring at this time in CH 8-30 (Fig. 3). The 1964 FIDS outbreak was mainly restricted to the CS plantations starting in 1966. Herein, the 1964 FIDS outbreak will be referred to as the 1966 FIDS outbreak. During this outbreak a one-year lag between the CS jack pine and scots pine plantations was also apparent with maximum suppression occurring in 1967 and 1968 in each species respectively (Fig. 3). Other periods of major radial growth suppression began between 1974 and 1977 and again in 1979 and 1984. The outbreak beginning in 1974 was also mainly observed in the CS plantations (5-33, 1-32 and 2-32) and in plantation CH 6-30 (Fig. 3). Again, a

2.6. Tree-ring signature and JPBW outbreaks To determine if JPBW outbreaks were associated with the formation of a thin latewood ring in the first year of defoliation followed by at least one narrow ring, a simple tree-ring signature (TRS) index was calculated for each tree using the mean value of the two cores collected and summed by plantation/species. This simple index used latewood proportion (LWP) and total ring width (RW) measurements. For any given measurement year, a value of 1 was given when (i) LWP(t) 6 20% and (ii) [RW(t + 1)/ RW(t)]  100 6 60% where LWP in year (t) is equal to or less than 20% and RW in year (t + 1) is equal to or less than 60% of the RW in year (t). In addition, the frequency of missing rings (MR) in all measured series was calculated to determine if (1) missing rings were more often associated with identified JPBW outbreaks and (2) which of the host species was more affected. The frequency of

Fig. 2. (A) Jack pine plantations (dark grey lines) and white spruce (black dotted line) standard chronologies for Spruce Woods Provincial Forest (SWPF). (B) Scots pine plantations (dark grey lines) and white spruce (black dotted line) standard chronologies for SWPF. The vertical dashed lines indicate the first year of outbreaks (1956, 1964, 1974, 1979, and 1984) in SWPF as identified by the FIDS records.

88

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

Fig. 3. Mean corrected normalized indices from all jack pine (left column) and scots pine (right column) series in a plantation. The mean annual suppression (grey shading) is indicated. Black shading corresponds to the years for which the corrected indices fell below 1.28 standard deviations (dotted line). The vertical dashed lines indicate the first year of outbreaks (1956, 1966, 1974, 1979, and 1984) as identified by OUTBREAK. Note that the 1964 vertical dashed line was moved to 1966 to more adequately depict the timing of suppression associated with the 1964 FIDS outbreak.

one-year lag was observed between jack pine and scots pine with maximum suppression observed in 1975 and 1976 respectively. The 1979 FIDS outbreak was detected in all three regions meeting the OUTBREAK program’s criteria in more than 70% of the plantations (Fig. 3). During the 1984 FIDS outbreak, radial growth suppressions were observed in jack pine plantations from all three regions but the OUTBREAK program’s criteria were only met at one plantation (CP 9-44). During this outbreak, scots pine plantations showed very little suppression in comparison to jack pine plantations (Fig. 3). In addition to the FIDS records of JPBW outbreaks, the OUTBREAK program also identified other periods of major radial growth suppression. Severe suppressions from 1992 to 1995 were observed in scots pine at CP and CH plantations with minor ones observed at the CS region (Fig. 3). Another suppression period with no equivalent in the FIDS records was identified beginning in 2006 at CS 2-32, CS 1-32, CH 6-30 and CP 5-46 (Fig. 3). During FIDS outbreaks, results revealed that over 50% of the tree-ring series within a plantation usually registered major radial growth suppressions (Fig. 4). The regional nature of the 1956, 1966, and 1974 outbreaks was again revealed. The results indicated that the 1974 FIDS outbreak started at CS and culminated in the large scale 1979 outbreak identified at all three regions. The 1984 FIDS outbreak was more specific to jack pine with suppression in scots pine never affecting more than 50% of the samples. The unique 1992 to 1995 suppression period identified in scots pine showed the highest percent of trees with synchronized growth suppression being observed at a large scale (Fig. 4). Comparing the radial growth suppression of the two host-species within mixed plantations (CS 2-32 and CP 5-46), where growing conditions and management practices were presumably similar to both species, also revealed specific differences (Figs. 3 and 4). At CS 2-32 during the 1966 and 1974 FIDS outbreaks the onset of suppression in scots pine lagged that of jack pine by one year. At CP 5-46, the first outbreak identified corresponded to the 1979 FIDS outbreak that was recorded by more than 50% of the

series in both species. The 1979 FIDS outbreak was not identified at plantation CS 2-32, however, radial growth suppressions were observed in more than 50% of the plantation’s series between 1977 and 1979. Within CS 2-32 the jack pine series did not record strong suppression in 1979 compared to the scots pine, however, during the 1984 FIDS outbreak major suppression was observed in the jack pine series compared to the negligible suppression in the scots pine (Figs. 3 and 4). 3.2. Tree-ring signature (TRS) Results indicated that the sequence of years with a high frequency of tree-rings with a thin latewood followed by a narrow ring generally corresponded to both the FIDS records and OUTBREAK suppressions (Fig. 5). The tree-ring signature was also observed more frequently in regions and plantations that recorded the most severe radial growth suppression. For example, the mean TRS frequency in 1966 among the CS plantation chronologies was 63.0% (±8.2% standard deviation; SD, n = 6) compared to 3.3% (±2.9% SD, n = 3) and 4.2% (±10.2% SD, n = 6) at CH and CP plantation chronologies respectively (Figs. 3 and 5). Overall, the TRS was more frequent at CS during identified outbreaks compared to CH and CP. During the 1979 outbreak, the mean TRS frequency at CS plantation chronologies was 63.6% (±24.1% SD, n = 6) compared to 46.7% (±33.3% SD, n = 3) and 33.2% (±23.1 SD, n = 6) in CH and CP respectively (Fig. 5). The TRS however also occurred during years that did not correspond to FIDS outbreaks (Fig. 5). In a majority of plantation both the 1950 and 1960 tree rings were characterized (when species chronologies existed) by a thin latewood followed in 1951 and 1961 by a narrow ring. In 1950 and 1960, the TRS had a respective mean frequency of 48.2% (±25.5 SD; n = 12) and 23.6% (±21.1 SD; n = 15). In both years, the TRS was also recorded more often in scots pine plantations (1950 mean: 59.1% ± 23.5 SD; n = 6 and 1960 mean: 34.2% ± 20.3 SD; n = 8) compared to jack pine plantations (1950 mean: 37.2% ± 24.4 SD; n = 6 and 1960 mean:

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

89

Fig. 4. Relative frequency of tree-ring series meeting the OUTBREAK criteria for JPBW outbreaks for jack pine (left column) and scots pine (right column) plantations. The solid black line indicates the number of series in each year and the horizontal dashed line indicates where 50% of the sample meets the OUTBREAK criteria. The vertical dashed lines indicate the first year of outbreaks (1956, 1966, 1974, 1979, and 1984) as identified by OUTBREAK. Note that the 1964 vertical dashed line was moved to 1966 to more adequately depict the timing of suppression associated with the 1964 FIDS outbreak.

