The Southern Pine Beetle

Chapter 4: Climatic, Site, and Stand Factors

Ray R. Hicks, Jr. – Associate Professor, Division of Forestry, West Virginia University, Morgantown, W. VA.

Introduction

Host-pest relationships are complex ecological phenomena that, in undisturbed systems, ultimately achieve a balance. The pest must be able to thrive and reproduce; but, equally important, it must permit some members of the host species to survive. This relationship is especially important when a highly specific association exists, such as that between the southern pine beetle and four or five species of southern pines.

The stands of pines that grow in the South today probably bear little resemblance to the natural forests that existed throughout most of the coevolution of southern pines and the SPB. During recent decades, SPB populations have fluctuated erratically, with an apparent trend toward increasing severity of damage (Hedden 1978b). The magnitude of fluctuations suggests that the system is in a state of imbalance, perhaps due to the abundance of host species. Given time and no interference from people, the system would reestablish an equilibrium through reduction in the pine component of forests. But this process would be counter to the goals of forest managers, many of whom want to achieve maximum productivity of softwood timber from southern forests. Three immediate questions arise: (1) Are there quantifiable causes responsible for the increase in SPB activity? (2) Can we manipulate these factors to achieve the goal of increased forest productivity by reducing SPB-induced losses? (3) Is it economically feasible to perform such manipulations? The site-stand-climatic investigators funded by the Expanded Southern Pine Beetle Research and Applications Program (ESPBRAP) have tried to answer the first question. Their results and those of other relevant studies are summarized here.

Researchers from Virginia to Texas took measurements on environmental factors affecting beetle-attacked and nonattacked (baseline) stands to determine what, if any, differences existed. These factors ranged from measures of tree competition and vigor to soil and site factors, including several climatic and disturbance characteristics. Prior research dealing with host-tree susceptibility to bark beetles had elucidated numerous relationships; on this foundation ESPBRAP initiated its research.

Site and stand factors affect several forest insect pests. For example, attacks by western spruce budworm (Choristoneura fumiferana Clem.) are found to be associated with high stand densities in some years but not in others (Fellin 1976). Research indicates that physiographic location, tree diameter, stand basal area, and species composition all play a role in susceptibility to spruce beetle, D. rufipennis (Kirby) (Schmid and Frye 1976). Logging residues apparently serve as reservoirs for spruce beetle brood (Schmid 1977), and logging practices that remove these residues lead to reduced infestation of residual stands (Beckwith, Wolff, and Zasada 1977).

For fir engraver, Scolytus ventralis Lec., competition among trees has been implicated as a prime factor in susceptibility (Schenk et al. 1977), as well as species diversity. In another study concerning susceptibility of white fir (Abies concolor [Gord. and Glen.] Lindl.) to fir engraver, induced water stress in trees permitted the beetles to enter more easily. Nonstressed trees were able to "pitch out" the beetles (Ferrell 1978).

Mountain pine beetle (D. ponderosae Hopkins), a species related to the southern pine beetle, has been the subject of numerous investigations on host susceptibility. Susceptibility of lodgepole pine (P. contorta var. latifolia Engelm.) to this beetle is largely a function of tree age, diameter, and phloem thickness. Beetles prefer larger older trees with thick phloem (Amman and Pace 1976, Amman et al. 1977, Berryman 1976, Cole and Cahill 1976). Mountain pine beetle also attacks ponderosa pine (P. ponderosa), and its susceptibility to this insect is related to excessive intertree competition. Investigators recommend thinning of overstocked stands as a means of reducing susceptibility (Sartwell 1971, Sartwell and Stevens 1975, Sartwell and Dolph 1976). These host and stand variables are proposed as important components of an integrated pest management system for mountain pine beetle (Stage and Long 1977).

Recent research has shown that many of the same site and stand factors affect susceptibility of southern pines to SPB. High stand density characterizes many infestations (Lorio and Bennett 1974), and this overstocking apparently leads to reduced radial growth (Bennett 1971; Coulson, Hain and Payne 1974). Trees growing on wetter sites seem to be more susceptible due to declining vigor (Bennett 1968, Lorio 1968). Stand disturbances such as lightning can trigger SPB infestations (Hodges and Pickard 1971, Lorio and Yandle 1978).

Obviously site, stand, and climatic factors play a significant role in the SPB-host relationship. Therefore, researchers in the ESPBRAP site-stand area have tried to quantify that relationship so that another piece of the complex model can be put into place and a truly integrated protection program for SPB can be developed, as Coster (1977) suggests. To do this, investigators established plots throughout the Southeast and took data on site, stand, and climatic variables. For comparison, most projects collected data on uninfested plots, which were located in an unbiased manner. The bulk of data was from the West Gulf Coastal Plain, with fewer plots from the Piedmont, North Georgia Mountains, and Atlantic Coastal Plain.

In an effort to summarize the contributions of this ESPBRAP coordinated regional project, and to incorporate appropriate background literature, this chapter discusses the effects of climatic, site, and stand factors on host susceptibility to SPB. ESPBRAP has also issued a Technical Bulletin to preserve basic statistics from the coordinated regional site-stand project (Coster and Searcy 1980).

