Weather conditions and phenological observations
Compared with weather conditions in the Wallis over the last 20 years, 2002 was wetter than the average, 2003 was extraordinarily dry and hot, and 2004 had one of the lowest amounts of rain during the vegetation period but the air temperature (TAir), the mean vapour pressure deficit (VPD), and the soil water potential (
Soil) were closer to 2002 than to 2003 (Table 2). In the ‘wet’ year 2002, there was a distinct drought period in the last three weeks of June which led to very low soil water potentials (–500 kPa). At the end of this period several rain events rehydrated the soil to moderate Soil-values for the rest of the summer. The weather conditions in 2003 led to large water deficits in the soil and plants from March onwards. A considerable number of trees of all species showed morphological reactions at both sites, but this was more obvious at Salgesch. At Salgesch, leaves of Q. pubescens turned yellow and had already been shed by July. Pinus sylvestris and P. abies shed more needles than usual in August and September at both sites (data not shown). During the vegetation period in 2004, the soil was wet in spring and remained moderately dry until mid-summer, despite little rain. From July onwards Soil dropped to low values ( kPa), but these climatic conditions did not lead to early leaf senescence as in 2003.
Intra-annual radial growth in 2003
In 2003, intra-annual radial growth of the ten trees was assessed with two different methods: (i) wood coring that allowed measurement of growth increments over 1–2 week periods throughout the season, and (ii) continuous stem radius measurements (
R) that allowed the estimation of continuous growth rates, but with reservations about the water-related component of the fluctuations. Both methods were able to detect the course of intra-annual radial growth and led to similar results (Fig. 2). Growth deduced from R included the increment in xylem and phloem cells. This led, in general, to larger growth estimates compared with the analysis of the wood cores which included the xylem only. Slight differences occurred for the date of the initial start of the growth (Fig. 2g, h) or the shape of the growth curve (Fig. 2c, e). Nevertheless, both methods allowed the growth patterns of Q. pubescens and the two conifer species, P. sylvestris and P. abies, to be distinguished (Fig. 3).
Radial growth of Q. pubescens began earlier in the season than that of P. sylvestris and P. abies despite the fact that the conifers had a high T PET–1 about 1–2 months before Q. pubescens. At both sites, about 25–30% of the total annual radial increment of Q. pubescens had been achieved by the date of budburst and 40–45% of the tree ring was already formed when the leaves reached a state of full expansion and thus full physiological functionality (Table 4). The time lag between the beginning of radial growth and budburst was longer at Jeizinen than at Salgesch. By contrast, initial radial growth of P. sylvestris and P. abies only occurred shortly before budburst (Fig. 3) and radial growth had reached no more than approximately 15% of the annual radial growth at this date.
The tree rings of the year 2003, analysed from the wood cores, were, on average, wider at Jeizinen than at Salgesch for both species. The reason for the wider tree rings of P. sylvestris at Jeizinen was not due to a higher number of cells but to a larger cell size. Q. pubescens formed, on average, one row of early wood cells at Salgesch but two to three at Jeizinen. The number of late wood cells in Q. pubescens were not counted because of the very small cell sizes and the irregular structure of the wood.
Radial growth from 2002 to 2004
The typical growth pattern observed in 2003 also appeared in 2002 and 2004 (deduced from R): Q. pubescens always started its growth period before the two conifer species and, at the time of full leaf expansion, a relevant part of the current tree ring was already built (Fig. 4). The extremely hot year 2003 led to smaller tree rings in all species compared with 2002 (Tables 5, 6) but the shapes of the growth curves did not differ from those of the other two years (Fig. 4).
The growth period of Q. pubescens was, on average, 97 d (i.e. 90% of the yearly increment, averaging over the years 2003 and 2004) and longer than that of P. sylvestris (60 d) and P. abies (64 d). Within this growth period of Q. pubescens, 55 d overlapped the 123 d when T PET–1 was >20% and thus a significant net CO2 assimilation was likely. The 55 d is equal to 45% of the ‘productive days’. Pinus sylvestris had the smallest ratio of ‘growth days’ to ‘productive days’ with 60 d out of 189 (32%) and the largest ratio was for P. abies with 64 d out of 95 (67%).
Impact of climatic conditions on radial growth
The potential impact of climatic conditions on the annual radial growth of P. sylvestris and Q. pubescens was investigated by comparing annual growth increments with the climatic conditions during the growth period over 7 years at Salgesch. The annual radial growth of both species responded significantly (Table 5) to variations in total rainfall during the growth period (Fig. 5): the more rain a growth period of a certain year had, the bigger the radial growth was. Over 90% of variance was explained with this factor (P Soil (Q. pubescens and P. sylvestris), low VPD (P. sylvestris), and low W (P. sylvestris) also had a positive impact on growth, but were less significant than rain (Table 5). No clear trends were found for temperature and radiation (P >0.1).
The same pattern was found for intra-annual growth at Salgesch and Jeizinen, analysed by dividing the growth periods into subsets of 10 d. The corresponding microclimatic factors were related to the growth increments. Again, rain strongly determined radial growth rates for Q. pubescens, P. sylvestris, and P. abies at both sites (Fig. 6; Tables 5, 6). The actual water status of soil (Soil) and tree (W) did not seem to have a strong impact on short-term growth and factors such as radiation and VPD did not show a consistent pattern (Tables 5, 6). Even small rain events in dry periods accelerated radial growth, sometimes without wetting the soil (Fig. 7) or substantially rehydrating the tree.
Nevertheless, more than 90% of the growth increment was strongly limited to the growth period of April to June: before or after this period very little growth occurred, even when there were wet conditions.
Physiological processes and radial growth
The timing of growth and the corresponding growth patterns were similar for all species, compared with the big differences that occurred in the pattern of T PET–1 between them. Obviously, T PET–1 of Q. pubescens was strongly determined by the seasonality of its leaf development. High ratios started about 2 weeks after budburst when the leaves were fully expanded (Table 4) and ended with leaf senescence in autumn (Fig. 4). With the exception of the summer of 2003, T PET–1 of Q. pubescens was less dependent on the actual microclimatic conditions than those of the conifers. The ratio T PET–1 of P. sylvestris followed W closely but hardly ever ceased completely across all four seasons, except during a couple of freezing days in wintertime and under very dry and hot conditions in the summer of 2003. Despite having evergreen needles, P. abies seemed to be unable to maintain a high T PET–1 in the winter of 2003/2004, but retained some activity in the first half of the winter 2004/2005. Both coniferous species had high T PET–1 in spring, several weeks before budburst (Fig. 4; Table 4).
The results of the effect of high T PET–1 of the tree species on their radial growth were not consistent. Intra-annual growth (10 d data) appeared to be largely independent of the actual T PET–1 for all three species at both sites (Tables 5, 6). In contrast to these results, the annual growth increments corresponded with increased T PET–1 before (P. sylvestris) and during the growth period (P. sylvestris and Q. pubescens) (Fig. 5). Q. pubescens seemed to be less influenced by T PET–1 during its growth phase than P. sylvestris.
Answers to all your Biology Questions