Fig. 5. Frequency of tree-ring characterized by a thin latewood ring followed by a narrow ring (see methods for details) occurring in (A) jack pine and (B) scots pine plantations by region. Stacked bars are shown for each plantation in a region with a possible maximum value of 700% for jack pine and 800% for scots pine. The vertical dashed lines indicate the first year of outbreaks (1956, 1966, 1974, 1979, and 1984) as identified by OUTBREAK. Note that the 1964 vertical dashed line was moved to 1966 to more adequately depict the timing of suppression associated with the 1964 FIDS outbreak.

11.4% ± 15.5; n = 7). In contrast to other suppression periods, few series and plantations recorded a TRS during the 1992 to 1995 scots pine suppression (Fig. 5). The frequency of missing rings in both species also generally corresponded to FIDS outbreaks with a one year lag (Fig. 6). Across the regions no missing rings were observed before 1957. For both host species the maximum frequency of missing rings was observed during the 1979 outbreak. Eleven of the 15 plantation species chronologies recorded missing rings from 1979 to

Fig. 6. Frequency of missing tree rings occurring in (A) jack pine and (B) scots pine plantations by region. Stacked bars are shown for each plantation in a region with a possible maximum value of 700% for jack pine and 800% for scots pine. The vertical dashed lines indicate the first year of outbreaks (1956, 1966, 1974, 1979, and 1984) as identified by OUTBREAK. Note that the 1964 vertical dashed line was moved to 1966 to more adequately depict the timing of suppression associated with the 1964 FIDS outbreak.

1980 at a frequency ranging from 2.5% to 40.9% (Fig. 6). In comparison, the most recent FIDS outbreak (1984–1986) had the least number of missing rings recorded. Overall, the six CS plantation chronologies recorded a greater total mean frequency of missing rings during the 15 years of the five identified FIDS outbreaks (mean: 15.9% ± 25.2 SD; n = 15) compared to the three CH plantation chronologies (mean: 10.8% ± 19.5 SD; n = 15) and the six CP plantation chronologies (mean: 0.6% ± 1.9 SD; n = 15). In contrast to FIDS outbreaks, the 1992–1995 suppression period in scots pine was not associated with a frequency of more than 2.5% (mean:

90

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

Fig. 7. Radial growth of host and non-host tree species, historical jack pine budworm outbreaks and associated climate variables. The vertical dashed lines indicate the onset of jack pine budworm outbreaks identified in the Spruce Woods Provincial Forest using both historical survey reports and radial growth suppression. (A) Mean of May minimum (black line) and maximum (grey line) temperatures. (B) Mean July Canadian drought code (black line) and total June–August precipitation (grey bar). Note that the drought code scale is inverted. (C) Standardized chronologies for white spruce (black), jack pine (dark grey) and scots pine (light grey). Climate data is from the Brandon CDA meteorological station located about 30 km west of the study area.

1.9% ± 1.3 SD; n = 4) missing rings at any plantation during the years of identified suppression (Fig. 6). 3.3. JPBW outbreaks and climate The identified JPBW outbreaks and associated growth suppression in the host species chronologies were compared to May temperatures, summer precipitation and the July drought code derived from the Brandon CDA meteorological station located approximately 30 km from the study area (Fig. 7). In SWPF, JPBW outbreaks corresponded to below average monthly minimum and maximum temperatures during the month of May. Cool May temperatures corresponded to the 1966, 1974 and 1979 outbreaks. Below average May temperatures were also observed in 1954 and in 1983 respectively, two and one year respectively preceding the FIDS records (Fig. 7). JPBW outbreaks also corresponded to pronounced drought (high July drought index value) and June to August total precipitation being below average during the 1966, 1974, 1979 and 1984 outbreaks. The 1956 outbreak was atypical with the closest major drought year being observed in 1951 (Fig. 7). 4. Discussion 4.1. JPBW outbreaks in central Canada In SWPF, JPBW outbreaks were successfully identified by severe radial growth suppressions observed in both host species and corresponded to the FIDS outbreaks of 1956 to 1958, 1966 to 1968, 1974 to 1977, 1979 to 1980 and 1984 to 1986. Prior to the recognition of JPBW and SBW as distinct species, it was noted in the FIDS records for 1937 that SBW was feeding heavily on jack pine in Sandilands Provincial Forest, about 270 km southeast of the SWPF plantations. Volney (1988) also stated that JPBW defoliation occurred from 1936 to 1938 and in 1944 in Manitoba, however these outbreaks were not recorded in SWPF plantations or observed in the tree-ring analysis. Most host trees in the plantations at that time were less than 12 years old and would not be

producing large crops of staminate cones needed to sustain large JPBW populations (Volney, 1988; Rudolph and Laidly, 1990; Sims et al., 1990; Skilling, 1990). These outbreaks were also not detected in the two oldest chronologies of scots pine in the dataset (CH 2-05 and CS 12-16). It may be noted that in SWPF both jack pine and scots pine were not native to this region prior to initial plantations established in 1904 thus the JPBW may not have been present in these initial plantations during reported provincial JPBW outbreaks. The FIDS records indicated that in 1956 and 1957 moderate to severe defoliation was occurring in pine plantations at CH with no damage at CP plantations (Hildahl et al., 1957, 1958). Tree-ring observations concurred with both severe growth suppression and the TRS mainly restricted to the CH plantations. In the province of Manitoba, the 1956 to 1958 JPBW outbreak was mainly restricted to SWPF (Hildahl et al., 1957, 1958, 1959; Brandt and McDowall, 1968; Volney, 1988; Walker, 1990). It was nonetheless reported farther south in northern Minnesota, U.S.A. (Kulman et al., 1963). In 1958, mild winter conditions and severe spring frosts caused mortality of both (i) the overwintering larvae population and (ii) the terminal host tree buds thus limiting the northern extent of this outbreak (Cayford et al., 1959; Brandt and McDowall, 1968; Hildahl et al., 1958). The severe spring frost injuries observed over a large portion of southern Manitoba in 1958 (Cayford et al., 1959) also resulted in a frost ring in numerous host and non-host trees from SWPF. The second outbreak (1966–1968) identified corresponded to FIDS records describing large JPBW populations and severe defoliation throughout central Canada (Manitoba and Saskatchewan) from 1963 to 1967. The FIDS records indicated that by 1964 isolated populations of the JPBW were beginning to increase at CS plantations (Brandt and McDowall, 1968; McDowall et al., 1967). In the SWPF plantations severe JPBW defoliation and major radial growth suppressions peaked in 1966 and 1967. McDowall et al. (1967) noted that by 1966 the JPBW outbreak had spread approximately 270 km from the southwestern corner of the province across into southeast Manitoba. By 1967, provincial JPBW populations were beginning to decline indicating the end of the outbreak (McDowall et al., 1968). Tree-ring data also indicated that JPBW outbreaks occurred from 1974 and 1977 and again from 1979 to 1980. Severe JPBW defoliation was recorded by the FIDS from 1974 to 1976 in SWPF before the major provincial outbreak in 1978 (Patterson et al., 1975; Petty et al., 1976; Campbell and Hildahl, 1977; Volney, 1988). By 1973, JPBW populations were increasing in southwestern Manitoba, specifically at SWPF (Robins et al., 1974). By 1976, JPBW defoliation in SWPF had increased to severe especially at the CS plantations (Patterson et al., 1975; Petty et al., 1976; Campbell and Hildahl, 1977). In 1979, the JPBW had again reached high populations throughout Manitoba (Hiratsuka et al., 1980) corresponding to the most widespread outbreak recorded at SWPF. In central Canada, the most recent JPBW outbreak (1984–1987) has been considered the most severe on record based on the size of forested area affected by severe defoliation within Saskatchewan, Manitoba and Ontario (Moody and Cerezke, 1984, 1985; Volney, 1988; Gross, 1992; Rhainds et al., 2012). During this outbreak, high populations of JPBW were also recorded within Minnesota and Wisconsin (U.S.A.) (Conway et al., 1999b). Despite defoliation being reported in SWPF plantations in 1984 (Moody and Cerezke, 1985) and 1985 (Rhainds et al., 2012) tree-ring data indicated that the 1985 outbreak was not so severe. Since 1987, no JPBW outbreak has been recorded in Manitoba (Pines et al., 1995; Pines, 2008) despite JPBW outbreaks having been documented in Wisconsin and Michigan (U.S.A.) and throughout northwestern Ontario from 1992 to 1997 and from 2004 to 2011 (Gross, 1992;

