Survival of Slash Pine Having Fusiform Rust Disease Varies with Year of First Stem Infection and Severity

Probabilities of death of young slashpine infected byfusiform rustpathogen varied with timing and severity of infection. Trees in nine slashpineplantations varying widely in site quality and initial number oftreesperacre hadsimilarprobabilities ofdeathfrom rust. About 90% oftrees with stem infections in thejirst three growing seasons died by age 15 if the gall spanned more than 50% of the circumference of the stem by age 5. If 50% gall encirclement occurred after age 5, mortality rates dropped to about 30% at age 15. Where first stem infection occurred after the fifth year, probability of death was essentially the same as for rust-free trees. Methods are given for using timing-severity data to estimate future stocking. South. J. Appl. For. 22(2):96-l 00. Efflclent management of slash pine (Pinus elliottii Engelm. var elliottii) stands infected by fusiform rust pathogen (Cronartium quercuum [Berk.] Miyabe ex Shirai f. sp. fusiforme) requires accurate predictions of preharvest rustassociated mortality. Death of trees with early stem infections of fusiform rust is common (Sluder 1977, Wells and Dinus 1978, Nance et al. 1981, Lloyd 1982, Shoulders and Nance 1987). Survival models (Nanceet al. 1983, Devine and Clutter 1985) or mortality simulation models (Geron and Hafley 1988) are available for estimating future stocking of fusiformrust-infected stands. Shoulders et al. (1991), showed however, that trees infected in the third growing season have a lower probability of fusiform rust disease-related mortality than trees infected in the second year. Percent of stem encircled by the gall, an indication of the disease development rate, also affects the survival of pine (Belanger and Zarnoch 1991). Other possible mortality associated factors include site quality, stand density, and host genotype. Applicability of survival prediction models to individual forests depends on the importance of the above factors. NOE: Manuscript received July 3, 1991, accepted May 26, 1997. R.C. Schmidtling can be reached at (601) 832-2747; Fax: (601) 832-0130; E-mail: [email protected]. Thanks to Georgia Pacific Corporation, Intemational Paper Company, and Interpine Lumber Company for providing land for experimental plantations. Technical assistance of L.M. Lott is gratefully acknowledged. The objectives of this study were: (1) to examine how timing and severity of fusiformrust pathogen infection affect mortality of planted slash pine, (2) to investigate the consistency of rust associated mortality predictions among locations with different stand densities and site qualities, (3) to determine whether probabilities of death vary among seed sources, and (4) to provide guidelines for using timingseverity data to estimate future stocking of slash pine. Materials and Methods Probabilities of death of planted slash pines having fusiform rust through age 16 are based on data from a study installed in 1974 at nine locations in coastal counties of Mississippi (Froelich and Snow 1986). Each plantation consisted of 25 rows of trees, with 30 trees per row. Spacing was 6 ft within rows and 10 ft between rows. Trees of one of three seed sources were planted in each row. Two sources originated from different seed production areas. The third source was from open-pollinated seed collected from seed orchard clones whose progeny had shown resistance to fusiform rust in previous tests. First-year survival of planted seedlings ranged from 236 to 583 trees/at (33 to 91%, Table 1). Mortality in the first year was due to excessive or inadequate soil moisture after planting. Fusiform rust disease infection, site index, and soil drainage varied substantially among locations (Table 1). 96 Reprinted from the Southern journal ofApplied Fores@, Vol. 22, No. 2, May 1998. Not for further reproduction. Table 1. Characteristics of nine plantations of slash pine used to study effects of fusiform rust on tree survival. Plantation no. Characteristic 1 2 3 4 5 6 7 8 9 No. live trees/at ’ 421 236 388 320 393 513 583 399 550 Site index * 59 67 59 69 67 Drainage’ M W-E M M-W M Percent rust’ 47 71 30 29 31 ’ Agel. 2 Baseage25yr. 3 Drainage: P = poor; M = moderate; W = well; E = excessive. 4 Drainage and site quality improved by bedding in Plantation 9. 5 (Cumulative stem infection at age 16)/(number live trees at age 1). 75 67 67 75 M-W M-W W p’ 12 58 68 41 Detailed observations on fusiform rust disease development rate were made each year for 16 yr. Data included: (1) location of each gall (distance above ground and distance to branch galls from the main stem); (2) year (since planted) when each gall developed; (3) year when branch galls spread into main stem; (4) year when more than 50% of the stem was encircled by one or more galls; (5) year of witches’ brooming (multiple stems and no primary stem) due to main stem mortality associated with fusifotm rust disease; (6) year of partial crown death at a stem gall; and (7) year of tree death identified as not rust related, broken at rust gall, or dead but standing with a stem gall or canker. stroyed Plantation 6. Plantation 7 was destroyed by wildfire after season 13 measurements. Moderate to heavy crown scorch from a wildfire probably hastened death of some infected trees in Plantation 1. Each tree was classified using a numeric/letter code that indicated the timing and severity of the infection, to aid presentation and analysis. The numeric prefix is the age when first stem infection occurred, or when a branch gall spread into the stem. The latter was defined as stem swelling due to assumed presence of the fungus. When stem swelling occurred after a nearby galled branch died, the year of stem infection was considered year of branch death. There were six age classes; 1, 2, 3,4,5, and 6+. Trees having stem infections occurring after age 5 were labeled 6+. Annual height measurements were used to reconstruct year of fust stem infection when direct stem infection was not detected for one or more years after it occurred. Probabilities of death were computed two ways; By actual year of infection, which includes infections discovered later and assigned to that particular year, and by infection that was actually visible at a given age. The former gives a better perspective of the total effects of rust development on stocking over time, the latter is more useful in using field observations made at a specified early stand age to predict future stocking. Because of the nature of the data, an overall factorial analysis was not feasible. The analysis was broken down into three levels. Mortality in the 18 infection classes (including those free of stem infection-rust-free) was examined for all plantations combined, for plantation differences, and for seed source differences. Class 6+A does not exist, because infection occurred after age 5. Classes lC, 2C, and 3C (stem infection in growing season 1, 2, or 3 and not 50% stem encircled at age 16) were dropped from the analysis because these classes had too few trees. The early infections nearly always became stem encircling. Probabilities of Death From Rust-All Plantings Combined Differences in mortality among infection classes for all plantings combined were large (Figure 1) and statistically significant at age 16 (P < 0.00 1). Mortality in classes 1 A, 2A, and 3A ranged from 90% to 95% at age 16. The differences were not statistically significant (P > 0.6). There were important differences, however, in the early time trends of the three. At age 5, 40% of the trees infected the first year (IA) were dead versus 15% for 2A and only 5% for 3A. The differences among these classes at age 5 were significant statistically (P <O.OOl). These differences show that rust infection has avery serious impact on early stand density when it occurs in the first or second growing season. This early impact has not been considered in growth and yield models that base projections on trees living at ages 3 or 5 or later. Severity of infection was defined as tree age when at least 50% of the circumference of the main stem of the tree was encircled by a stem gall. “A” is equivalent to 50% encirclement by age 5, “B” is equivalent to 50% encirclement after age 5, “c” is equivalent to not 50% encircled by age 16. Age 5 and 50% are not arbitrary numbers. Experience has shown that 50% encirclement by age 5 results in reduced growth and early tree mortality in plantations (Froelich et al. 1983). 100