Cut or keep: What should a forest owner do after a windthrow?
Introduction
Windstorms are one of the most important risks for forest management. The impacts of catastrophic events and the risk of total destruction on the optimal rotation age have been addressed by previous works focussing on even-aged forests (e.g. Martell, 1980, Routledge, 1980, Reed, 1984). In these earlier works three major simplifying assumptions have been used to obtain analytical expressions for the model’s optimal solutions: (i) The probability of a catastrophic event is exogenous; (ii) whenever a catastrophic event hits the forest stand, it is completely destroyed as a result, hence damages are also exogenous; and (iii) the salvage value after the catastrophe is equal to zero.
While the assumption of exogeneity of a catastrophic event is rather easy to justify, the other two are harder to meet in the real world. Therefore, several studies have tried to relax these assumptions, often through numerical methods.
Valsta (1992) relaxed the assumption of total destruction of the stand by using an uniformly-distributed random variable to represent the fraction of the stand destroyed, and assumed a salvage value. Haight et al. (1995) introduced an age-dependent damage risk and a salvage proportion. Successively, Thorsen and Helles (1998) computed the optimal rotation length and thinning regime using an endogenous damage function that depends on the forest’s characteristics as well as on the forest management (i.e. thinning operations). While Thorsen and Helles (1998) and Haight et al. (1995) assumed an exogenous land value, Loisel, 2011, Loisel, 2014 computed the optimal rotation age and thinning regime that maximise the land expectation value (LEV) – hence taking into account the value of successive rotations – under the risk of destructive events while accounting for the salvage value of windthrown timber.
In an evaluation context, rather than in an optimisation one, Bright and Price (2000) proposed a simple and elegant way to compute the expected LEV under the risk of hazards whose realisation implies the end of the rotation. A similar methodology was applied by Deegen and Matolepszy (2015) to compute site-dependent optimal rotation ages and expected LEV based on historical management data in Germany.
All these studies have assumed total clearing of the stand after a windthrow, independently of the level of damage. However, following a windthrow, it might be profitable to keep the standing trees until the “prescribed” end of the rotation. The continuation of the rotation after a disturbance was firstly discussed by Xu et al., 2016a, Xu et al., 2016b, who analysed the forest management problem under the risk of two sequential disturbances. They tested three alternative strategies: (i) the systematic clearing and replanting of the stand after the first disturbance (the common assumption in previous literature); (ii) keeping the standing trees after the first disturbance and cutting and replanting after the second disturbance; (iii) keeping the standing trees also after the second disturbance2 and harvest the stand at maturity. Their results showed that ignoring the continuation options could lead, under some circumstances (i.e. disturbances’ arrival rates, survival probabilities and post-disturbance tree-growth losses), to suboptimal harvest decisions and land expectation values.
In their work, Xu et al., 2016a, Xu et al., 2016b assumed and compared the outcomes of fixed continuation strategies that are kept constant along the planning horizon. However, whether keeping the standing trees is profitable or not will ultimately depend on the cost structure as well as the characteristics of the stand after the windstorm.
In this paper we relax the hypothesis of a fixed continuation strategy and introduce a condition to determine when it is profitable to continue with the remaining trees until maturity and, alternatively, when it is best to cut them immediately and replant the whole stand. We assume (i) an exogenous rate of return for windstorms as in previous works, as well as (ii) an endogenous damage function determining the fraction of windthrown trees after a storm; (iii) a positive salvage value for the windthrown trees, and (iv) explicit clearing costs.
In Section 2 of this paper we present the theoretical model and in Section 3 we introduce the cut-or-keep decision rule. In Section 4 we calibrate our model with real data for maritime pine in Landes Region in France and run a Monte Carlo simulation. Sections 5 and 6 are devoted to presenting and discussing the results.