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

Kouki et al., 1997; Conway et al., 1999b; Radeloff et al., 2000; OMNR, 2012). 4.2. Climate control on JPBW outbreaks In SWPF all identified JPBW outbreaks (to a lesser extent the 1956 one) were associated with reduced summer precipitation (and/or high July drought index) and cool May temperature. Models of future climate conditions for central Canada are predicting increasing temperatures and higher evapotranspiration rates leading to increased drought frequency and severity (Henderson et al., 2002; PaiMazumder et al., 2013). Increased drought stress will presumably stimulate staminate cone production in jack pine trees potentially leading to more frequent or severe JPBW outbreaks (Volney, 1988; Fleming, 2000). Our results challenge the view that increased droughts by itself may lead to more frequent or severe JPBW outbreaks. In SWPF, JPBW outbreaks were associated with below average May temperatures. The development of major JPBW outbreaks appears to be synchronized with the dual occurrence of both dry summer and cool May. This result was unexpected given that attempts to correlate weather and JPBW population dynamics have been rather inconclusive (Nealis, 1995). In JPBW, feeding of early larval stages is almost exclusively in staminate cones. As staminate cones mature and shed their pollen, larvae will move to current-year needles (Nealis, 1990, 1995). It is thus speculated that cool May temperatures may play an important role in JPBW larvae survival by maintaining developing staminate cones and delaying maximum pollen release. Di-Giovanni et al. (1996) established that maximum pollen release in jack pine was attained with the accumulation of 288.58 degree days above 4 °C starting at Julian day 107. Using the same criteria, both minimum and maximum May temperatures in SWPF were significantly correlated to the date of maximum pollen release (respectively 0.837 and 0.859; p < 0.001, n = 99) indicating that during a cool May maximum pollen release is delayed and occurs in early to mid June. May temperatures may thus play an important role in maintaining the synchrony between JPBW larval emergence and host tree phenology including staminate cone development and timing of bud break. Simulated climate warming experiments have shown that the timing of an early spring may play an important role in determining plant–herbivore interactions and outbreak dynamics (Schwartzberg et al., 2014). The importance of May temperature in the development of JPBW outbreak is also supported by the fact that the JPBW outbreaks observed from 1992 to 1997 and 2004 to 2011 (despite being less severe than mid-1980s) throughout northwestern Ontario (Gross, 1992; Kouki et al., 1997; Conway et al., 1999b; Radeloff et al., 2000; OMNR, 2012) did not occur in SWPF or in Manitoba. Only slight increases in JPBW moths captured in monitoring pheromone traps were observed in 1997 and 2002 to 2003 (Pines, 2008). In SWPF, the dry summer/cool May combination may not have been optimal. The climate data for our study area indicated that minimum temperatures in May 2002 were the coldest recorded since 1910. At the other end of the spectrum, extremely late flush and severe spring frost may also be detrimental for larvae survival. 4.3. JPBW outbreak dynamics and forest management The tree-ring data showed that the onset, end date and severity of suppression associated with FIDS outbreaks varied regionally and locally within SWPF. The variability observed ultimately reflects the forest conditions (staminate cone abundance, stand age, and stand density) that affect local JPBW populations (Hodson and Zehngraff, 1946; Kulman et al., 1963; Clancy et al., 1980; Volney, 1988; Mallet and Volney, 1990; McCullough et al.,

91

1996; Kouki et al., 1997; Conway et al., 1999b; Radeloff et al., 2000; Nealis et al., 2003). In SWPF, management practices such as thinning, insecticide spraying and salvage logging were often implemented in response to JPBW outbreaks (Walker, 1990). Thinning operations including the removal of suppressed or damaged trees lowers stand density and has been shown to improve tree vigor and reduce both staminate cone production and damages caused by defoliating insects such as JPBW (Hodson and Zehngraff, 1946; McCullough et al., 1996; Muzika and Liebhold, 2000; Radeloff et al., 2000; Veteli et al., 2006). In Europe, intensive thinning of scots pine stands has also been shown to reduce defoliation damage caused by Lepidoptera and Hymenoptera species (Veteli et al., 2006). In SWPF, the FIDS records indicated high staminate cone production and moderate to severe JPBW defoliation at plantations CH 6-30 and CH 8-30 from 1956 to 1957 (Hildahl et al., 1957, 1958, 1959) corresponding to major radial growth suppression at the CH plantations. During this outbreak, the scots pine plantation (CH 8-30) showed more severe suppression than the jack pine (CH 6-30) and recorded the greatest suppression and frequency of missing rings compared to any other identified outbreak. The FIDS records also indicated that during this outbreak the CH scots pine plantations with trees between the ages of 16 and 35 years with high staminate cone production suffered more damage than jack pine plantations of a similar age (Hildahl et al., 1958; Brandt and McDowall, 1968). In contrast, the oldest CH plantation (CH 2-05) experienced light defoliation during the 1956 outbreak (Hildahl et al., 1957, 1959). The negligible suppression at CH 205 may be related to thinning operations occurring in 1947 (Walker, 1990) ten years prior to the outbreak. Hodson and Zehngraff (1946) demonstrated that thinning could improve tree vigor and reduce staminate cone production for up to 15 years thus limiting the increase of JPBW populations. In the CP and CS regions, the FIDS records indicated only light defoliation (<30%) from 1956 to 1957 (Hildahl et al., 1957, 1958) concurring with the negligible radial growth suppression observed in these plantations. The regional variability observed during the 1966 to 1968 outbreak may again reflect thinning operations that occurred in SWPF. During this outbreak, the tree-ring data and FIDS records indicated that severe radial growth suppression and severe JPBW defoliation beginning in 1965 were restricted to the CS region (McDowall et al., 1967). The lack of thinning in the CS plantations during the 1960s and 1970s (Walker, 1990) may have allowed for the development of larger JPBW populations. The limited spread of this outbreak to other regions in SWPF may also relate to aerial applications of insecticides for JPBW control that were initiated for the first time in SWPF in 1967 following the increase of populations in 1965 (DeBoo and Hildahl, 1967; Brandt and McDowall, 1968). Insecticide control continued throughout the 1970s and by 1975 the insecticide ‘‘Sumithion’’ was considered successful in reducing larvae populations by 70–75% (Petty et al., 1976; Campbell and Hildahl, 1977). Despite insecticide spraying programs, severe radial growth suppressions were observed from 1974 to 1977 and again in 1979 corresponding to FIDS records of high JPBW populations (Patterson et al., 1975; Petty et al., 1976; Campbell and Hildahl, 1977; Hiratsuka et al., 1980). The variations in the onset of suppression from 1973 to 1976 among the CS and CH plantations probably reflected the normal progression of a JPBW outbreak. The production of staminate cones, which are necessary for JPBW larvae survival, will decrease on individual host trees’ following a year of severe defoliation and budworms will migrate to more suitable hosts during spring emergence in subsequent outbreak years (Hodson and Zehngraff, 1946; Kulman et al., 1963; Nealis and Lomic, 1994; Nealis et al., 1997, 2003; Rhainds et al., 2012). At CP plantations, the less severe suppressions may again relate to

