COMPARISONS OF TREE HEIGHT GROWTH ON BROADCAST-BURNED, BULLDOZER-PILED, AND NONPREPARED SITES 15 TO 25 YEARS AFTER CLEARCUT LOGGING

Bob Bosworth
Dan Studer

ABSTRACT

Tree height growth was compared on clearcuts that were broadcast burned, bulldozer piled, or had no preparation for regeneration in the Bonners Ferry Ranger District, northern Idaho. Differences in heights and growth are related to treatment. Height growth on burned sites exceeded that on piled sites in 8 of 9 years. Questions arise as to effects on site productivity of piling. Reasons for differences include nutrients available to regeneration related to site preparation method, and compaction effects from piling.

INTRODUCTION

The Bonners Ferry Ranger District of the Idaho Panhandle National Forests lies in northern Idaho. Generally, it is bounded by Canada to the north, Montana to the east, the crest of the Selkirk Mountains to the west, and extends south to the Boundary/Bonner County line. Elevations range from 2,500 to over 6,000 feet. Habitat types range from those of the relatively warm, dry Douglas-fir (Pseudotsuga menziesii) series to those of the cool, moist subalpine fir (Abies lasiocarpa) series (Daubenmire and Daubenmire 1968).

Subsoils in the district are dominated by glacial tills from both continental and alpine glacial action. Some are subsoils formed from glacial-fluvial outwashes. Most of the surface soils are volcanic ash in origin. This ash cap is a very good medium for growing trees (Ford 1985).

Clearcutting as a regeneration system has been practiced since as early as 1916 (Roe and DeJarnette 1965) in the district. However, use of the clearcut system became increasingly routine from the late 1950's to the mid 1970's. Most of this cutting has been in overmature to decadent timber stands where large quantities of slash and debris dictate some sort of site preparation to achieve adequate regeneration and to meet fire hazard reduction objectives. Broadcast burning and bulldozer piling have been the most frequent treatments. Some stands have had no site preparation treatments.

The district is continually juggling its program, considering the limited times suitable for broadcast burning and the total acres to be treated for planting each year. Typically, debris on slopes less than 35 percent is piled, while steeper slopes are burned. In years of larger programs, bulldozers are moved onto even steeper slopes to balance the program and get all sites prepared in a timely manner. With increasing concerns over air quality, there is more pressure to move away from broadcast burning and to piling and burning to reduce impacts on airsheds.

As our knowledge base on the complex interactions of our forest ecosystems expands, the questions of short- and long-term impacts on growth related to site treatments are repeatedly asked. Bonners Ferry District personnel have been intensively monitoring progress of regenerating stands for many years. We wondered if the data from monitoring might be able to shed some light on the effect of treatment on regeneration performance. We were particularly interested in looking at performance of stands 15-25 years after harvest.

This paper reports the extension of a study that was presented in the Prescribed Fire in the Intermountain Region Symposium Proceedings (Bosworth 1989). The 1989 paper discussed differences in height growth found on clearcuts that were logged in the years 1960 to 1965. Since then the study has been extended to include clear- cuts that were logged in 1966, 1967, and 1968.

METHODS

District personnel monitor the progress of stands after harvest with a regeneration status survey. On a plot, regeneration is tallied by species, height to the nearest foot, and last year's leader growth to the nearest inch, and trees are classified "crop" or "excess." Relatively high numbers of plots are surveyed, since one of our main concerns early in the rotation is poor distribution of regeneration. The high number of plots monitored was important to this study, as this gave us many trees to assess. Plots are systematically distributed throughout the clearcuts on a grid basis.

The stand database was queried for all clearcuts harvested during the period of interest. Stands were selected for analysis if both tree height and leader growth estimates were recorded for the plots. A total of 104 stands had adequate data from 2,752 plots.

Stands were stratified by site preparation method; 58 were treated by broadcast burning, 38 by bulldozer piling, and eight had no site preparation.

Each regeneration status plot in a selected stand was analyzed. The dominant tree was selected, and species, total height, and last year's leader growth of that tree were recorded. Since the stands were on differing inventory schedules, and year of last examination varied, it was necessary to normalize the data for analysis. For phase 1, 1960-65, of the study trees on plots examined in years prior to 1982 were "grown" to 1982 by adding appropriate numbers of annual leader growths to the total height estimate at time of exam. For phase 2, 1966-68, a similar process was applied for trees on plots examined prior to 1986. For comparison between the two phases of this study, years 1963/1966; 1964/1967; and 1965/1968 had been clearcut for the same number of years at time of analysis.

A mean total height (1982 basis, phase 1 and 1986 basis, phase 2) and mean leader growth were computed for each site preparation treatment for each year of clearcut origin. Analysis of variance was then used to compare treatments to see if differences in the means were related to treatment (Freese 1967).

RESULTS AND DISCUSSION

In phase 1, we determined that, in general, bulldozer piling for site preparation had a negative impact on both height growth and total heights of trees, when compared with broadcast burning and no site preparation. From the analysis of phase 2 data, it is apparent that these same trends are still evident, although there is some evidence that the differences between treatments are not as great in the newer stands.