Section snippets
Stand dynamics and harvesting
Let us assume a forest owner who plants a mono-specific and single-aged forest stand3 at the beginning of the planning horizon, . The forest owner begins by planting at that time, and the plantation costs are represented by . The age of the forest stand – denoted by – is equal to zero both at the initial period (i.e. ) and the year after every harvest. Time is discrete and the
Cut-or-keep rule
Most papers modelling windstorms and wind damage make the simplifying assumption that whenever a storm hits the stand either all or none of the trees are windthrown. A few papers allow for a fraction of the trees to be windthrown, but after the windstorm the remaining standing trees are felled –by default– and a new rotation begins. This assumption is both restrictive and unrealistic, but makes modelling simple and analytical solutions possible. For the sake of realism, in our model, only a
Model calibration and Monte Carlo simulations
The model described above cannot be solved analytically because of both the non-linear nature of the stand dynamics and the use of the cut-or-keep condition. Hence, we solve the model numerically using Monte Carlo technique with MATLAB. The code is organised as shown in Table 1. For a given rotation age, the model simulates the evolution of the stand for 300 years projecting the stand growth, harvesting operations and windstorms. The code then sums all the discounted revenues and costs
Results
Using Monte Carlo technique we ran simulations for a number of scenarios with different windstorm return rates (i.e. 0%, 1%, 2% and 4%). For all these scenarios we computed both the (approximated) mathematical expectation of the Land Expectation Value () and the optimal rotation age. Two cases of figure were considered: (i) when no decision rule is applied, which means that after a windstorm the whole stand is, by default, cut down and then replanted; and (ii) with the cut-or-keep
Discussion
Whenever a windstorm hits a forest stand it may create severe economic damages. The importance of these damages is a function of the wind speed and the trees’ height. We model the decision problem that a forester faces after a windstorm hits the stand: The owner may decide to remove both the windthrown trees and the standing trees and replant or, alternatively, continue with the standing trees until maturity. We propose a new decision rule that allows to compare the economic gains of each
CRediT authorship contribution statement
Claudio Petucco: Conceptualization, Methodology, Software, Formal analysis, Writing - original draft, Writing - review & editing. Pablo Andrés-Domenech: Conceptualization, Methodology, Writing - original draft, Writing - review & editing, Supervision. Lilian Duband: Methodology, Formal analysis.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors appreciate the helpful suggestions and comments of Robert Haight (USDA Forest Service – Northern Research Station), Anne Stenger (Université de Lorraine, Université de Strasbourg, AgroParisTech, CNRS, INRA, BETA) and Yves Ehrhart (LERFob, AgroParisTech – INRA). The authors would also like to thank Alexandra Niedzwiedz at the Forest Economics Observatory for the support with the empirical data. The UMR 1443 is supported by a grant overseen by the French National Research Agency (ANR)
References (27)
- et al.
Risk aversion and optimal management of an uneven-aged forest under risk of windthrow: a Markov decision process approach
J. For. Econ.
(2016) - et al.
Economic balancing of forest management under storm risk, the case of the Ore Mountains (Germany)
J. For. Econ.
(2015) - et al.
How does forest composition and structure affect the stability against wind and snow?
For. Ecol. Manage.
(2017) Faustmann rotation and population dynamics in the presence of a risk of destructive events
J. For. Econ.
(2011)Impact of storm risk on Faustmann rotation
For. Policy Econ.
(2014)- et al.
Optimal spatial harvest planning under risk of windthrow
For. Ecol. Manage.
(2001) The effects of the risk of fire on the optimal rotation of a forest
J. Environ. Econ. Manage.
(1984)Economics of rotation and thinning revisited: the optimality of clearcuts versus continuous cover forestry
For. Policy Econ.
(2016)- et al.
Optimal stand management with endogenous risk of sudden destruction
For. Ecol. Manage.
(1998) - et al.
Optimal forest management with sequential disturbances
J. For. Econ.
(2016)
Valuing forest land under hazards to crop survival
Forestry
Volatilité des cours du bois par essence et qualité: perspectives pour la gestion forestiere
Revue Forestiere Française
Berechnung des Werthes, welchen Waldboden, sowie noch nicht haubare Holzbestände für die Waldwirtschaft besitzen
Allgemeine Forst- und Jagd-Zeitung
Cited by (6)
Faustmann formula and its use in forest asset valuation: A review and a suggestion
2024, Forest Policy and EconomicsThe Management Response to Wind Disturbances in European Forests
2021, Current Forestry Reports
- 1
The contribution of Claudio Petucco in this paper is the result of his PhD thesis work carried out at the former Laboratory of Forest Economics (INRA, AgroParisTech), now BETA (UMR 1443, Université de Lorraine, Université de Strasbourg, AgroParisTech, CNRS, INRA).