92

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

thinning operations occurring from 1970 to 1974 in the region (Walker, 1990). In 1974 and 1975 CP was also aerially sprayed with the Sumithion insecticide, whereas CH and CS were not (Knowles, personal communication). In SWPF, the 1979 outbreak was the most widespread and severe as indicated by maximum missing rings occurring between 1979 and 1980 at all three regions. Spacing and thinning operations were occurring at CP plantations during 1978 and 1979 (Walker, 1990), however, these plantations appeared to suffer the greatest suppression during 1979 compared to the other two regions. The last outbreak identified in SWPF corresponding to FIDS records of moderate to severe defoliation occurred from 1984 to 1986 (Moody and Cerezke, 1985; Rhainds et al., 2012). This outbreak was only identified at one jack pine plantation (CP 9-44) with less severe suppressions also recorded among other jack pine plantations at CP and CS. However, the 1984–1986 outbreak was characterized with the least missing rings and the fewest identified suppressions compared to previous outbreaks which may be an effect of thinning and salvage operations that followed severe JPBW defoliation from 1974 to 1977 and in 1979 (Walker, 1990). Tree-ring data also revealed that scots pine plantations were least affected during this outbreak showing very little suppression compared to the jack pine. 4.4. JPBW tree-ring signature Severe defoliation events have often been associated with treering signatures involving reduced secondary cell wall thickening. For example, deciduous conifers such as eastern larch may produce a light ring (pale latewood ring) in the first year of severe defoliation by the larch sawfly (Pristiphora erichsonii [Hartig]; Harper, 1913; Girardin et al., 2001). In deciduous species, white rings (vessels with reduced cell wall thickening) were associated with severe defoliation by the forest tent caterpillar (Malacosoma disstria Hubner; Hogg et al., 2002; Sutton and Tardif, 2005). In black spruce and jack pine, white earlywood rings (reduced cell wall thickening in earlywood tracheids) were also associated with major defoliation or crown damaging events occurring during the dormant season (Waito et al., 2013). In this study, the first year of severe JPBW defoliation was often associated with the formation of a tree ring with normal earlywood followed by a reduced proportion of latewood cells. This pattern was also noted in other JPBW studies (Kulman et al., 1963; Gross, 1992). During the first year of an outbreak, feeding on the current years’ foliage occurs in the later portion of the growing season thus not affecting the earlywood development (O’Neil, 1962; Ericsson et al., 1980; Gross, 1992; Moody and Amirault, 1992). The latewood production may however be negatively impacted by the loss of current year’s foliage (O’Neil, 1962; Kulman et al., 1963; Gross, 1992; Moody and Amirault, 1992). In the subsequent year, the ratio of earlywood to latewood may also shift strongly towards the latewood (Vaganov et al., 2006). In Minnesota, Kulman et al. (1963) noted that the earlywood in jack pine tree rings was nearly absent in 1957 following severe JPBW defoliation in 1956. This phenomenon was also observed in many tree-ring series from SWPF (not presented) and requires further investigation. In SWPF, the TRS varied in frequency and synchronicity among plantations. Overall it was most common at CS plantations where analysis of the FIDS records suggests that thinning was employed less often compared to CH and CP. While not a perfect index, the TRS proved to be a useful complement to host and non-host comparison using tree-ring data. For example, 75% of the jack pine trees in plantation CH 6-30 displayed the TRS during the 1979 to 1980 JPBW outbreak when radial growth suppressions were not sufficient to be detected by the OUTBREAK program. In addition,

the near absence of the TRS from 1992 to 1995 and in 2006 support the interpretation that the observed radial growth suppressions were not associated with JPBW defoliation despite OUTBREAK criteria being met. The TRS used alone however may not be a satisfactory indicator of JPBW outbreaks. It was observed at a high frequency in 1950 and 1960 with no outbreak equivalent in the FIDS record. Very narrow rings were observed in both the host species and the non-host species during 1951 and 1961 indicating the thin latewood in 1950 and 1960 followed by a narrow ring may reflect other large-scale environmental signals. In pine trees, annual rings with reduced latewood proportion have been related to early growth cessation due to moisture deficits at the end of a growing season (Cregg et al., 1988; Gross, 1992; Peltola et al., 2007; Stahle et al., 2009). Peltola et al. (2007) also noted that reduced latewood production was a result of moisture deficits in scots pine plantations that were not thinned. Narrow rings in 1951 and 1961 in the non-host white spruce chronology from SWPP have been previously associated with drought stress (Chhin et al., 2004). In the Canadian prairies, the 1961 drought was the most severe and extensive single-year drought in the 20th century (Maybank et al., 1995). In southwestern Manitoba, McGinn and Byrant (1999) also noted that between 1955 and 1988 significant droughts occurred in 1957–58, 1960–61, 1965– 66, 1966–67, 1974–75, 1976–77, 1979–82, 1984 and 1987–88 that also correspond to the FIDS JPBW outbreaks. Volney (1988) indicated that the extent of severe JPBW outbreaks in Manitoba and Saskatchewan showed a strong positive correlation with the forest fire cycle, which is an indicator of drought conditions. In a prediction model developed by Clancy et al. (1980) dry conditions during larval development in June and July favour larva survival and lead to the onset of an outbreak in the following year. As jack pine, the major host for JPBW, is a fire regenerating species the relationship between droughts, fire cycles and JPBW outbreaks requires further investigation.

