Introduction 

There is a growing increase in the impact of elevated atmospheric CO₂ on forest trees and forest ecosystems. This is not surprising as forests cover some 27% of the total land surface but account for some 70% of terrestrial net primary production ([27]). Moreover, more than 85% of the total plant C on earth and between 60-70% of the total soil C is contained in forests ([10]). The atmospheric concentration of CO₂ is currently increasing at the rate of 3.3 billion tons of C per year ([17]). Very good reviews have been published about the consequences of elevated atmospheric CO₂ on forests ([5], [34], [6], [19], [23]). Ceulemans et al. ([6]) argue that the question is not so much by what we do not know, but rather how to best integrate our knowledge in order to predict the performance and productivity of future ecosystems to global climatic changes. Newman et al. ([29]) have the opinion that our current ability to detect and predict changes in forest ecosystem productivity is constrained by several limitations. These include a poor understanding of belowground productivity, the short duration of most analyses, and a need for greater examination of species or community-specific variability in productivity studies. From studies to date, we know that the life-long above ground growth response of forest trees cannot be accurately predicted from short-term experiments ([30]). Karnosky ([19]) suggests long-term studies using free-air CO₂ enrichment (FACE) technologies or forest stands around natural CO₂ vents so as to increase the knowledge base on forest ecosystem responses to elevated atmospheric CO₂. There is no doubt that FACE experiments can offer valuable information. However, there are some shortcomings with those experiments. First is the representativity. If the FACE experiments try to cover the whole range of forest ecosystems types, the cost will be tremendously high. The second is that FACE experiments will have a higher rate of CO₂ enrichment than the actual rising rate of CO₂. Forest stands around natural CO₂ vents are rather limited. The establishment of new plots in natural atmosphere has, of course, the disadvantage of the very long-term expectancy of conclusions. Hopefully, there are already experimental plots, which have a past of ecological measurements. These are the intensively monitored plots (800 in total) which were installed in Europe under the auspices of the ICP-Forests ([16]). Those plots were installed when concern about the threat of atmospheric pollution on forest vitality was widespread. In this work I will try to stress the importance of keeping monitoring these plots. More specifically, I will focus on the potential contribution the intensively monitored plots can offer to meet the knowledge gaps with regard to tree responses to elevated atmospheric CO₂. These knowledge gaps were set by Karnosky ([19]). The measures suggested concern long-term monitoring so they must be as cost efficient as possible.

  Tree growth 

It has been found that short-term assimilation of CO₂ is significantly stimulated by increased CO₂ in nearly all plant species ([34], [25]). However, in the longest study (25-30 years) of continuous exposure of forest trees to elevated atmospheric CO₂ with forest stands of holm oak growing in the vicinity of two natural CO₂ springs in Italy, it was found that the trees showed a moderate, age dependent increase in stem biomass production, but had significantly lower biomass of 6-year-old branches, decreased branching, and lower leaf area per unit branch biomass, compared to control trees at a nearby site ([13]). In the intensively monitored plots growth has been measured since 1995. At that time tree diameter was measured either with measuring tapes or calipers. Recently, during the Pan European FutMon project (⇒ http:/­/­www.­futmon.­org/­) some diameters were measured with girth bands. These bands are relatively cheap and very accurate. This kind of measurement can be extended in all plots to include many more trees. In some years there will be valuable results all over Europe.

  Water balance 

Long-term studies of forest trees in an enriched atmosphere with CO₂ have shown a significant 21% decrease in stomatal conductance ([26]). However, as tree biomass increases the question remains whether trees use more or less water even if stomatal conductance decreases. The installation of soil moisture sensors in many plots help in the calculation of the whole water balance in forest stands. The actual magnitude of transpiration cannot be calculated but a comparison between forest stands can be made. For instance, for the same amounts of rainfall and interception the stand with the less water loss in drainage (assuming there is no surface runoff) will have a higher transpiration rate. The soil water holding capacity of each stand will be a covariate in the statistical comparison.

  Soils 

An important aspect that has not been tackled so far by the ICP-Forests projects is the form of C in soils. Elevated atmospheric CO₂ will bring about a stimulation of soil respiration, which can be much higher than the enhancement of root biomass ([4]). A major consequence of the increase in microbial activity and consequently in CO₂ production is a potential negative effect on the accumulation of organic C in soils and thus C potential sequestration of soils ([15]). However, there is a plateau beyond which organic matter cannot be easily attacked by microbes. Mineralization of soil organic matter strongly depends on the size of the labile C pool in addition to the effects of the microbial community and abiotic environmental conditions like temperature and moisture ([7]). A simple method to measure the labile C pool in soils is by water extraction ([18]). Although water extractable organic C makes up only a small portion of total soil organic matter (usually less than 1% - [3]), it might provide a measure of microbially available C. Zhao et al. ([37]) found a strong relationship between water extractable organic C and the rates of C mineralization using arable and forest soils.

In the intensively monitored plots the two soil surveys have a distance of 10 years from each other (1997 and 2007). Most counties, if not all, have stored the samples so they can be reanalyzed. If a third survey takes place in another 10 years, there will be a set of data with the amount of microbially available C covering a time space of 30 years. Comparisons can be made having in mind a fourth or fifth survey for the next generations.

  Plant nutrition and nutrient cycling  

The growth response of forest plants to the rising concentrations of CO₂ depends on their ability to acquire soil nutrients and water ([12]). Nutrient imbalances are reflected in foliar analysis. Every two years foliar analysis is carried out in all plots. A valuable data bank has been formed and should be enriched in the following years.