Figure 1 shows average leader growth by year of harvest. In 8 years out of 9, growth on burned sites exceeded growth on bulldozer-piled sites. Growth on burned sites also exceeded that on nonprepared sites in 7 years out of 9. These differences are related to treatment and are significant at the 0.01 level in all years except 1962. In 1962 there is no significant difference in growth between burned and piled sites, but the differences between growth on nonprepared sites are significant when compared with the other treatments.

Figure 1—Mean growth of crop trees; comparison of site preparation methods. [view larger image - 68K] [Text description of this figure]

Where leader growth differences were significant, growth on burned sites was from 1.6 to 5.2 inches per tree better than on piled sites, and from 1.7 to 2.0 inches better than on nonprepared sites in years where burned areas had the best growth. When growths were best on nonprepared sites (1964, 1967) they exceeded growths on burned sites by 0.6 and 1.4 inches per tree.

When average total heights are compared, in 7 of 9 years, trees on burned sites are taller than trees on bulldozer-piled sites (fig. 2). The exceptions are years 1962 and 1966. In the 5 years when all three treatments occurred, burned sites had the tallest trees twice, nonprepared sites had the tallest trees twice, and bulldozer-piled sites had the tallest trees once. Analysis of variance shows that the differences between treatments for all 9 years is significant at the 0.01 level.

Figure 2—Mean height of crop trees; comparison of site preparation methods. [view larger image - 56K] [Text description of this figure]

In the 7 years where average heights on burned sites exceeded the heights on piled sites, the differences in means ranged from 0.9 foot in 1968 to 5.7 feet in 1960, with an average 2.9-foot height advantage on burned sites. In the 2 years when piled sites had the taller trees, the differences were 5.3 and 1.3 feet.

Caution is advised in drawing conclusions about the total height differences expressed by the nonprepared site data. Examination of the trees included in these plots showed that they include some very old (70-year) advanced regeneration, and this factor is accentuated by the fact that in years where the differences are greatest, there is a very small sample. This advanced regeneration was not removed from the study because it is possible that some old advanced regeneration may also have survived piling or broadcast burning and may be in those plots also. Their effect is less in the pile or burn plots because there are generally many more trees in this part of the analysis, and a few trees do not affect the average as much.

The implications of the differences in height growth displayed in figure 1 in favor of burned sites over bulldozer piled sites, combined with the differences in total height displayed in figure 2, generally in favor of burned sites, cannot be ignored. A total of 94 stands comparing 2,496 trees are represented. Smith (1962) stated, "Within a stand, few trees ever recover a dominant position after they have fallen behind in the race for the sky." The trees in this study on the bulldozer-piled sites have not lost their dominant position in the stand, but even so, on the average, they are falling behind in height growth when compared with trees on burned sites. Based just on differences in total height, with no allowance for differences in diameter growth, if this trend continues for the remaining four-fifths of the rotation, the trees on the burned sites will average 10 to 22 percent greater sawlog volume than those on piled sites. It is reasonable to expect that if factors are reducing height growth of trees, they would also be adversely affecting diameter growth, thus producing even greater volume differences between treatments.

There are several reasons that these differences in height growth are being observed. One possibility is the flush of nutrients available for plant growth following broadcast burning (Holdorf 1982). There are also at least two possibilities for the reduced growth on piled sites. Compaction of soils has been shown to reduce height growth (Froehlich 1978). Even with one or two passes of a tractor over the soil while piling, some increase in soil bulk density occurs, particularly if the soils are wet. Another, and perhaps more important, reason may be related to nitrogen fixation. Nitrogen-fixing activity is normally between 5 and 10 times greater in organic versus soil horizons (Jurgensen and others 1980). It is not uncommon for much, if not all, of the litter and humus layers to be shoved into the piles during bulldozer site preparation. This is especially true when high volumes of logging debris exist. In that case, it is very difficult to keep from scraping all the organic matter off the site. This would greatly reduce the soil nitrogen-fixing process over much of the area piled.

It is likely that the differences in height growth demonstrated in this study are the result of a combination of these factors. The design of this study was not sufficient to ferret out precisely what the causes of the differences were.

Results of the phase 2 analysis are encouraging from several aspects. First, differences in height growth on burned and piled sites are not as great as they were in phase 1. This indicates that management decisions to use brush blades versus straight bulldozer blades, and to not pile slash as thoroughly, were in fact good decisions and are resulting in less impacts to the site.

The second encouragement from phase 2 is the indication that newer work is producing better trees than older work. Phase 1 trees-year 1963 are the same age as phase 2-year 1966; year 1964 the same age as 1967; and year 1965 the same age as year 1968. In figure 2, it is apparent that newer work is producing taller trees at younger ages in both the burning and the piling treatments. Some factors that may be causing these improved tree heights may be: (1) improving coordination to get prompt planting accomplished following site preparation; (2) improving planting stock handling and storage prior to planting; (3) improving attention to detail of properly planting each tree; (4) decreasing site preparation effects (less intense broadcast burns and less site disturbance during piling operations); and (5) earlier reaction to scattered mortality in plantations by interplanting.

