On the Ecology of Mountainous Forests in a Changing Climate: A ...

More documents

Recommendations

Info

172 Chapter 7 European conditions is scarce. On the other hand, the influence of FORCLIM-S on the dynamics of FORCLIM-P appears to be stronger, although it is still less important than the coupling between FORCLIM-E and FORCLIM-P. Based on these considerations, the fact that most forest gap models developed to date ignore the turnover of soil organic matter but include a stochastic weather generator gets an empirical, quantitative underpinning. FORCLIM-S is the first submodel for belowground carbon and nitrogen turnover used in a gap model for central European conditions, and the simulation studies with it were encouraging. However, this submodel should be scrutinized carefully, several of its equations should be reformulated on a more mechanistic basis, and it should be validated extensively. Model improvements should deal primarily with adding a “slow” compartment (Parton et al. 1987, Verberne et al. 1990), and with a reformulation of the mineralization rate of the humus. Moreover, there is a serious problem of mapping the compartments in FORCLIM-S (“litter”, i.e. material that immobilizes nitrogen, and “humus”, i.e. material that releases nitrogen) to field measurements typically distinguishing “forest floor” and “mineral soil” primarily on a morphological basis (e.g. Vogt et al. 1986). In FORCLIM-E, the sensitivity of the drought stress index (Cramer & Prentice 1988, Prentice & Helmisaari 1991) to small changes of actual evapotranspiration raises the question whether it is robust enough to be used for parametrizing the ecological effects of drought on tree growth. It is suggested that further research should address this issue. Two aspects that have made forest gap models especially elegant could be maintained in FORCLIM: The simple representation of crown geometry, i.e. that all the leaves are concentrated at the top of the bole, and the lack of spatial interactions among forest patches. While the former assumption might have to be changed if boreal forests were to be simulated (Leemans 1992), the latter would have to be relaxed if migration phenomena were to be considered, e.g. for validation studies using pollen proxy data (Lotter & Kienast 1992), or if the spatial dynamics of landscapes were to be simulated (Urban et al. 1991). 7.3 Parameter sensitivity The analysis of the parameter sensitivity of FORCLIM-E/P/S revealed that the model is comparably robust to the values of its species parameters when they are varied within their range of plausibility. Thus the simulated species composition is not an artifact of ar-
Discussion 173 bitrarily chosen parameters, which increases our confidence that the simulation results obtained from FORCLIM represent reliable hypotheses on the forests under study. The parameter describing the tolerance to low nitrogen availability (kNTol) proved to be most important for determining the simulated species composition. This sensitivity calls for a careful scrutinization of the simple formulation used for modelling the effects of nitrogen availability on tree growth in FORCLIM (Pastor & Post 1985). With an improved formulation of this growth factor the coupling between FORCLIM-S and FORCLIM-P may become more important, which in turn would underline the significance of improved modelling of soil carbon and nitrogen turnover (e.g. Perruchoud 1994). The simulated species composition is also sensitive to the scaling constant in the tree growth equation (kG). Since tree growth is directly linked to competitive ability, this sensitivity appears to be quite realistic. The parameter determines tree growth during the early stage of tree life, where competition is especially strong due to extensive shading. The other parameters of the growth equation (kHm, kDm) have a stronger influence on older trees only. Competition for light is a major factor both in real and in model forests, and FORCLIM therefore is correct in producing a high sensitivity to the species parameters describing the tolerance of low light availability (kL y , kL a ). Thus we may conclude that the equation determining the maximum diameter increment is among the most sensitive parts of FORCLIM. These findings suggest that the basic assumptions such as the carbon balance of trees and the various allometric relationships used in the current growth equation should be scrutinized carefully. For example, the parabolic relationship between tree height and diameter may be questionable because it requires that diameter growth always comes to a halt when height growth ceases, which certainly is unrealistic and makes it difficult to estimate the parameters of the current growth equation (cf. Appendix II). Finally, the sensitivity analysis revealed that the precision of the biomass estimates obtained from FORCLIM is low, i.e. the abundance of a given species varies considerably depending on the values of the parameters used to characterize its natural history. It is interesting to view this finding within the framework proposed by Levins (1966): FOR- CLIM appears to be realistic (it produces plausible species compositions) and general (it is applicable under a wide range of climatological and ecological conditions), but the species composition simulated by FORCLIM is not precise.