It has been well documented that the nitrogen level in the foliage of trees growing under elevated atmospheric CO₂ generally decreases ([24]). As an example for repeated measures analyses, Fig. 1 shows N concentrations in fir foliage over time in one of the intensively monitored plots in Greece ([28]). In that particular case no significant changes were observed.

Fig. 1 - Nitrogen concentration in current year’s needles of fir (Abies borisii regis)in Greece.

  Enlarge/Shrink   Download   Full Width  Open in Viewer

Nitrogen concentrations also decrease in litterfall ([31]). By contrast to N concentration, the quantity of litterfall increases 20-30% under elevated CO₂ ([9]). It is not known how nutrient mineralization will change due to the higher quantity of CO₂ in soil. In the intensively monitored plots litterfall is monitored. Litterfall quantities are measured and elemental analyses are carried out. A simple test is a time series comparison between the successive N concentrations in foliage and foliar litterfall.

If the N content in litterfall decreases so must decrease the litter decomposition rates and the N mineralization rates. Both of them can be measured with the classic method of the litterbags technique. This is a relatively easy technique that can be employed in the future.

  Belowground carbon allocation 

Allocation of C to belowground plant structures often equals or exceeds aboveground litterfall C and aboveground respiration in forest ecosystems, making it the single most important fate for gross primary productivity ([20]). Despite its importance, total belowground allocation (TBCA) remains poorly quantified because it is difficult to quantify root and mycorrhizal processes by any method ([11], [14]). If growth is enhanced, there will be more C input into the soil. Therefore, a major indirect response to an increase in atmospheric CO₂ consists in the greater below-ground C allocation through root exudation and turnover, which is likely to lead to changes in the size and the activity of soil microflora ([22]).

In the absence of direct measurements of TBCA, Raich & Nadelhoffer ([33]) and Davidson et al. ([8]) proposed the following equation (eqn. 1):

where soil respiration is the sum of root respiration, root litter C decomposition and aboveground litter C decomposition.

A critical assumption of this approach to estimate TBCA is that that the ecosystem is at a steady state, which means that the annual inputs of C below ground are equal to annual rates of decomposition.

If the system is not at a steady state the previous equation becomes (eqn. 2):

where ΔC_soil, ΔC_litter, ΔC_root are the changes in C stocks of mineral soil, forest floor and root biomass, respectively, and C_xport is C loss through leaching. From the above equation the parameters R_soil, C_litterfall and C_export in soil solution can be measured continuously in the intensively monitored plots. ΔC_soil and ΔC_litter can be measured in the following soil survey and compared with the survey in 2007. ΔC _root can be calculated from an equation between ABD (aboveground biomass density) in Mg ha^-1 and RBD (root biomass density) also in Mg ha^-1.

Cairns et al. ([2]) used the following equation (eqn. 3):

The above equation was based on 160 studies covering tropical temperate and boreal forests. The use of this equation is encouraged by other workers ([1]).

The total above ground biomass includes foliage apart from woody biomass. There are equations connecting foliage and tree diameters and heights for the main forest species ([21]), so that all components of the equation can be found.

In order to apply the above equations, the soil respiration rates have to be measured. It is probably the only expensive measurement, but it will help in drawing valuable conclusions, as soil respiration depends on soil moisture and temperature repeated measures have to be taken in time and space. Vincent et al. ([35]) used soil respiration chambers on which a portable infrared gas analyser was connected. They measured respiration in daytime during the growth period in a temperate deciduous forest. Raich ([32]) measured soil respiration with the soda-lime technique in three Hawaiian rain forests in a whole year period. The soil respiration fluxes should be expressed as the littefall fluxes, i.e., in kg ha^-1 yr^-1. So a whole year of measurements is necessary. The number of samplers (chambers) depends on the variability of respiration rates and the required probability level. A trial can be carried out with a limited number of samplers after which the required sample size for a given confidence interval can be calculated according to statistical formulas ([36]).

In my opinion, the first simpler equation may be used in most cases and the second one may be applied when there is a third soil survey in the intensively monitored plots.

  Phenology 

Phenology is the study of the periodicity of leafing, flowering and fruiting in plants in relation to climate and other environmental factors. There is the hypothesis that elevated CO₂ can affect development of leaf area in the spring so that trees could potentially benefit from an earlier onset of C assimilation at the start of the growing season. This could be an important factor in the expansion of tree populations into areas currently too cold for their growth ([34]). Phenological observations are carried out regularly in the intensively monitored plots and if leaf area increases that will be written down. What I now suggest is the extension of phenological observations to mycorrhizal fungi. This is crucial because the plants’s ability to acquire soil resources in an elevated CO₂ environment is mediated by symbiotic associations with mycorrhizal fungi. A new distribution of mycorrhizal fungi may not be beneficial to plants as the plant-fungi symbiosis was evolved over time due to natural selection.

  Conclusions 

The intensively monitored plot can be a valuable tool in assessing the impact of elevated CO₂ on forests. The activities already existing should be complemented by the following actions:

1. Tree growth is suggested to be monitored with girth band in all plots;
2. A third soil survey must be carried out. In this survey, together with stored samples of the previous ones, the labile C should be determined by water extraction.
3. Soil respiration must be measured so as to draw conclusions on the C allocated belowground.
4. Organic matter mineralization rates should be determined in all plots.
5. Phenology observations must include the distribution of mycorrhizal fungi.

  References