Climatic Factors

Rainfall

The amount and timing of rainfall probably affect southern pine beetle activity, but we do not know how. For instance, moisture could directly affect the survival and vigor of adult or brood beetles, or it could indirectly affect beetle populations by altering host tree resistance. Several studies have investigated the relationships of tree water balance to successful beetle attack. Others have dealt with the relationship of rainfall to areawide fluctuations in SPB populations (Craighead 1925). The implication is that rainfall affects tree water stress, which in turn affects resistance to attack; the latter affects the population of beetles. Unfortunately, experimental data concerning this chain of events is lacking.

Studies dealing specifically with internal tree water balance and southern pine beetle attack have produced interesting results. Lorio and Hodges (1968) found that oleoresin exudation pressure of large loblolly pines was reduced during periods of drought-induced moisture stress. They contend that the beetles are more successful in attacking trees with lower oleoresin exudation pressures since the beetles are less likely to be pitched out of such trees. Indeed, they found that artificially stressed trees succumbed to induced beetle attack more readily than nonstressed trees (Lorio and Hodges 1977).

Obviously, a relationship exists between host tree resistance and oleoresin exudation pressure as affected by internal water balance. This relationship may in part account for SPB population fluctuations ostensibly related to rainfall. Several workers have investigated the relationship between rainfall and changes in SPB population. The number of beetle spots in an area correlates to some degree with rainfall in previous months. In east Texas, rainfall for the previous summer, fall, and spring was associated with the number of beetle spots in the following summer (Kroll and Reeves 1978). Abundant rain in the previous summer was conducive to more spots in the following year, while previous fall and spring precipitation was negatively correlated with current year infestations. This procedure needs additional refinement before being used. The Texas Forest Service (1978) reported a similar finding regarding previous summer rainfall. However, these data do not substantiate the hypothesis that lowered host resistance due to water stress causes SPB population increases. On the contrary, one would expect low rainfall during the summer growing season to be associated with reduced host resistance and greater beetle activity. Since stress-related host susceptibility would operate for short periods of time during droughts, it may be difficult to show water stress relationships where historical weather records reporting monthly averages are used to predict current or future trends in beetle activity.

At Virginia Polytechnic Institute and State University, researchers studied the associations between several climatic variables and monthly SPB spots per 1,000 acres in Arkansas and North Carolina. They found that precipitation 2 and 4 months prior to the month in question was weakly associated with SPB activity. But in both cases, lower precipitation was associated with higher beetle activity. Since these researchers were predicting SPB activity for the summer months, their lagged precipitation data would be mostly for the winter and spring months and therefore consistent with the negative relationship for spring rainfall found by Kroll and Reeves (1978).

Rainfall almost certainly has an effect on host resistance. But quantifying this effect is difficult because of continuous variations in the host, the beetle, and the weather. These variations account in part for the yearly and regional differences reported concerning the effect of rainfall on SPB populations. For example, King (1972) compared epidemic and nonepidemic years and found low summer rainfall in Georgia, high winter rainfall in Texas, and high spring coupled with low early summer rainfall in the Carolinas to be associated with epidemic years. Hansen, Baker, and Barry (1973) found that outbreaks in the Atlantic Coastal Plain of Delaware, Maryland, and Virginia are associated with extended drought periods. Conversely, Kalkstein (1974) found that SPB activity in Texas and Louisiana was associated with increased late winter moisture.

Although these results are somewhat confusing, it seems that infestations in the Western Gulf Coastal Plain are frequently associated with prior periods of abundant or superabundant rainfall. Infestations along the Atlantic Coastal Plains, however, are frequently associated with prior periods of drought. The Western Gulf relationship seems consistent with findings of Lorio (1968) and studies in Texas and Louisiana, where infestations are frequently found on low-lying or wet sites. On such sites, excessive rainfall tends to accentuate the poor drainage problem. In the Atlantic Coastal Plain, beetle attacks may be more often related to drought stress, as evidenced by their occurrence following dry years.

Temperature

The primary effect of temperature seems directly related to insect survival, but a potential effect on host resistance is related to tree water balance. Kalkstein (1976) found that potential evapotranspiration of trees was a useful variable in predicting beetle population trends in Louisiana and Texas. Evapotranspiration potential relates to several environmental variables, including moisture supply and temperature. This evidence for the involvement of temperature on host resistance is circumstantial at best. Campbell and Smith (1978) did not detect a similar relationship of potential evapotranspiration in an Arkansas study.

Weather-Related Stand Disturbances

Previous research had elucidated the associations of several stand disturbances with initiation of southern pine beetle infestations (Lorio and Bennett 1974, Lorio and Yandle 1978, Hodges and Pickard 1971). Several of these are weather related (e.g., lightning strikes and wind, ice, and hail damage). Investigators in the ESPBRAP site-stand group also recorded the occurrence of these disturbances and others at each site visited. ESPBRAP data for weather-related disturbances generally show a marked association of certain disturbances with SPB occurrence (table 4-1). Most notable is the occurrence of lightning strikes on beetle-infested plots, where 10 to 40 percent were found to have this disturbance, in comparison to less than 1 percent for randomly located, nonattacked (baseline) plots.

