Forest–water dynamics within a mountainous catchment in Austria

Abstract

In this study, we assessed the role of forests in the local water budget within a 7.3 km² catchment that has measured river runoff and a long history of forest exploitation and disastrous flooding and debris flow events. Forests retain and absorb water from turning into runoff which is a possible trigger for catastrophic events. Forest water budgets (i.e. interception, transpiration, evaporation, sublimation, soil water storage and outflow) interact with ecosystem processes that are related to the carbon, nutrient and energy cycles and consequently affect forest growth rates. Therefore, we employed a biogeochemical–mechanistic ecosystem model, Biome-BGC, as a diagnostic tool to evaluate the dynamic relationships between key forest ecosystem characteristics and the water cycle. Our study was conducted in the Schmittental catchment with about 70 % forest coverage, situated in the Greywacke Zone of the Austrian Alps. Using stand and site information from 21 Norway spruce stands from the region and 29 years of total catchment runoff data for model validation, we demonstrated that the process-based ecosystem model mimics the interaction of forest growth and the water budget realistically. The weekly catchment runoff calculations based on Biome-BGC grid simulations compared well with observed runoff data. The analysis of the forest–water dynamics/relations showed that the water budget is affected by the size of the canopy and the physiological canopy behaviour in response to daily weather. The results suggested that for fully stocked stands and with a standing volume of >250 m³ ha⁻¹, forest water outflow was minimised.

Appendix

Maximum stomatal conductance (g _{s_max}, Table 3) is limited with multipliers (m) [0,1] that depend species specifically on VPD, soil water potential (Ψ) and daily minimum temperature (Tmin) and generally on intercepted solar radiation (Tables 4, 5, Eqs. 1, 2). For VPD below and Ψ and Tmin above a certain threshold level (_open), the multiplier is 1, and thus, there is no conductance reduction; for VPD above and Ψ and Tmin below another threshold level (_close), the multiplier is 0, and thus, there is complete stomatal closure. In between the two threshold levels, the conductance multiplier for VPD and Ψ changes linearly (Table 5, Eqs. 3–5). The solar radiation multiplier is calculated separately of the sun and the shade canopy. It considers the amount of radiation intercepted in the canopy (PPFD _pLAI, photosynthetically active photon flux density per projected LAI, μmol m⁻² s⁻¹). This hyperbolic relationship needs the PPFD ₅₀ value, giving the level of PPFD_pLAI for which m _Srad is 0.5, defined as 75 μmol m⁻² s⁻¹ (Thornton 1998). The intercepted radiation (PPFD_pLAI) differs between sun and shade leaves; thus, the conductance multiplier is calculated separately for sun and shade leaves (Table 5, Eq. 6). Consequently, also m _total (Table 5, Eq. 2) is calculated separately for sun and shade canopy. Final leaf-level conductance to transpired water (g _trans) uses an electric circuit analogy and is stomatal conductance (g _s) and cuticular conductance (g _c) in parallel and boundary conductance (g _bl,Table 3) in series (Table 5, Eq. 7).

Table 3 Selected ecophysiological constants relevant for the water and carbon budgets

Table 4 Equations for soil water potential and volumetric water content in Biome-BGC (Thornton 1998)

Table 5 Equations for the regulation of stomatal conductance in Biome-BGC (Thornton 1998)