4.5. Other radial growth suppressions in host trees In addition to identifying JPBW outbreaks, the host and nonhost comparison revealed a major suppression period from 1992 to 1995 in numerous scots pine plantations with no equivalent in jack pine plantations. The lack of a JPBW tree-ring signature and missing rings during this period suggests it did not result from JPBW defoliation. From 1986 to 1994, pheromone traps near CS also recorded very low moth captures and annual surveys revealed minimal JPBW defoliation (Pines et al., 1995). It is speculated that this suppression may have resulted from defoliation by the introduced pine sawfly (Diprion similis [Hartig]) that first appeared in Manitoba in 1969 after being introduced to North America in 1914 (Wong and Tidsbury, 1983). In Europe, two very similar diprionid sawfly species, European pine sawfly (Neodiprion sertifer [Geoffr.]) and the large pine sawfly, are among the most important defoliators causing widespread defoliation to scots pine (Lyytikäinen-Saarenmaa, 1999; Lyytikäinen-Saarenmaa and Tomppo, 2002; Lyytikäinen-Saarenmaa et al., 2003; Veteli et al., 2006; Augustaitis, 2007). In Canada, the introduced pine sawfly has shown a preference for its native host, scots pine, as evidenced by defoliation damage that occurred in 1982 at forest plantations 270 km southeast of the SWPF plantations (Wong and Tidsbury, 1983). The sawfly was also reported to feed on jack pine (Wong and Tidsbury, 1983) and it may account for the small suppression observed in 1992 in the jack pine trees growing in SWPF mixed plantations. However, the damage and suppression caused by the introduced pine sawfly in Canada’s jack pine forests requires further research as the range and frequency of outbreaks of its related

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

species in Europe, European pine sawfly, have recently increased due to warming climates (Netherer and Schopf, 2010). Excluding the 1992–1995 suppression, jack pine plantations showed more identified outbreaks than scots pine plantations suggesting the species may be more sensitive to JPBW defoliation. However, during outbreaks the occurrence of missing rings was more common in the scots pine. The FIDS records have also stressed that scots pine was more sensitive to defoliation and suffered greater damage and mortality during JPBW outbreaks (Hildahl et al., 1958; Brandt and McDowall, 1968; Reeks, 1971; Walker, 1990). Furthermore, thinning and salvage logging were frequently done in SWPF following severe JPBW outbreaks (Reeks, 1971; Walker, 1990) precluding strict comparison among host species. In many of the SWPF plantations a one year lag in severe suppression was observed between species especially within the mixed plantations (CS 2-32 and CP 5-46). This indicates that JPBW larvae may first defoliate jack pine and then move to scots pine as the staminate cone abundance on the primary host is reduced. This observed preference for jack pine and more severe response in scots pine merits further research as mixed host-species plantations may affect outbreak dynamics in comparison to natural jack pine forests.

5. Conclusion Five periods of major radial growth suppression associated with JPBW outbreaks were identified in jack pine and scots pine plantations at SWPF. Severe radial growth suppressions beginning in 1956, 1966, 1974, 1979 and 1984 corresponded to documented JPBW outbreaks in SWPF and throughout the unmanaged forests of central Canada. The variability observed among regions and plantations also reflected the impact of documented forest management practices. Based on the onset of radial growth suppression, jack pine was the preferred host but scots pine may be more sensitive to defoliation as evidence by the occurrence of missing rings during outbreaks. This lag in the onset of suppressions in scots pine plantations could suggest that JPBW outbreaks may be more severe or prolong in regions/plantations where both species co-occur. Comparison of the outbreak history and severity with unmanaged forests will be needed to test this hypothesis. Further comparison between unmanaged jack pine forest and managed plantations of non-native host species would help to assess how forest composition may influence the extent of JPBW outbreaks. Interestingly, major radial growth suppression was identified in scots pine plantations from 1992 to 1995 and may be attributable to defoliation from the introduced pine sawfly. Further study is also needed to better quantify the impact of the introduced pine sawfly on host trees within North America. In this study, the host and non-host comparison proved to be a powerful tool for reconstructing past insect outbreaks and for documenting the variability in the severity of JPBW defoliation at the local scale. The occurrence of severe JPBW defoliation also corresponded to the formation of a thin latewood followed by narrow rings and this tree-ring signature proved to be complementary in the identification of JPBW outbreaks. In SWPF, the connection between JPBW outbreaks and both summer drought and cool spring (May temperatures) suggested that spring temperatures play an important role in maintaining the phenological synchrony between the host tree species and the JPBW. While increased drought associated with future climate change have been linked to increased probability of severe JPBW outbreaks, the warmer springs also predicted could partly offset the risk by altering the phenological synchrony between the host and the defoliator. This association between spring temperatures and JPBW outbreaks merits further investigation.

93

Acknowledgments We thank Johanna Hallmann, Nia Perron, Justin Waito and Zabrina Yaremko for their help with field and/or laboratory work. We thank Fiona Ross from Manitoba Conservation for providing access to FIDS records and historical plantation maps. We thank Dr. Martin Girardin from the Canadian Forest Service for providing the Canadian drought code. We also thank the editor and the reviewers for their constructive feedback and comments on earlier manuscripts. We also acknowledge the financial support obtained from the University of Winnipeg and the Natural Sciences and Engineering Research Council (NSERC) of Canada.

References Augustaitis, A., 2007. Pine sawfly (Diprion pini L.) – related changes in scots pine crown defoliation and possibilities of recovery. Pol. J. Environ. Stud. 16, 363– 369. Bogdanski, B.E.C., 2008. Canada’s boreal forest economy: economic and socioeconomic issues and research opportunities. Info. Rep. BC-X-414. Canadian Forest Service, Pacific Forestry Centre, Victoria, BC. 72pp. Brandt, N.R., McDowall, L.L., 1968. Jack pine budworm in Manitoba and Saskatchewan. Info. Rep. MS-X-8. Forest Research Laboratory, Winnipeg, MB. 26pp. Brooks, J.R., Flanagan, L.B., Ehleringer, J.R., 1998. Responses of boreal conifers to climate fluctuations: indications from tree-ring widths and carbon isotope analyses. Can. J. For. Res. 28, 524–533. Brown, A.W.A., MacKay, M.R., 1943. The jack pine budworm and the spruce budworm, Cacoecia fumiferana Clem. (Tortricidae). Can. Entomol. 75, 207–211. Campbell, A.E., Hildahl, V., 1977. Summary of forest insect and disease conditions in Manitoba, 1976. Info. Rep. NOR-X-183. Canadian Forest Service, Northern Forestry Research Centre, Edmonton, AB, 29pp. Cayford, J.H., Hildahl, V., Nairn, L.D., Wheaton, M.P.H., 1959. Injury to trees from winter drying and frost in Manitoba and Saskatchewan in 1958. Forest Chron. 35, 282–290. Chhin, S., Wang, G., Tardif, J., 2004. Dendroclimatic analysis of white spruce at its southern limit of distribution in the Spruce Woods Provincial Park, Manitoba, Canada. Tree-Ring Res. 60, 31–43. Clancy, K.M., Giese, R.L., Benjamin, D.M., 1980. Predicting jack-pine budworm infestations in northwestern Wisconsin. Environ. Entomol. 9, 743–751. Conway, B.E., Leefers, L.A., McCullough, D.G., 1999a. Yield and financial losses associated with a jack pine budworm outbreak in Michigan and the implications for management. Can. J. For. Res. 29, 382–392. Conway, B.E., McCullough, D.G., Leefers, L.A., 1999b. Long-term effects of jack pine budworm outbreaks on the growth of jack pine trees in Michigan. Can. J. For. Res. 29, 1510–1517. Cook, E.R., Holmes, R.L., 1999. User’s manual for program ARSTAN. Laboratory of Tree-Ring Research. University of Arizona, Tucson, AZ. 13pp. Cregg, B.M., Dougherty, P.M., Hennessey, T.C., 1988. Growth and wood quality of young loblolly pine trees in relation to stand density and climatic factors. Can. J. For. Res. 18, 851–858. DeBoo, R.F., Hildahl, V., 1967. Aerial spraying for control of the jack pine budworm in Manitoba. Man. Entomol. 1, 21–26. Di-Giovanni, F., Kevan, P.G., Caron, G.-É., 1996. Estimating the timing of maximum pollen release from jack pine (Pinus banksiana Lamb.) in northern Ontario. For. Chron. 72, 166–169. Environment Canada, 2014. Canadian climate normals 1981–2010 station data, Brandon CDA. . Government of Canada, Ottawa, ON (accessed on 01.03.14). Ericsson, A., Larsson, S., Tenow, O., 1980. Effects of early and late season defoliation on growth and carbohydrate dynamics in scots pine. J. Appl. Ecol. 17, 747–769. Fleming, R.A., 2000. Climate change and insect disturbance regimes in Canada’s boreal forests. World Resour. Rev. 12, 521–555. Freeman, T.N., 1953. The spruce budworm, Choristoneura fumiferana (Clem.) and an allied new species on pine (Lepidoptera: Tortricidae). Can. Entomol. 85, 121– 127. Genries, A., Drobyshev, I., Bergeron, Y., 2012. Growth-climate response of jack pine on clay soils in northeastern Canada. Dendrochronologia 30, 127–136. Girardin, M.-P., Tardif, J., 2005. Sensitivity of tree growth to the atmospheric vertical profile in the boreal plains of Manitoba, Canada. Can. J. For. Res. 35, 48–64. Girardin, M.-P., Tardif, J., Bergeron, Y., 2001. Radial growth analysis of Larix laricina from the Lake Duparquet area, Québec, in relation to climate and larch sawfly outbreaks. Écoscience 8, 127–138. Girardin, M.-P., Tardif, J.C., Epp, B., Conciatori, F., 2009. Frequency of cool summers in interior North America over the past three centuries [online]. Geophys. Res. Lett. 36, L07705. http://dx.doi.org/10.1029/2009GL037242. Grandmaison, M., 1991. Forest insect and disease conditions in Manitoba in 1990. File Report. Forestry, Canada, Forest Insect and Disease Survey, 31pp.