Apparently these stand disturbances, which weaken trees, favor initial beetle attack. Such trees form epicenters for beetle spots, and the infestation may enlarge depending on the availability of SPB populations and other environmental conditions prevailing at the time of attack. For example, beetles attacking a lone lightning-struck tree may not successfully initiate an active spot if the surrounding trees are vigorous and not stressed. Overstocking of pine stands seems to be a prime factor in promoting such spot growth (Hedden and Billings 1979).

Site Factors

Landform

Investigators in the site-stand group classified sites according to landform categories outlined in the U.S. Forest Service’s Soils Resource Guide: Southern Region (U.S. Department of Agriculture Forest Service 1972a). Most of the data were from the Coastal Plain from Mississippi to Texas (Rowell 1978 unpublished). Here a greater-than-expected frequency of infestations was found on low-lying landforms such as swamps, flood plains, stream terraces, bays, and lower slops. SPB infestations occurred on 27.6 percent of these landforms but on only 17.6 percent of the baseline plots. The most common landform for both infested and baseline plots was upland flat (45.2 percent and 39.6 percent, respectively), and the data indicate that beetles were slightly more frequent on upland flat sites. Likewise, beetles seemed to prefer trees growing on ridge sites (14 percent infested v. 5.7 percent baseline). The only sites showing a lower-than-expected frequency of attack were side slopes and steep side slopes (13.2 percent attacked v. 37.4 percent baseline).

Stress seems to be the key to the beetles’ preference for attacking trees on the higher upland and lowland landform categories. That is, excessive moisture in low-lying areas can induce root damage and stress, while trees growing on ridge tops are prone to drought stress during dry periods. Trees growing on sloping sites (where attacks were less frequent than expected) likely avoid either of these extremes.

Limited data were collected for the Georgia Piedmont and Mountains (Belanger, Hatchell, and Moore 1977; Belanger, Osgood, and Hatchell 1979b). Both differed considerably from the Gulf Coastal Plain in the relationships of landform to beetle attack preference. In the Piedmont steep side slopes accounted for 60.7 percent of the infestations but only 38.6 percent of the baseline plots. Conversely, ridge sites had a 32 percent attack frequency, compared to almost 50 percent of the baseline plots in this category. Beetles do not prefer trees on low-lying sites in the Piedmont. Apparently, susceptibility of trees to SPB in the Piedmont is not associated with water relations as suggested for the Coastal Plains.

Table 4-1. – Summary of weather related stand disturbances in infested (attacked) and baseline (nonattacked) plots in the Gulf Coastal Plain.

Disturbances	Study Area
	Arkansas		Louisiana		Texas		Georgia
	Attacked	Non-Attacked	Attacked	Non-Attacked	Attacked	Non-Attacked	Attacked	Non-Attacked
	(%)	(%)	(%)	(%)	(%)	(%)	(%)	(%)
Severe ice or hail (>50% stems affected)	4.0	4.0	0.0	--	0.0	0.0	1.0	0.0
Light ice or hail (<50% stems affected)	20.0	20.0	0.0	--	0.2	0.7	5.0	4.0
Lightning	39.0	0.4	10.2	--	31.6	0.9	23.0	1.0
Wind	1.0	0.4	4.0	--	4.6	0.0	2.0	1.0

In the North Georgia Mountains, a totally different set of landform categories was used; therefore, these results are not directly comparable with those from other physiographic provinces. Infestations in the Mountains occurred more frequently on south-facing slopes, but this is the slope face where most of the pines grow in the southern Appalachians (Belanger et al. 1979b).

Water Regime

Previous discussions of the relationship between landform and occurrence of beetle infestations in the Coastal Plains imply that soil moisture may directly influence beetle susceptibility since infestations occurred more often than expected on low-lying areas and ridge tops. Both Lorio (1968) and Bennett (1968) pointed out that infestations in Texas and Louisiana were common on wet sites and that damage by root-rotting fungi may cause reduced vigor of trees on these sites and predispose them to beetle attack. Hick et al. (1978) noted differences in radial growth rates of trees in infested v. baseline plots on wet and waterlogged sites in Texas. On such sites, trees in beetle infestations were growing comparatively slower than those on other water regimes. But trees in baseline plots on wet and water-logged sites were growing faster than on other water regimes. The researchers also noted a greater-than-expected frequency of occurrence of beetle spots on wet and waterlogged sites and a lower-than-expected frequency of occurrence on dry sites in Texas. In the Coastal Plain of North Carolina (Belanger et al. 1977), all infestations visited were classified as being on either moist or wet sites. None were found on dry sites.

Although the evidence is circumstantial, it does appear that at least some of the association between landform and SPB activity in the Coastal Plain is due to differences in moisture regime, with infestations being more frequent on wetter sites. It should be noted that these data were collected during a period when rainfall was adequate. The results might be different during drier periods.

Soil Texture

Considerable regional variation was found in soil texture both within and between the two major physiographic provinces (Piedmont and Coastal Plain). Piedmont infestations uniformly occurred on fairly heavy clay soils; sandy or loamy soils characterized Coastal Plain infestations (tables 4-2 and 4-3).