CONCLUSIONS

These data indicate that trees on broadcast-burned sites will have a height growth advantage over trees growing on bulldozer-piled sites. This advantage will be evident for at least 20 years. If the manager has a choice of methods, burning should be chosen over piling.

However, in situations where there is a big program, the manager needs the flexibility to use both methods in order to accomplish timely reforestation. Consequently, when piling must be used, it is important that everyone involved in the process be aware of the potential for long- term growth losses on those sites. The logger, the contract administrator, and the bulldozer operator need to know the effects of excessive scraping of the organic layers of the soil and also the potential for compaction of the site from "overworking" the area with a tractor.

In addition, silviculturists and fuel managers must take a critical look at some stands to determine if site preparation is in fact needed. Unnecessary working of a site by heavy equipment may cause hidden long-term damage.

REFERENCES

Bosworth, Bob. 1989. Height growth on burned, piled and nonprepped clearcuts, 17 to 22 years after harvest. In: Baumgartner, David M.; Brewer, David W.; Zamora, Benjamin A.; [and others], eds. Prescribed fire in the intermountain region: forest site preparation and range improvement; 1986 March 3-5; Spokane, WA. Pullman, WA: Washington State University, Conferences and Institutes: 65-67.

Daubenmire, R.; Daubenmire, Jean B. 1968. Forest vegetation of eastern Washington and northern Idaho. Tech. Bull. 60. Pullman, WA: Washington Agricultural Experiment Station, College of Agriculture, Washington State University. 104 p.

Ford, Gary L. 1985. Preliminary report, soil survey of the Bonners Ferry Ranger District. Coeur d'Alene, ID: U.S. Department of Agriculture, Forest Service, Idaho Panhandle National Forests. 139 p.

Freese, Frank. 1967. Elementary statistical methods for foresters. Agric. Handb. 317. Washington, DC: U.S. Department of Agriculture, Forest Service. 87 p.

Froehlich, Henry A. 1978. The effects of soil compaction by logging on forest productivity. Final report to Bureau of Land Management, Portland, OR.

Holdorf, Herbert. 1982. Effects of site preparation and fuel management practices on soil productivity. In: Baumgartner, David M., ed. Site preparation and fuels management on steep terrain symposium proceedings; 1982 February 15-17; Spokane, WA. Pullman, WA: Washington State University Cooperative Extension: 63-65.

Jurgensen, M. F.; Arno, S. R.; Harvey, A. E.; [and others]. 1979. Symbiotic and nonsymbiotic nitrogen fixation in Northern Rocky Mountain forest ecosystems. In: Symbiotic nitrogen fixation in the management of temperate forests workshop: proceedings; 1979 April 2-5; Corvallis, OR: National Science Foundation: 294-308.

Roe, A. L.; DeJarnette, G. M. 1965. Results of regeneration cutting in a spruce-subalpine fir stand. Res. Pap. INT-17. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station. 14 p.

Smith, David Martyn. 1962. The practice of silviculture. 7th ed. New York: John Wiley and Sons. 578 p.

Speakers answered questions from the audience after their presentations. Following are the questions and answers on this topic:

Q. (from Jeff Collins) Your graph indicates that height growth is equal or greater on nonprepared sites compared to bulldozer piling. Would you consider not piling if you could not burn?

A. Yes, I would consider no treatment, if neither site preparation or hazard reduction are needed to meet objectives. Caution is warranted in placing strong interpretation on the nonprepped sites, due to the very small number of trees in those samples and the fact that inspection of those data shows that many of those trees are very old (70 year) advance regeneration.

Q. We heard earlier that tree height may not be a good indicator of fertility response in conifers. How might this observation affect the outcome of your study results?

A. In trees the age that were studied, growth as measured in leader growth is simple to estimate. This leader growth indicates the sum of all factors that are affecting the tree. One uniqueness of this study is that it involves many trees, in many stands, on many sites, from several years of treatment. These trees are demonstrating differences that are related to the different treatments. As I stated in my conclusions, we do not know precisely why these differences in growth are demonstrated, but we do know that it is related to piled versus burned sites. The reduction in height growth on piled sites, regardless of the reason, cannot be ignored. It does seem very probable that nutrient availability is a part of the differences shown, given our current understanding of the importance of organic material to forest soil nutrients and knowing the extent of scraping off of organic material in the bulldozer piling that was typical during the 1960's.

Paper presented at the Symposium on Management and Productivity of Western-Montane Forest Soils, Boise, ID, April 10-12, 1990.

Bob Bosworth is Fire Management/Silviculture Staff Officer, and Dan Studer is Forest Technician, Silviculture, Bonners Ferry Ranger District, Idaho Panhandle National Forests, Bonners Ferry, ID 83805.