Paper presented at BioEnergy '98: Expanding Bioenergy Partnerships, Madison, Wisconsin, October 4-8, 1998.

ABSTRACT

Biomass energy is being considered as an alternative energy source for reducing dependence on fossil fuels, achieving greater energy independence, its environmental values and rejuvenating the farm sector. Short - rotation intensively managed plantations of genetically improved fastgrowing trees represent a substantial component of bioenergy feedstock supplies. In 1997, a genetic selection trial of forty willow and poplar clones was simultaneously established in central New York (NY) and north central Wisconsin (WI) to evaluate the relative performance of the clones and relate it to ecophysiological characteristics. These clones have good potential for use as commercial biomass energy crops. At the end of the first growing season, statistically significant differences (p<0.05) were found between clones in terms of height and diameter growth and clonal leaf area and biomass production. The clones also exhibited varying patterns of biomass partitioning and leaf area deployment. This paper reports on the relative performance of the clones based on first year performance and evidence is presented to support the use of foliar studies and environmental parameters in understanding the ecophysiological basis of biomass production.

Keywords: willow, poplar, clones, ecophysiology, specific leaf weight, productivity

INTRODUCTION

The twin problems of diminishing fossil fuel reserves (Campbell and Laherrére 1998) and the need for environmentally sound, renewable fuel supplies provides impetus for research and development in and the subsequent commercialization of biomass energy crops. The viability of bioenergy as a fuel requires the establishment of reliable sources of adequate and cost effective feedstock of biomass. Intensively cultivated energy plantations of woody or herbaceous biomass offer a means to attain this objective (Anderson et al. 1983; Lee et al. 1987; Börjesson 1996).

Research at the State University of NY College of Environmental Sciences and Forestry (SUNY ESF) has shown that willows and some hybrid poplar clones are well suited biomass crops for New York and the surrounding region (White et al. 1988). The Short Rotation Intensive Culture System (SRIC) used in this research is based on development work from Sweden involves growing willow and poplar clones at very close spacing (around 15,000 stems/ha) under intensive cultural management and for short coppice rotations of 3- 4 years. The guiding principle is to promote maximization of leaf area index and photosynthetic productivity, thus enhancing yield. Rapid canopy closure at the above spacing would also inhibit weed competition, a factor essential for high production.

In addition to identifying and evaluating clones capable of high biomass production and tolerance to environmental factors, it is also important that the processes that influence primary biomass production be better understood. In this regard, one of the key factors is to relate photosynthetic productivity to biomass yield and then develop indices of productivity (Larson and Isebrands 1971). Biomass yield is dependent on photosynthetic production and photosynthate redistribution and utilization (Michael et al. 1988). This process is a complex interplay of genetic, physiological and environmental factors. Information about the seasonal growth pattern of bioenergy crops under field conditions can help provide insight into these relationships. This, when combined with the knowledge of fundamental biological processes, can help in the development of growth models for predicting growth performance as a function of clonal characteristics, environmental and cultural conditions (Nilsson, 1985).

Primary Objectives

The current study has assessed the growth and production parameters for 40 clones during their establishment year from cutting, in both central NY and north central WI. This is the first such test of this type of genetic material in the north central United States in anticipation of expanding the SRIC system to that region. This work represents a collaborative effort between SUNY ESF and USFS North Central Experimental Station, Rhinelander with support from the Bioenergy Feedstock Development Program (BFDP) at Oak Ridge National Laboratory, Tennessee.

MATERIALS AND METHODS

In central New York, the planting site was in the village of Tully (42 0 47 ' 30 " N, 76 0 07 ' 30 " W) on SUNY ESF property. The soil was a palmyra gravelly silt loam (Glossoboric Hapludalf), a good quality agricultural soil with corn production generally in the range of 10. o.d.t ha -1 yr -1 ( Hutton and Rice 1977). In north central Wisconsin, the site was in the town of Rhinelander on US Forest Service property. Details of the Wisconsin site and the methods used are not reported here, but are available from the authors.