In the Gulf Coastal Plain, infested and baseline plots were very similar in soil particle size distribution, although a considerable degree of variation existed across geographic areas within the province. For example, in east Texas surface soil sand content was about 70 percent; another project dealing principally with industrial landholdings in Louisiana, Mississippi, and Texas reported an average sand content of only about 50 percent (table 4-2). Clay contents were relatively similar across areas; the variation in sand was thus compensated for by variations in silt content (table 4-2). These variations reflect real differences in soil texture across the Gulf Coastal Plain, with a trend toward sandier soils from northeast to southwest. But they do not imply any selectivity on the part of the SPB based on soil texture.

Table 4-2. – Particle size distribution of soils from beetle-infested plots in the Gulf Coastal Plain.

Geographic Area	Surface Soil			Subsoil
	% sand	% silt	% clay	% sand	% silt	% clay
Arkansas	53.8	36.1	10.1	44.6	33.7	21.7
Louisiana	58.9	31.2	9.9	42.9	31.3	25.8
Louisiana, Mississippi, Texas	49.6	39.3	11.1	42.2	35.6	22.2
Texas	68.7	22.4	8.8	50.9	21.5	27.6

Table 4-3. – Particle size distribution of soils from beetle-infested plots in the Piedmont.

Geographic Area	Surface Soil			Subsoil
	% sand	% silt	% clay	% sand	% silt	% clay
Georgia	56.4	18.9	24.2	40.7	17.1	42.1
Virginia	37.8	36.5	25.3	28.5	34.8	36.2
North Carolina	54.5	32.2	14.3	35.5	30.6	33.9

In the Piedmont, quite a different situation exists regarding soil texture. First, the soils are much higher in clay content than those of the Coastal Plain. Here, however, a relationship of surface soil clay content with the presence of beetle infestations is apparent. The average surface soil clay content for all Piedmont infested plots is about 21 percent (table 4-3), while the baseline plots had only 18 percent. Conversely, the surface sand content was 55 and 59 percent, respectively, for infested and baseline plots. In the Georgia Piedmont, the situation was even more dramatic. Surface soil clay content for Georgia infested plots was 25 percent, as compared to only 19 percent for baseline plots. Belanger, et al. (1977) reported that both high clay content and abundance of shortleaf pine were associated with beetle infestations in their Georgia Piedmont study. Both these conditions also favor the development of littleleaf disease (Phytophthora cinnamoni Rand.), a particular problem on the eroded clay soil of the lower Piedmont. Belanger’s team believes that many infestations in the Piedmont may result from "locus" points of low-vigor trees associated with littleleaf sites. Such a hypothesis is consistent with the notion that beetles gain easier access to trees growing under stressful conditions, as proposed by Lorio and Hodges (1968b).

Soil Chemical Properties

Relationship of soil chemical properties to southern pine beetle occurrence has had little prior study. One Program-supported project at Stephen F. Austin State University undertook the quantitative measurement of mineral nutrients in soils of beetle-infested and noninfested plots in Texas. Researchers conducted laboratory analyses for organic matter; phosphorus, calcium, magnesium, potassium, zinc, sodium, and manganese; percent base saturation; and cation exchange capacity.

In the surface soil, several minerals differed significantly between infested and baseline plots: potassium, sodium, calcium, and zinc. Infested plots had lower levels of all these except for sodium, which was elevated in infested plots. In the subsoil, only zinc showed a significant difference, occurring at a lower level in infested plots. The hypothesis generated from these data is that nutrient deficiencies were responsible, in part, for infestations in east Texas. All the elements present at significantly lower amounts in infested plots contribute either directly or indirectly to soil fertility. In the case of sodium, higher amounts characteristic of infested plots could contribute to moisture stress and therefore to increased beetle susceptibility.

These data consist of averages from many plots, some of which were surely susceptible to southern pine beetle attack for reasons other than soil fertility. But even when plots found to have predisposing stand disturbances (e.g., lightning and recent logging damage) were eliminated from the analysis, all elements but potassium still differed significantly among infested and baseline plots. It seems logical that at least some stands were susceptible due to nutrient deficiencies and imbalances in the soil.

Attempts to evaluate the effect of fertilization on SPB resistance have, however, met with little success. Haines, Haines, and Liles (1976) recorded SPB occurrence in 176 Southwide fertilization plots and in adjacent unfertilized stands. Only 4 percent of the fertilized stands had SPB, and even fewer of the unfertilized stands were attacked. The Haines team felt that most of the infestations resulted from stand disturbances that happened during thinning some of the treated stands and had little to do with fertilizer treatment. Since the level of SPB activity was very low in both fertilized and unfertilized stands, we cannot draw broad conclusions from these observations. In another study, Moore and Layman (1978) applied 10-10-10 fertilizer to the fringe of active SPB spots. They reported that this practice did not significantly increase resistance of 9- to 11-year-old loblolly pines.