94

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95

Grandmaison, M., 1993. Forest insect and disease conditions in Manitoba in 1992. File report. Forestry Canada, Forest Insect and Disease Survey. 25pp. Gross, H.L., 1992. Impact analysis for a jack pine budworm infestation in Ontario. Can. J. For. Res. 22, 818–831. Haack, R.A., Byler, J.W., 1993. Insects and pathogens: regulators of forest ecosystems. J. Forest. 91, 32–37. Henderson, N., Hogg, E., Barrow, E., Dolter, B., 2002. Climate change impacts on the island forests of the great plains and the implications for nature conservation policy: the outlook for Sweet Grass Hills (Montana), Cypress Hills (Alberta– Saskatchewan), Moose Mountain (Saskatchewan), Spruce Woods (Manitoba) and Turtle Mountain (Manitoba – North Dakota). Final Rev. Rep. Prairie Adaptation and Research Collaborative, Regina, SK. 116pp. Hildahl, V., McDowall, L.L., Campbell, A.E., Mortensen, K.L., Lawrence, J.J., Martin, J.B., Pratt, M.R., Drouin, J.A., McLeod, B.B., Machuk, A., Lalor, G.T., 1957. Annual reports of forest biology rangers, Manitoba and Saskatchewan, 1956. Interim Rep. 1956-1. Canada Department of Agriculture, Forest Biology Division, Winnipeg, MB. 270pp. Hildahl, V., McDowall, L.L., Campbell, A.E., Mortensen, K. L., Lawrence, J.J., Martin, J.B., Pratt, M.R., Drouin, J.A., McLeod, B.B., Machuk, A., Lalor, G.T., 1958. Annual reports of forest biology rangers, Manitoba and Saskatchewan, 1957. Interim Rep. 1957-3. Canada Department of Agriculture, Forest Biology Division, Winnipeg, MB. 72pp. Hildahl, V., McDowall, L.L., Campbell, A.E., Pratt, M.R., Mortensen, K., Lawrence, J.J., Martin, J.B., Machuk, A., Drouin, J.A., McLeod, B.B., Lalor, G.T., 1959. Annual reports of the forest biology rangers, Manitoba and Saskatchewan, 1958. Interim Rep. 1958-6. Canada Department of Agriculture, Forest Biology Division, Winnipeg, MB. 159pp. Hiratsuka, Y., Cerezke, H.F., Petty, J., 1980. Forest insect and disease conditions in Alberta, Saskatchewan, Manitoba and the Northwest Territories in 1979 and predictions for 1980. Info. Rep. Nor-X-225. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 17pp. Hodson, A.C., Zehngraff, P.J., 1946. Budworm control in jack pine forest management. J. Forest. 44, 198–200. Hoffer, M., Tardif, J.C., 2009. False rings in jack pine and black spruce trees from eastern Manitoba as indicators of dry summers. Can. J. For. Res. 39, 1722– 1736. Hofgaard, A., Tardif, J., Bergeron, Y., 1999. Dendroclimatic response of Picea mariana and Pinus banksiana along a latitudinal gradient in the eastern Canadian boreal forest. Can. J. For. Res. 29, 1333–1346. Hogg, E.H., Hart, M., Lieffers, V.J., 2002. White tree rings formed in trembling aspen saplings following experimental defoliation. Can. J. For. Res. 32, 1929–1934. Holmes, R.L., 1983. Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bull. 43, 69–78. Holmes, R.L., Swetnam, T.W., 1996. Program OUTBREAK Users Manual. Laboratory of Tree-Ring Research, University of Arizona, Tucson, AZ. 6pp. Huang, J., Bergeron, Y., Berninger, F., Zhai, L., Tardif, J., Denneler, B., 2013. Impact of future climate on radial growth of four major boreal tree species in the eastern Canadian boreal forest. PlosOne. http://dx.doi.org/10.1371/journal. pone.0056758. Hughes, J.S., Fortin, M.-J., Nealis, V., Régnière, J., 2014. Pollen cone production in jack pine: spatial and temporal patterns subject to natural disturbance by the jack pine budworm. Can. J. For. Res. 44, 195–211. Husch, B., Beers, T.W., Kershaw, J.A., 2003. Forest Mensuration, fourth ed. John Wiley and Sons, Inc., Hoboken, NJ, 443pp. Ives, W.G.H., 1981. Environmental factors affecting 21 forest insect defoliators in Manitoba and Saskatchewan, 1945–69. Info. Rep. NOR-X-233. Canadian Forestry Service, Northern Forestry Research Centre, Edmonton, AB. 146pp. Kouki, J., McCullough, D.G., Marshall, L.D., 1997. Effect of forest stand and edge characteristics on the vulnerability of jack pine stands to jack pine budworm (Choristoneura pinus pinus) damage. Can. J. For. Res. 27, 1765–1772. Kulman, H.M., 1971. Effects of insect defoliation on growth and mortality of trees. Annu. Rev. Entomol. 16, 289–324. Kulman, H.M., Hodson, A.C., Duncan, D.P., 1963. Distribution and effects of jack-pine budworm defoliation. For. Sci. 9, 146–157. Lysyk, T.J., 1989. A multiple-cohort model for simulating jack pine budworm (Lepidoptera: Tortricidae) development under variable temperature conditions. Can. Entomol. 121, 373–387. Lyytikäinen-Saarenmaa, P., 1999. The responses of scots pine, Pinus sylvestris, to natural and artificial defoliation stress. Ecol. Appl. 9, 469–474. Lyytikäinen-Saarenmaa, P., Tomppo, E., 2002. Impact of sawfly defoliation on growth of scots pine Pinus sylvestris (Pinaceae) and associated economic loss. Bull. Entomol. Res. 92, 137–140. Lyytikäinen-Saarenmaa, P., Niemelä, P., Annila, E., 2003. Growth responses and mortality of scots pine (Pinus sylvestris L.) after a pine sawfly outbreak. In: Kamata, N., Liebhold, N.M., Quiring, D.T., Clancy, K.M., (Eds.), Proceedings of the IUFRO Kanazawa 2003 International Symposium: Forest insect population dynamics and host influences, Kanazawa, Japan, 14–19 September 2003, Kanazawa University, Japan, pp. 81–85. Mallet, K.I., Volney, W.J.A., 1990. Relationships among jack pine budworm damage, selected tree characteristics and Armillaria root rot in jack pine. Can. J. For. Res. 20, 1791–1795. Maybank, J., Bonsal, B., Jones, K., Lawford, R.G., O’Brien, E.G., Ripley, E.A., Wheaton, E., 1995. Drought as a natural disaster. Atmos.-Ocean. 33, 195–222. McCullough, D.G., 2000. A review of factors affecting the population dynamics of jack pine budworm (Choristoneura pinus pinus Freeman). Pop. Ecol. 42, 243– 256.