The existing vegetation on the site was killed in fall 1996 by applying Glyphosphate and 2,4 D at 1.0 and 0.25-0.5 kg active per acre respectively. In spring 1997, pre-emergent herbicide Goal (Oxyflurofen) was applied at 1kg pounds per acre. In December 1997, the trees were cut back to the ground to allow for coppice development. In the spring of 1997 at both sites, a randomized block design was established and 40 clones were planted at 0.9m X 0.6m spacing with 48 trees per plot. These clones represent diverse parentage (Table 1). Hybrid poplar clone NM6 cuttings that had been stored for a year were also planted in this trial. This plot was labelled as NM6(2yr). The clones at both sites were surveyed for survival, diameter, tree height and stem biomass at the end of their first growing season in late November 1997. Diameter measurements were made at 2.5 cm above ground. At the central New York site, ten clones with contrasting morphology and phenology were selected for more detailed study. In these plots, periodic destructive measurements of leaf area and biomass partitioning were conducted in the buffer rows. Weather conditions were recorded at a weather station at the site.

The performance of clones at the end of the season was evaluated using the general linear models procedure of analysis of variance followed by mean separations were done the Tukey's student range test (HSD). Cluster analysis was used to broadly classify the clones according to growth parameters. Discriminant analysis was done to validate the clusters and also identify the variables that constituted the Linear Discrimination Factors (LDF). All statistical calculations and analyses were performed at alpha=0.05, using SAS version 6.08 (SAS Institute 1985) and Microsoft EXCEL version 5.0 (Microsoft Corporation 1993).

RESULTS

Clone site comparison study

The results at the end of the establishment year from both the sites are summarized below. In Central New York during 1997 growing season, a total of 1470 growing degree days were accumulated and the total precipitation during the period was 50 cm. Data from Wisconsin is available from the authors. First year tree survival percentage for all clones was 88 % in Central New York and 97 % in Wisconsin. There was greater variation in New York, with five clones having survival percentages lesser than 75 percent.

Statistical differences were found among clones in height at the end of the season (Fig1). In central New York, mean heights ranged from 1.4 m for hybrid poplar clone NM6 to 0.20 - 0.30 m in the case willow clones S 646 and S566 (p=0.0001). In Wisconsin, mean heights ranged from 0.80 m for willow clone 94003 to 0.35m for hybrid poplar clones DN 34, DN5 and willow clone S365 (p=0.0001). Overall the clones had greater height in New York (Fig 1). In central New York, large differences in stem biomass production among clones were also observed (p=0.0001). Mean biomass per tree ranged from 0.85 kg in the case of poplar clone NM6 to around 0.10 kg in the case of willow clones S19, S566 and S646.

Figure 1. Mean Height at both sites

Based on the survival and tree growth data collected in central New York, cluster analysis was performed to identify groups of similar performing clones. The complete linkage method was used to provide the groupings (Aldenderfer and Blashfield 1990). The Cubic Clustering Criterion (CCC), pseudo t² and pseudoF plots of the data indicated that the clones could be grouped into homogenous clusters (Fig 2).

Figure 2. Clusters of clones and linkage distances between them

Discrimination analysis was used to validate the cluster analysis. Since the primary objective of the analysis was to separate the clones (not necessarily classify future observations) canonical discriminant analysis was used to obtain linear discrimination functions( (LDF) based on canonical variables (linear combinations of the quantitative variables). The two LDF's were significant based on the test of canonical correlations. Graphical representation of the two LDF's is presented in Fig 3. The clusters and their main characteristics are presented in Table 2.

Figure 3 Discrimination analysis plot of clusters along LDF1 and LDF2

Ecophysiology Study

The study was conducted on a set of ten clones from within the forty deployed.

Foliar characteristics

Leaf expansion and height growth began immediately after stem expansion from planted cuttings in late May and continued until cessation of growth in early September. Accelerated increase in leaf area occurred from mid August to late September.Clones retained leaves for varying periods of time, with willow clones SV1 and Pur12 retaining leaves through the first week of November. Maximum leaf area attained ranged from 0.12m 2 to 0.53m 2 for willow clones PUR12 and S365, respectively and 1.2m 2 for poplar clone NM6 (Table 3). Mean leaf area was significantly different (p=0.001). NM6 maintained consistently large leaf area through the entire season and willow clone PUR12 retained the least leaf area. In the case of all the clones; specific leaf area (SLA), the ratio of leaf area to leaf mass and leaf weight ratio (LWR), the ratio of leaf biomass to total above ground biomass fluctuated through the season. Leaf area ratio, a measure of the leaf area per unit stem weight decreased during the season, albeit to varying degrees depending on the clone. There was substantial variation in stem biomass production between the clones (p=0.001). Stem biomass ranged from 0.89 kg in hybrid poplar NM6 to 0.34 kg for willow clone PUR12.