Site Index A weak but significant positive correlation existed between loblolly pine site index and SPB activity in the Gulf Coastal Plain (Rowell 1978 unpublished). Figure 4-1 shows that infestations developed more frequently in stands with higher site index classes. Some regional deviations were apparent from the different projects. For example, in the upper Coastal Plain of Arkansas, little difference in site index existed between baseline and infested plots. In the other projects located along the lower Gulf Coastal Plain, higher site index was associated with infested plots. Two other facts seem pertinent here. First, shortleaf pine — the most abundant species — was the preferred host species in the Arkansas study, whereas loblolly pine was the preferred host in the other West Gulf studies. Second, neither wet sites nor low-lying landforms were associated with SPB outbreaks in Arkansas as they were in Louisiana and Texas. It is a widely recognized fact that loblolly pine is more abundant than shortleaf pine on moister sites and usually occurs in stands having a higher site index than shortleaf pine (Zahner 1954). Apparently then, the regional differences seen within the West Gulf Coastal Plain relate to differences in species preferred by the beetle and a complex of contributory factors.

Figure 4-1 – Distribution of infested and baseline plots by site index class for the Gulf Coastal Plain (after Rowell 1978 unpublished).

Soil Depth, pH, Modifiers, and Accessory Characteristics

Shallow topsoil depth was associated with infestation in the Georgia Piedmont. As previously discussed, this trait is often characteristic of sites prone to littleleaf disease of shortleaf pine, further substantiating the role of this disease in SPB susceptibility in this region. Soil depth was not associated with SPB occurrence elsewhere in the Gulf Coastal Plain (Rowell 1978 unpublished).

Neither soil pH, modifiers, nor accessory characteristics as defined in the Soil Resources Guide: Southern Region (U.S. Department of Agriculture Forest Service 1972a) were found to be associated with SPB activity.

Stand Factors Stand Density High stand density (fig. 4-2) has long been implicated as a causal factor of southern pine beetle infestations (Bennett 1968, Leuschner et al. 1976, Lorio and Bennett 1974). The rationale is that overstocking causes reduced vigor of trees and therefore predisposes them to SPB attack (Bennett 1971). This hypothesis was borne out, particularly on wet sites in east Texas, where Hicks et al. (1978) noted that low tree vigor was indeed correlated with stand density and that low-vigor stands were markedly more susceptible to beetle attack.

Figure 4-2 – A high-BA stand of pines that have been attacked by the beetle.

Program-supported studies used two measures of stand density — trees per acre and basal area (BA) per acre. The former was not very useful relative to SPB incidence, since trees per acre are dramatically affected by tree diameter. Also, with the exception of Arkansas, where infested plots had significantly more trees per acre, this variable was not found to be associated with SPB attack. However, spatial arrangement of trees, reflected by trees per acre, may be associated with the rate of growth of active SPB spots. Basal area, on the other hand, was found to be quite highly associated with SPB occurrence. BA was tallied for pines and hardwoods separately; in most cases, it was the pine component that showed the highest degree of association with SPB.

In the Gulf Coastal Plain studies, beetle occurrence was strongly associated with high stand BA. Rowell (1978 unpublished) reported that the average pine BA for 2,021 infested plots in the Gulf Coastal Plain was 114.4 ft²/acre, as compared with 72.5 ft²/acre for 1,396 uninfested plots from the same region. Hardwood BA showed a similar trend, with infested plots having 36.7 ft²/acre, compared to 26.4 for uninfested plots. Hardwood BA was not as consistent as pine BA across projects; i.e., Arkansas data showed little difference for this variable between infested and baseline plots.

Averages for total BA (combining pine and hardwood) were 140.7 and 109.2 ft2/acre for infested and uninfested plots, respectively, in the Gulf Coastal Plain. Figure 4-3 illustrates the distributions of total BA for infested and uninfested plots. These data leave no doubt that stand density is indeed important in determining susceptibility of stands to SPB attack in the Gulf Coastal Plain. In Arkansas, where site factors assumed less importance than in other parts of the Gulf Coastal Plain, high pine BA was associated with infestation. Thus, it appears that for the Gulf Coastal Plain this is a variable consistently associated with SPB activity and is potentially useful in identifying stands that are more likely to be attacked. The correlation between BA and southern pine beetle occurrence seems to relate to competition and its effect on tree vigor. This hypothesis is further supported by the fact that higher hardwood BA is associated with beetle attack. Since hardwoods are nonhosts, their effect must be related to their competition with the host pines and resulting loss of vigor.

Figure 4-3 – Distribution of baseline and infested plots by basal area classes for the Gulf Coastal Plain (after Rowell 1978 unpublished).

Studies in the Piedmont regions of Georgia and North Carolina did not find similar relationships between BA and beetle infestation. In Georgia, as previously discussed, site factors, littleleaf disease, and host species (shortleaf pine) were related to SPB occurrence. In the North Carolina study, data were collected primarily on trees within the fluctuating water line of Kerr Reservoir. Here, flooding and related root damage were associated with the infestations (Maki 1978 unpublished).

In the North Georgia Mountains, as in the Coastal Plain, higher pine and hardwood BA were associated with SPB attack (Belanger et al. 1979b).

Radial Growth

Bennett (1968, 1971) suggested that poor tree vigor, expressed as reduced radial growth, was consistently associated with southern pine beetle infestations. Data of Coulson et al. (1974) further substantiated this observation in east Texas.