McCullough, D.G., Marshall, L.D., Buss, L.J., Kouki, J., 1996. Relating jack pine budworm damage to stand inventory variables in northern Michigan. Can. J. For. Res. 26, 2180–2190. McCullough, D.G., Werner, R.A., Neumann, D., 1998. Fire and insects in northern and boreal forest ecosystems of North America. Annu. Rev. Entomol. 43, 107–127. McDowall, L.L., Campbell, A.E., Mortensen, K.L., Arthurs, J.W., Tidsbury, R.C., Devlin, R.J., McLeod, B.B., Crawford, W.B., Rentz, C.L., Still, G.N., Homann, M.J., 1967. Annual district reports: Forest insect and disease survey; ManitobaSaskatchewan region, 1966. Info. Rep. MS-X-4. Canadian Department of Forestry, Forest Research Laboratory. Winnipeg, MB. 166pp. McDowall, L.L., Mortensen, K.L., Still, G.N., Shepherd, D.N., Tidsbury, R.C., Crawford, W.B., McLoed, B.B., Beveridge, W.J., Rentz, C.L., Simpson, A.N., Homann, M.J., Anderson, P.W., 1968. Annual district reports: Forest insect and disease survey; Manitoba-Saskatchewan region, 1967. Info. Rep. MS-X-7. Canadian Department of Forestry, Forest Research Laboratory. Winnipeg, MB. 148pp. McGinn, R.A., Byrant, G.J., 1999. The drought severity index and the recollection of drought by agriculturalists in the Palliser Triangle, southwestern Manitoba. Prairie Persp. 2, 39–54. Manitoba Conservation and Water Stewardship (MCWS), 2013. Manitoba’s provincial forests [online]. . Manitoba Conservation and Water Stewardship: Forestry Branch (accessed on 25.09.13). Mitchell, K., 2007. Quantitative analysis by the point-centered quarter method. Dept. of Math. and Comp. Sci. Hobart and William Smith Colleges, Geneva, NY, 34pp. Moody, B.H., Cerezke H.F., 1984. Forest insect and disease conditions in Alberta, Saskatchewan, Manitoba, and the Northwest Territories in 1983 and predictions for 1984. Info. Rep. NOR-X-261. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 20pp. Moody, B.H., Amirault, P.A., 1992. Impacts of major pests on forest growth and yield in the praririe provinces and the Northwest Territories: a literature review. Info Rep. NOR-X-324. Forestry Canada, Northwest region. Northern Forestry Centre, Edmonton, AB. 41pp. Moody, B.H., Cerezke, H.F., 1985. Forest insect and disease conditions in Alberta, Saskatchewan, Manitoba, and the Northwest Territories in 1984 and predictions for 1985. Info. Rep. NOR-X-269. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 21pp. Muzika, R.M., Liebhold, A.M., 2000. A critique of silviculture approaches to managing defoliating insects in North America. Agric. For. Entomol. 2, 97–105. Nealis, V.G., 1990. Jack pine budworm populations and staminate flowers. Can. J. For. Res. 20, 1253–1255. Nealis, V.G., 1991. Parasitism in sustained and collapsing populations of the jack pine budworm, Choristoneura pinus pinus Free. (Lepidoptera: Tortricidae) in Ontario, 1985–1987. Can. Entomol. 123, 1065–1075. Nealis, V.G., 1995. Population biology of the jack pine budworm. In: Volney, W.J.A., Nealis, V.G., Howse, G.M., Westwood, A.R., McCullough, D.R., Laishley, B.L., (Eds.), Proceedings of the jack pine budworm symposium: Jack pine budworm biology and management, Info. Rep. NOR-X-342, Winnipeg, MB, 24–26 January 1995, pp. 55–72. Nealis, V.G., Lomic, P.V., 1994. Host-plant influence on the population ecology of the jack pine budworm, Choristoneura pinus (Lepidoptera: Tortricidae). Ecol. Entomol. 19, 367–373. Nealis, V.G., Lomic, P.V., Meating, J.H., 1997. Forecasting defoliation by the jack pine budworm. Can. J. For. Res. 27, 1154–1158. Nealis, V.G., Magnussen, S., Hopkin, A.A., 2003. A lagged density–dependant relationship between jack pine budworm Choristoneura pinus pinus and its host tree Pinus banksiana. Ecol. Entomol. 28, 183–192. Netherer, S., Schopf, A., 2010. Potential effects of climate change on insect herbivores in European forests: general aspects and the pine processionary moth as specific example. For. Ecol. Manage. 259, 831–838. Ontario Ministry of Natural Resources (OMNR), 2012. Forest health conditions in Ontario, 2011. In: Scarr, T.A., Ryall, K.L., Hodge, P., (Eds.), Natural Resources Canada, 109pp. O’Neil, L.C., 1962. Some effects of artificial defoliation on the growth of jack pine (Pinus banksiana Lamb.). Can. J. Bot. 40, 273–280. PaiMazumder, D., Sushama, L., Laprise, R., Khaliq, M.N., Sauchyn, D., 2013. Canadian RCM projected changes to short- and long-term drought characteristics over the Canadian Prairies. Int. J. Climatol. 33, 1409–1423. Patterson, V.B., Still, G.N., Emond, F.G., Gautreau, E.J., Tidsbury, R.C., Petty, J., Smith, G.J., Caltrell, R.M., Susut, J.P., 1975. Annual district reports: Forest insect and disease survey; Prairies region, 1974. Info. Rep. NOR-X-125. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 60pp. Peltola, H., Kilpeläinen, A., Sauvala, K., Räisänen, T., Ikonen, V.-P., 2007. Effects of early thinning regime and tree status on the radial growth and wood density of scots pine. Silva Fenn. 41, 489–505. Petty, J., Caltrell, R.M., Campbell, A.E., Emond, F.J., Hildahl, V., Patterson, V.B., Still, G.N., Susut, J.P., Tidsbury, R.C., 1976. Annual district reports: Forest insect and disease survey; Prairies region, 1975. Info. Rep. NOR-X-154. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 33pp. Pines, I.L., 2008. Forest pests in Manitoba. Forest Health and Renewal, Manitoba Conservation, National Forest Pest Management Forum. Available at: . (accessed on 25.09.13). Pines, I.L., Knowles, K., Westwood, A.R., 1995. Monitoring jack pine budworm in Manitoba with pheromone-baited traps and reference to pollen cone levels. In: Volney, W.J.A., Nealis, V.G., Howse, G.M., Westwood, A.R., McCullough, D.R., Laishley, B.L., (Eds.), Proceedings of the jack pine budworm symposium: Jack