Page 1:
Diss. ETH No. 10638 On the Ecology
Page 4 and 5:
ii Table of contents A BSTRACT.....
Page 6 and 7:
iv APPENDIX .......................
Page 8 and 9:
vi Harald BUGMANN, 1994: On the eco
Page 10 and 11:
viii Harald BUGMANN, 1994: Aspekte
Page 13 and 14:
1 1 . Introduction 1.1 Climatic cha
Page 15 and 16:
Introduction 3 1.2 Methods for the
Page 17 and 18:
Introduction 5 in a changing climat
Page 19 and 20:
Introduction 7 Their integrative ca
Page 21 and 22:
Introduction 9 (1984) provides a mo
Page 23 and 24:
Introduction 11 The main advantage
Page 25 and 26:
13 2 . Analysis of existing forest
Page 27 and 28:
Analysis of existing forest gap mod
Page 29 and 30:
Page 31 and 32:
Page 33 and 34:
Page 35 and 36:
Page 37 and 38:
Page 39 and 40:
The forest model FORCLIM 45 carbon
Page 41 and 42:
The forest model FORCLIM 47 TREE GR
Page 43 and 44:
The forest model FORCLIM 49 gBFlag
Page 45 and 46:
The forest model FORCLIM 51 Disturb
Page 47 and 48:
The forest model FORCLIM 53 decay o
Page 49 and 50:
The forest model FORCLIM 55 the est
Page 51 and 52:
The forest model FORCLIM 57 3.3 Mod
Page 53 and 54:
The forest model FORCLIM 59 Light a
Page 55 and 56:
The forest model FORCLIM 61 Overall
Page 57 and 58:
The forest model FORCLIM 63 D (cm)
Page 59 and 60:
The forest model FORCLIM 65 a) b) c
Page 61 and 62:
The forest model FORCLIM 67 Growth
Page 63 and 64:
The forest model FORCLIM 69 Stress-
Page 65 and 66:
The forest model FORCLIM 71 Tab. 3.
Page 67 and 68:
The forest model FORCLIM 73 3.3.2 F
Page 69 and 70:
The forest model FORCLIM 75 where k
Page 71 and 72:
The forest model FORCLIM 77 NITROGE
Page 73 and 74:
The forest model FORCLIM 79 peratur
Page 75 and 76:
The forest model FORCLIM 81 of degr
Page 77 and 78:
The forest model FORCLIM 83 microcl
Page 79 and 80:
The forest model FORCLIM 85 Tab. 3.
Page 81 and 82:
The forest model FORCLIM 87 3.4.3 F
Page 83 and 84:
The forest model FORCLIM 89 All the
Page 85 and 86:
The forest model FORCLIM 91 The mas
Page 87 and 88:
The forest model FORCLIM 93 FORCLIM
Page 89 and 90:
Behaviour of FORCLIM along a transe
Page 91 and 92:
Page 93 and 94:
Page 95 and 96:
Page 97 and 98:
Page 99 and 100:
Page 101 and 102:
Page 103 and 104:
Page 105 and 106:
Page 107 and 108:
Page 109 and 110:
Page 111 and 112:
Page 113 and 114:
Page 115 and 116: Parameter sensitivity & model valid
Page 147 and 148: 153 6 . Model applications Climatic
Page 149 and 150: Model applications 155 The simulate
Page 151 and 152: Model applications 157 6.2 Possible
Page 153 and 154: Model applications 159 simulation s
Page 155 and 156: Model applications 161 Simulation e
Page 157 and 158: Model applications 163 At the site
Page 159 and 160: Model applications 165 Bever Biomas
Page 161 and 162: Model applications 167 tainty inher
Page 163 and 164: Model applications 169 scenario cho
Page 165: Discussion 171 Finally, the analysi
Page 169 and 170: Discussion 175 comparably small imp
Page 171 and 172: Discussion 177 end of the 21st cent
Page 173 and 174: Conclusions 179 Ecological factors
Page 175 and 176: Conclusions 181 species composition
Page 177 and 178: References 183 Begon, M., Harper, J
Page 179 and 180: References 185 Burger, H., 1951. Ho
Page 181 and 182: References 187 Faber, P.J., 1991. A
Page 183 and 184: References 189 Huntley, B. & Birks,
Page 185 and 186: References 191 Leemans, R. & Prenti
Page 187 and 188: References 193 Olson, J.S., 1963. E
Page 189 and 190: References 195 Rudloff, W., 1981. W
Page 191 and 192: References 197 Smith, T.M., Leemans
Page 193 and 194: References 199 Whittaker, R.H., 195
Page 195 and 196: Appendix 201 II. Derivation of para
Page 197 and 198: Appendix 203 Tab. A-4: Tree species
Page 199 and 200: Appendix 205 Tab. A-7: Values for m
Page 201 and 202: Appendix 207 The minimum winter tem
Page 203 and 204: Appendix 209 Tab. A-11: Shade toler
Page 205 and 206: Appendix 211 Tab. A-14: Climatic pa
Page 207 and 208: Appendix 213 IV. Source code of the
Page 209 and 210: Appendix 215 FROM SimBase IMPORT De
Page 211 and 212: Appendix 217 END; (* IF *) prevDay
Page 213 and 214: Appendix 219 PROCEDURE InitializeFW
Page 215 and 216: Appendix 221 InstallSeparator( fMen
Page 217 and 218:
Appendix 223 FROM FCPFileIO IMPORT
Page 219 and 220:
Appendix 225 modelSp^.p.kHm := sens
Page 221 and 222:
Appendix 227 DeclMV( meanLitt[leafS
Page 223 and 224:
Appendix 229 Swiss Federal Institut
Page 225 and 226:
Appendix 231 (*********************
Page 227 and 228:
Appendix 233 BEGIN WITH sp^ DO d :=
Page 229 and 230:
Appendix 235 Definition module FCPB
Page 231 and 232:
Appendix 237 Purpose Simulation mod
Page 233 and 234:
Appendix 239 END Immobilization; PR
Page 235 and 236:
Appendix 241 CONST lem = 3; VAR ef:
Page 237 and 238:
Appendix 243 Purpose Provides the b
Page 239 and 240:
Appendix 245 Example of a text file
Page 241 and 242:
Appendix 247 Tab. A-15: Lower end o
Page 243 and 244:
Appendix 249 Tab. A-17: Percentage
Page 245 and 246:
Appendix 251 Tab. A-19: Significant
Page 247 and 248:
Appendix 253 VI. Derivation of para
Page 249 and 250:
Appendix 255 Tab. A-21: Species-spe
Page 251 and 252:
Appendix 257 Tab. A-22 (continued)
show all

On the Ecology of Mountainous Forests in a Changing Climate: A ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?