Data collected on more than 3,000 infested and noninfested plots in the Gulf Coastal Plain during 1975-1979 have verified and quantified the relationship of reduced growth rate with SPB attack. Investigators collected data from three to five dominant or codominant trees within each plot and measured the width of the most recent five annual rings and the preceding five rings. Current 5-year increment for infested plots in the Gulf Coastal Plain was 15.03 mm, compared to 18.02 mm for uninfested plots (Rowell 1978 unpublished). Respective data for the preceding 5-year period were 16.23 mm and 19.51 mm. These results were consistent for all projects in the Gulf Coastal Plain region, although data for current 5-year radial growth in east Texas showed the most marked difference between infested and baseline plots.

In the Piedmont studies a similar association of reduced radial growth with beetle attack was found, but the degree of difference was less than for the Coastal Plain. Table 4-4 shows the results for the Coastal Plain and the Piedmont studies. Although infested plots were consistently slower growing than their noninfested complements, considerable variation existed between regions and among projects within regions. Generally, growth was slower for Piedmont plots compared to the Gulf Coastal Plain. Within the Coastal Plain, Texas and Louisiana trees grew more slowly than those in Arkansas. The Arkansas plots were generally in younger, smaller-diameter stands, which would be expected to grow more rapidly than older, larger trees.

Table 4-4. – Radial growth for infested and baseline plots in the Gulf Coastal Plain and Piedmont

	Gulf Coastal Plain						Piedmont
	AR		LA		LA, MS, TX		TX		GA		NC
	Inf.	Base.	Inf.	Base.	Inf.	Base.	Inf.	Base.	Inf.	Base.	Inf.	Base.
Recent: 5 yrs. (mm)	17.9	19.2	14.6	--	16.5	19.3	13.9	17.6	11.5	14.1	8.6	9.1
Previous: 5 yrs. (mm)	19.2	21.0	15.3	--	18.0	20.6	15.9	18.7	15.1	18.0	10.2	10.4

Presumably, radial growth is a reflection of many factors affecting tree vigor. Perhaps this is why radial growth is consistently associated with SPB attack throughout the South. Within the Coastal Plain, it appears that overstocking and wet site conditions — either alone or in combination — contribute to reduced vigor, hence reduced radial growth. In the Georgia Piedmont, the cause of reduced radial growth is more commonly associated with site factors that predispose shortleaf pines to littleleaf disease. The growth reduction reflects the poor vigor of littleleaf trees. In the North Carolina study, periodic flooding along Kerr Reservoir seems to bring about the same effect — reduced vigor of trees. Therefore, radial growth may indeed be a consistent variable that can be used as a Southwide index to SPB susceptibility. To apply such an index, however, researchers need good baseline data for each region or subregion. Some means for adjusting radial growth data for tree or stand age will most likely be needed in order to make comparisons across different age classes (Hicks et al. 1978).

Workers in the site-stand group of the ESPBRAP measured bark thickness at the fissures and ridges of three to five trees per plot using a standard bark gage. Results were confusing: Southwide, investigators found no distinct differences between baseline and infested plots. This discovery may reflect two facts: data were combined for several pine species, and different species were preferred hosts in the various study areas. Certainly, it is well known that shortleaf has thinner bark than loblolly pine.

In the Coastal Plain, trees in infested plots had significantly thicker bark than their uninfested counterparts. Fissure values were 0.28 inches v. 0.27 inches and ridge values of 0.92 inches v. 0.83 inches for infested and baseline plots, respectively (Rowell 1978 unpublished).

Certain projects found very strong association of bark thickness with infestation. For example, in east Texas, fissure bark thickness for infested trees averaged 0.23 inches in comparison to 0.15 for baseline trees. Researchers in this study found that fissure bark thickness was the best single variable to distinguish infested from baseline plots. Another Coastal Plain study dealing with industry-owned lands in Louisiana, Texas, and Mississippi also noted that thicker bark in fissures was associated with SPB attack but had much higher values (0.44 inches v. 0.39 inches). These project-related differences could be due to variations in measurement methods.

In Arkansas, researchers found almost no difference between infested and baseline trees relative to bark thickness. In this study, it should be emphasized that shortleaf pine was the preferred host while loblolly pine was the preferred host for other Coastal Plain studies. In the Georgia Piedmont, where again shortleaf pine was the preferred host, no apparent relationship between attack and bark thickness was evident.

Apparently, bark thickness is related to beetle attack in loblolly pine but perhaps not in shortleaf pine, or at least for the latter species the data are obscured by other factors. The exact cause-and-effect relationship between bark thickness and SPB infestation, if one exists, is not immediately obvious. Perhaps beetles are capable of producing more and healthier brood to reinfest adjacent trees when feeding on trees with thicker bark. Such is the case with mountain pine beetle in lodgepole pine (Amman and Pace 1976). In any event, bark thickness is somewhat difficult to measure, owing to within- and between-tree variability and the limitations of currently available measurement devices.