J.R.M. Robson et al. / Forest Ecology and Management 347 (2015) 83–95 pine budworm biology and management, Info. Rep. NOR-X-342, Winnipeg, MB., 24–26 January 1995. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. pp. 89–96. Pisaric, M.F.J., St-Onge, S.M., Kokelj, S.V., 2009. Tree-ring reconstruction of earlygrowing season precipitation from Yellowknife, Northwest Territories, Canada. Arct. Antarct. Alp. Res. 41, 486–496. Radeloff, V.C., Mladenoff, D.J., Boyce, M.S., 2000. The changing relation of landscape patterns and jack pine budworm populations during an outbreak. Oikos 90, 417–430. Reeks, W.A., 1971. Impact of insect outbreaks in Manitoba. Man. Entomol. 5, 5–16. Rhainds, M., Pines, I., Thurston, G., 2012. Anatomy of an outbreak of jack pine budworm, Choristoneura pinus pinus (Lepidoptera: Tortricidae). Appl. Entomol. Zool. 47, 119–123. Robins, J.K., Patterson, V.B., Still, G.N., Emond, F.J., Gautreau, E.J., Tidsbury, R.C., Petty, J., Smith, G.J., Caltrell, R.M., Susut, J.P., 1974. Annual district reports: Forest insect and disease survey; Prairies region, 1973. Info. Rep. Nor-X-73. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 58pp. Rudolph, T.D., Laidly, P.R., 1990. Pinus banksiana Lamb. Jack pine. In: Burns, R.M., Honkala, B.H., (Eds.), Silvics of North America, Conifers, vol. 1. Agric Handb. 654. U.S. Dep. Agric. Washington, DC. pp. 280–293. Schwartzberg, E.G., Jamieson, M.A., Raffa, K.F., Reich, P.B., Montgomery, R.A., Lindroth, R.L., 2014. Simulated climate warming alters phenological synchrony between an outbreak insect herbivore and host trees. Oecologia 175, 1041–1049. Schweingruber, F.H., 2007. Wood Structure and Environment. Springer-Verlag, Berlin Heidelberg, Germany, 279pp. Sims, R.A., Kershaw, H.M., Wickware, G.M., 1990. The autecology of major tree species in the north central region of Ontario. Canada-Ontario Forest Resource Development Agreement Rep. 3302. Ontario Ministry of Natural Resources, Ottawa, ON. 126pp. Skilling, D.D., 1990. Pinus sylvestris L. Scotch pine. In: Burns, R.M., Honkala, B.H., (Eds.), Silvics of North America, Conifers, vol. 1. Agric. Handb. 654. U.S. Dep. Agric. Washington, DC. pp. 489–496. Smith, R.E., Veldhius, H., Mills, G.F., Eilers, R.G., Fraser, W.R., Lelyk, G.W., 1998. Terrestrial ecozones, ecoregions, and ecodistricts of Manitoba: an ecological stratification of Manitoba’s natural landscapes. Bulletin 1998-9E. Agriculture and Agri-food Canada Land Resource Unit. Brandon Research Centre, Brandon, MB. Tech. 319pp.

95

Speer, J.H., 2010. Fundamentals of Tree-ring Research. The University of Arizona Press, Tucson, AZ, 333pp. Stahle, D.W., Cleaveland, M.K., Grissino-Mayer, H.D., Griffin, R.D., Fye, F.K., Therrell, M.D., Burnette, D.J., Meko, D.M., Villanueva Diaz, J., 2009. Cool- and warmseason precipitation reconstructions over western New Mexico. J. Clim. 22, 3729–3750. Sutton, A., Tardif, J.C., 2005. Distribution and anatomical characteristics of white rings in Populus tremuloides. IAWA J. 26, 221–238. Swetnam, T.W., Thompson, M.A., Sutherland, E.K., 1985. Using dendrochronology to measure radial growth of defoliated trees. Agric. Handb. 639. U.S. Department of Agriculture, Forest Service, Tucson, AZ. 39pp. Tardif, J.C., Girardin, M.P., Conciatori, F., 2011. Light rings as bioindicators of climate change in interior North America. Glob. Planet. Change. 79, 134–144. Vaganov, E.A., Hughes, M.K., Shashkin, A.V., 2006. Growth dynamics of conifer tree rings: Images of past and future environments. Ecological Studies, vol. 183. Spring-Verlag, Berlin Heidelberg, Germany, p. 354. Veteli, T.O., Koricheva, J., Niemelä, P., Kellomäki, S., 2006. Effects of forest management on the abundance of insect pests on scots pine. For. Ecol. Manage. 231, 214–217. Volney, W.J.A., 1988. Analysis of historic jack pine budworm outbreaks in the prairie provinces of Canada. Can. J. For. Res. 18, 1152–1158. Volney, W.J.A., 1998. Ten-year tree mortality following a jack pine budworm outbreak in Saskatchewan. Can. J. For. Res. 28, 1784–1793. Volney, W.J.A., Fleming, R.A., 2000. Climate change and impacts of boreal forest insects. Agric. Ecosyst. Environ. 82, 283–294. Volney, W.J.A., Fleming, R.A., 2007. Spruce budworm (Choristoneura spp.) biotype reactions to forest and climate characteristics. Glob. Change Biol. 13, 1630– 1643. Volney, W.J.A., Mallet, K.I., 1992. Light rings and the age of jack pine trees. Can. J. For. Res. 22, 2011–2013. Volney, W.J.A., McCullough, D.G., 1994. Jack pine budworm population behaviour in northwestern Wisconsin. Can. J. For. Res. 24, 502–510. Waito, J., Conciatori, F., Tardif, J.C., 2013. Frost rings and white earlywood rings in Picea mariana trees from the boreal plains, central Canada. IAWA J. 34, 71–87. Walker, N.R., 1990. Planting conifers in the Spruce Woods, 1904–1979. For. Manage. Demonstr. Note 14. Manitoba Conservation, Winnipeg, MB. 28pp. Wong, H.R., Tidsbury, R.C., 1983. Introduced pine sawfly in Manitoba. Forest Management Note 26. Canadian Forestry Service, Northern Forest Research Centre. Edmonton, AB. 2pp.