DISCUSSION

Clone site comparison study Mean survival at both sites was excellent. The clones with relatively low survival at both sites were willow clones Streamco and S646. Additionally, in central New York, NM6 plots that had been planted with 2year old cuttings also had poor survival. There were substantial differences in the height growth between the two sites. Excessive weed infestation might explain the lower mean heights for the clones at Wisconsin. On the other hand in NY, weed control was excellent, site productivity is also greater at this site. At both the study sites willow clones 94001 to 94015 (S.pupurea) and poplar clones of the NM series had the best height growth.

The mean heights and biomass of first year poplar clones recorded in central New York compare well with those reported in other studies conducted in the USA (Gottschalk and Dickmann 1978) and Europe (Impens et al, 1988). Production of first year willow clones has not been previously reported. However it has to be noted that the performance attributes often change during the course of the rotation. Zsuffa (1975) states that the performance of poplar clones often change between the establishment and harvest years (Zsuffa, 1975). Thus first year growth or yield data should only be used to study patterns of growth rather than predict absolute yields.

The discriminant analysis (Figure 3) reveals that LDF1 represents the greatest proportion of the difference between the clusters. LDF1 represents all variables used (diameter, height, stem biomass and survival). LDF2 represents height and survival. However based on the standardised canonical coeffecients, it appears that stem biomass and diameter and to a smaller extent survival, are the main factors differentiating between the clusters. It is evident that the clusters do not differentiate between genera or species. Members of each species, type of cross and origin appear in all the clusters. However, most of the S. purpureas, and all the S. udensis and hybrid poplars occur within the top two clusters (2 and 3). The S. eriocephala clones exhibit greater variation and can be found in clusters 1 and 3. These clusters are likely to change as the trees age and juvenile - mature correlations of the clusters can be assessed.

Ecophysiology study

Leaf area has been shown to be one of the most important factors (Watson 1956, Isebrands et al. 1996) that determine biomass production. In this study, this relationship does not appear to be a direct linear one. In addition to leaf area, factors like leaf longevity, leaf thickness (Gottschalk and Dickmann 1978) and degree of efficiency in biomass partitioning between leaves and stem influence this relationship. For example, hybrid poplar clone NM6 deployed the maximum mean leaf area and also accumulated the highest above stem biomass. Whereas, willow clone SX61 with low leaf area, turned out to be one of the better biomass producers. Thus a better approach to understanding the relationship between photosynthesis and dry matter production may be to study the clone's growth strategy by examining all these factors in combination. As an illustration of this, the growth strategy of two clones is discussed below (Fig 4). Hybrid poplar NM6 exhibited the maximum mean leaf area and stem biomass production. This is further complemented by a high LWR ratio (70 %) early in the season. This ratio is initially high, however through the season it follows a decreasing pattern. These initial high values are an important factor in early establishment and growth. This feature has also been reported in some poplar clones (Gottschalk and Dickmann 1978). Specific Leaf Area (SLA) is the reciprocal of specific leaf weight (SLW), a measure of leaf thickness. Leaf thickness tends to increase with age. Crops having the same leaf area, but different leaf thickness can differ in productivity as thicker leaves have a larger photosynthetic surface when light is not limiting (Pieters, 1960). Conversely SLA decreases during the season and with age. Canopy dynamics also influence SLA and the presence of more sun leaves versus shade leaves reduces the SLA as shade leaves are thinner. Lower SLA often times is associated with increased photosynthetic activity (Djikstra, 1990). In NM6, SLA decreased during the season from around 14 m²/kg to around 11m²/kg.

In the case of willow clone SX61, a different growth strategy was adopted. Although the clone had very low leaf areas throughout the season, it was ranked as a moderate biomass producer. This can be explained by the SLA and LWR patterns. There was a considerable reduction in SLA from around 13 m²/Kg to 8 m² /Kg in mid august. Similarly, the LWR that was around 50 % early in the season changed to 30 percent by the end of the season. This implies that the clone was partitioning increasing amounts of biomass away from the leaves to the stem.