Species Composition

Species of southern pines differ in their susceptibility to the beetle. Longleaf and slash pines are fairly resistant, a phenomenon that has been attributed to their ability to "pitch out" attacking beetles in resinous exudations. However, Program-sponsored studies have found that, even among so-called susceptible host species such as loblolly and shortleaf pines, preferential attack can occur and preferences vary by regions. For example, in east Texas, loblolly pine constituted approximately 30 percent of the species mix, but about 50 percent of the infested trees were loblollies. This species was also found to be the most frequently attacked in other lower Coastal Plain studies. But in the upper Coastal Plain of Arkansas and the Piedmont of Georgia, shortleaf pine was more frequently attacked. These differences probably relate to the unique factors within the various provinces that predispose trees to beetle attack. In the lower Coastal Plain, many infestations occur in overstocked stands on wet sites. Loblolly pine — a more hydric species — is more frequent than shortleaf pine on such sites and thus becomes the preferred host under these conditions.

In the upper Coastal Plain of Arkansas, wet site conditions do not seem to predispose trees to southern pine beetle attack. Except for the extreme southern counties of Arkansas, along the Red River, shortleaf pine is the most abundant species. The greater abundance of shortleaf pine therefore accounts for its higher frequency of attack.

In the Georgia Piedmont the previously noted soil conditions that influence the susceptibility of shortleaf pine to littleleaf disease also seem to be the major predisposing factors to SPB attack. Loblolly pine is less frequently infested with littleleaf and is not the preferred host species in the Piedmont.

In the North Georgia Mountains, shortleaf, pitch, and loblolly pines were found to be more susceptible to SPB than Virginia pine or eastern white pine (Belanger et al. 1979b). These authors recommended species selection as a means to reducing losses from SPB.

Average Stand Age, Height, and Diameter

These highly correlated characteristics were infrequently associated with SPB infestation across the South. Rowell (1978 unpublished) reported that average stand age of infested plots in the Gulf Coastal Plain is about 40 years for both infested and uninfested plots. Considerable regional variation in average stand age, height, and diameter was reported. At the project level, some investigators found them to be associated with SPB occurrence.

Part of the variation of the projects is due to differences in ownership and management objectives. For example, a Louisiana study dealing specifically with the Kisatchie National Forest reported the largest and oldest trees of any Gulf Coastal Plain study. Trees in infested stands averaged 44 years of age, 12.2 inches in d.b.h., and 67 ft in height. Trees in National Forests are generally larger than trees on adjoining ownerships. But even so, beetles apparently selected the older and larger trees within the National Forest for attack (fig. 4-4). Examining only infestations on the Davy Crockett National Forest in Texas, Leuschner et al. (1976) also found that infestations occurred more often in larger-diameter trees. A study dealing with industrial landholdings in Louisiana, Texas, and Mississippi reported the second-oldest and –largest trees in the Coastal Plain, with an average age of 41 years and d.b.h. of 11.2 inches. No differences in age and size between infested and uninfested stands were detected. Projects in Arkansas and east Texas reported the smallest trees in the Gulf Coastal Plain. In both areas, plots were selected in an unbiased manner and without regard for landowner. Tree diameters in these studies averaged about 24.5 cm and average heights varied between 17.1 and 19.2 m. In Arkansas younger, smaller-diameter trees were infested more frequently than older, larger ones. This relationship was amplified when plots with stand disturbances were excluded from the analysis (Ku, Sweeney, and Shelburne 1976 and 1977). In east Texas, infestation incidence was not related to age of d.b.h., but infested trees tended to be somewhat taller (fig. 4-5). This observation could relate to the fact that this lower Coastal Plain study found a high proportion of infestations on wet sites, which have a higher site index for loblolly pine, the preferred host species.<

Figure 4-4 – Percentages of SPB infestations and general forest area by stand age classes in the Kisatchie National Forest, Louisiana (from Lorio, in Coster and Searcy 1979).

Figure 4-5 – Distribution of baseline and infested plots by total tree height classes for east Texas.

In the Georgia Piedmont, infested pines were much smaller than in the Gulf Coastal Plain (8.5 inches d.b.h., 52 ft tall), but their age (39 years) was about the same (Belanger et al. 1977). Reduced size for age may, in part, be related to the fact that shortleaf pine was the preferred host and many infestations occurred on sites conducive to littleleaf disease. On such sites, stunted growth is common for shortleaf pine.

In the North Georgia Mountains, infested pines were considerably older than elsewhere (Belanger et al. 1979b), but trees were only slightly larger (11.6 inches d.b.h., 67 ft tall). The inherently slower growth of the species attacked and the harsher growing conditions encountered probably account for the reduced size of these older trees.

Diseases and Other Insects

Several other diseases and insects are known to reduce vigor of southern pine and therefore they may predispose trees to attack by the beetle. One such disease is annosus root rot (Heterobasidion annosum). This disease is a particular problem in thinned plantations and especially on sandy soils. Under such conditions the disease causes outright death of trees, often spreading in a radiating fashion from an infection point, such as a cut stump. A more subtle condition is the endemic infection of annosus root rot that affects varying proportions of otherwise living pine roots and causes reduced tree growth (Bradford and Skelly 1976). Further, Skelly (1976) noted that approximately 30 percent of the roots of SPB-attacked trees were infected with annosus root rot, compared with 20 percent for unattacked trees on the same site.

Virginia investigators established a series of plots in Virginia, Georgia, and Texas on high-hazard annosus sites (> 70 percent sand in topsoil). They excavated tree root systems in beetle infestations and adjacent noninfested stands. All trees showed some sign of annosus root rot, but infection was more severe in plantations than in natural stands and on SPB-infested than noninfested plots. These results suggest that under certain conditions annosus root rot is an important predisposing agent for SPB attack.

In another study in the Georgia Piedmont, littleleaf disease of shortleaf pine has also been implicated as a predisposing agent of beetle attack (Belanger et al. 1977). Several facts support this conclusion. First, shortleaf pine, the host of littleleaf disease, is also the preferred SPB host species in the Georgia Piedmont. It constitutes roughtly 45 percent of the pine component in the region, and 69 percent of the beetle infestations occur in shortleaf pine. And conditions favoring the development of littleleaf disease (eroded heavy clay soils) are also associated with SPB infestation. It appears that in the Piedmont, littleleaf disease causes reduced growth and vigor of shortleaf pine and this, in turn, predisposes trees to SPB attack. Other bark beetles, such as black turpentine beetle and Ips, are known to weaken trees and lead to infestations of SPB. Preliminary attacks by other bark beetles are often associated with stand disturbances such as logging, lightning, or wind damage (fig.4-6). Their effect is similar to that of other causal agents in reducing the vigor of the host tree and hence its ability to resist SPB attack. Summary and Conclusions Coastal Plain

Figure 4-6 – Pitch tubes signify the attack of black turpentine beetle after this tree was damaged by logging equipment.

Infestations studied in all Coastal Plain projects typically occurred in overstocked stands of slower-than-normal growth. In the lower Coastal Plain of Mississippi, Louisiana, and Texas, a higher-than-expected frequency of infestations occurred on wet or low-lying sites of higher site index. Loblolly pine was the preferred host species. In Coastal Plain stands, increased beetle activity on wet sites often follows periods of abnormally high rainfall. Pines in beetle-infested stands in this region have thicker bark than those in uninfested plots. Infestations in the Kisatchie National Forest of Louisiana occur in stands of older and larger pines. But the age and size of infested pines varies on other ownerships in the lower Coastal Plain, where timber harvesting has eliminated many of the mature and overmature stands.

In the upper Coastal Plain of Arkansas, shortleaf pine was the preferred host species. Overstocking and reduced radial growth were associated with SPB attack, as in the lower Coastal Plain. But trees on low-lying landforms were not preferentially attacked (shortleaf pine is seldom found on these sites). Beetles tended to prefer younger, smaller trees in Arkansas — a reversal of the trend in the Kisatchie National Forest of Louisiana.

A few plots were established in the Atlantic Coastal Plain from Virginia to Georgia. Although inconclusive, the results are generally similar to those of the Gulf Coastal Plain in that overstocked stands of loblolly pine with reduced radial growth growing on wet sites are most frequently attacked by SPB.

Piedmont

The only factor similar for Piedmont and Coastal Plain infestations was reduced radial growth of infested trees. Otherwise, infestations typically occurred in shortleaf pine growing on greatly eroded, heavy clay soils. Also, littleleaf disease is more prevalent in the Piedmont.

Mountains

Relatively few plots were established in the North Georgia Mountains. Again, overstocking and reduced radial growth were associated with SPB infestation. Also, shortleaf and pitch pines were preferred host species as evidenced by the proportion of beetle attacks in stands of those species relative to their presence in the forest. Virginia pine, the most abundant species in the North Georgia Mountains, was attacked less frequently than expected.

Southwide Conclusions

The underlying factors contributing most to host susceptibility to SPB throughout the South are low tree vigor and/or stress. This is evidenced by the consistent Southwide association of SPB with reduced radial growth and certain stand disturbances. Reduced radial growth (low vigor) occurs in overstocked stands in the Coastal Plain and Mountains and in the presence of certain soils conditions in the Piedmont. Lightning is associated with initial SPB attack throughout the South. Locally, many other factors may work independently or in combination with those mentioned above to predispose trees to beetle attack. For example, plantations on sandy sites are vulnerable to annosus root rot, which in turn reduces vigor and SPB resistance.

Very few "new" findings were elucidated by the Program. Researchers had previously identified the aforementioned factors in numerous independent studies. What the ESPBRAP did do was to quantify the effects of these variables, and through a coordinated effort provide geographic replication to identify regional similarities and differences. Our results, unlike those of preceding studies, are useful for developing stand hazard rating models and management recommendations. In fact, several such models are either complete or in preparation (see Chapter 8). These models will permit land managers to identify those stands that warrant special attention and remedial action to minimize SPB losses.

Another use of these data will be in developing recommendations for preventing or reducing further SPB damage. For example, problem sites can either be avoided or amended, overstocked stands can be thinned or harvested, and highly susceptible species can be phased out in high-hazard areas (Hicks, Coster, and Watterston 1979).

The work done to date is only the beginning, but it provides a solid foundation on which to build future research.

[  Contents  ]   [  Previous  ]   [  Next  ]   [  Home  ]