CONCLUSION

The results illustrate the variation in potential productivity and growth strategies of different willow and hybrid poplar clones. Results also indicate good potential for willow biomass growth in the north central region of the U.S.A and the utility of cluster analysis to group clones in productivity studies. A preliminary analysis of the leaf characteristics indicates that greater leaf area in combination with high SLW and efficient partitioning would contribute to greater stem biomass production during the establishment year. In addition to foliar and environmental studies, research is needed in the area of canopy light interception and seasonal mean light use efficiencies (Cannell, 1989). Knowledge of these parameters and their interaction will help devise improved growth process models, tree improvement efforts and plantation management techniques. This study, at both locations will be continued. At the end of the 1997, trees at both locations were cut and subsequent studies will examine coppice regrowth.

REFERENCES

1. Aldenderfer, M.S and R.K. Blashfield . Cluster analysis Series : Quantitative Applications in the Social Sciences . 44. Sage Univ. Paper.1990

2. Anderson D.C., H.W., Papadopol, C.S., Zsuffa, L. Wood energy plantation in temperate climates. Forest Ecology and Management. Vol 6 . pp 281- 306. 1983.

3. Börjesson, P.I.I. Energy Analysis of biomass production and transportation. Biomass and Bioenergy. Vol 11, No 4 pp 305-318, 1996

4. Campbell, C.J and Jean H. Laherrčre. The end of cheap oil. Scientific American, March, 1998

5. Cannell, M.G.R. Physiological basis of wood production: A review. Scand. J. For. Res. 4: 459- 490, 1989

6. Dikjstra. P.,Cause and effects of differences in specific leaf area. In. Causes and Consequences of variation in growth rate and productivity of higher plants. ed. anon. SPB Academic publishing, pp. 125- 140. 1990

7. Gottschalk, K.W and Donald I. Dickmann.. First year growth, biomass yield and leaf area development of four intensively cultured Populus clones in Southern Michigan. In. Proc. of the central hardwood forest conference. pp. 140 - 158. 1978

8. Hutton, F.Z., and C.E. Rice..Soil survey of Onondaga County. Central New York. U.S.D.A Soil Cons. Serv 235pp. 1977

9. Impens, I.I.. First and second year intensively cultured poplar cloned as related to leaf photosynthetic performance and radiation intercepting foliage canopy. In Proceedings of 5th International conference on Biomass and energy and Industry. ed G.Grassi, G. Gosse and G.dos Santos pp.1.439 - 1.445. 1990

10. Lambers, H.,. Does variation in photosynthetic rate explain variation in growth rate and yield. Neth. J. Agric .Sc., 35, 505-519. 1987

11. Larson, P.R and J.G Isebrands.. The relation between leaf production and wood weight in first year root sprouts of two Populus clones. Can. J. For. Res.. Vol 2, pp98-104. 1971

12. Lee, K.D., Gordon, J.C., Prominitz, L.C.,. Three year growth and yield of Populus hybrids grown under intensive culture. Biomass 13. 117- 124. 1987

13. Michael, D.A., Isebrands, J.G., Dickmann, D.I., Nelson, N.D..Growth and development during the establishment year of two Populus clones with contrasting morphology and phenology. Tree Phys. 4, 139- 152. 1988

14. Microsoft Corporation,. Microsoft EXCEL Version 5.0 Edition. Redmond, Washington, 1993

15. Nilsson, L O. Growth and yield of Willow stands in relation to climate and nutrition. Dissertation. Report 21. Swedish Univ. of Agric.Sciences, Uppsala. Sweden. 1985.

16. Pieters, G.A. On the relation between the maximum rate of photosynthesis and the thickness of the mesophyll in sun and shade leavs of Acer pseudoplanatus. L. Meded. Lanbouwhogeschool, Wageningen 60 (17): 1-6. 1960.

17. SAS Institute, SAS User's Guide: Statistics, Version 5Ed. SAS Inst. Cary, NC. 1985.

18. White.E.H., Abrahamson, L.P., Gambles.R L. and Zsuffa.L.,. Experiences with Willow as a Wood Biomass Species. In Energy from Biomass and Wastes XII. ed D.L.Klass. Inst. of Gas Technology. New Orleans LA, PP. 125-152. 1988

19. Zsuffa, L.. Some problems of hybrid poplar selection and management in Ontario. For Chron., 51: 240- 242. 1975

ACKNOWLEDGEMENTS

This research project was conducted with financial assistance (Sub contract no: 19X -SW561C) from the Bioenergy Feedstock Development Program at the Oak Ridge National Laboratory, Tennessee.