Papers by Katarina Merganicova
Ecological Modelling, 2011
Natural disturbances play a key role in ecosystem dynamics and are important factors for sustaina... more Natural disturbances play a key role in ecosystem dynamics and are important factors for sustainable forest ecosystem management. Quantitative models are frequently employed to tackle the complexities associated with disturbance processes. Here we review the wide variety of approaches to modelling natural disturbances in forest ecosystems, addressing the full spectrum of disturbance modelling from single events to integrated disturbance regimes. We applied a general, process-based framework founded in disturbance ecology to analyze modelling approaches for drought, wind, forest fires, insect pests and ungulate browsing. Modelling approaches were reviewed by disturbance agent and mechanism, and a set of general disturbance modelling concepts was deduced. We found that although the number of disturbance modelling approaches emerging over the last 15 years has increased strongly, statistical concepts for descriptive modelling are still largely prevalent over mechanistic concepts for explanatory and predictive applications. Yet, considering the increasing importance of disturbances for forest dynamics and ecosystem stewardship under anthropogenic climate change, the latter concepts are crucial tool for understanding and coping with change in forest ecosystems. Current challenges for disturbance modelling in forest ecosystems are thus (i) to overcome remaining limits in process understanding, (ii) to further a mechanistic foundation in disturbance modelling, (iii) to integrate multiple disturbance processes in dynamic ecosystem models for decision support in forest management, and (iv) to bring together scaling capabilities across several levels of organization with a representation of system complexity that captures the emergent behaviour of disturbance regimes.
Forest Ecology and Management, 2005
The impacts that stand treatments have on carbon, nitrogen and water cycles is a key question rel... more The impacts that stand treatments have on carbon, nitrogen and water cycles is a key question relating the sustainability of forests to individual forest management decisions. While the response of the above ground carbon cycle may be relatively easy to measure and is well understood, e.g. volume growth response due to thinning, the impact of stand treatment and different harvesting scenarios on the water, and nutritional status of the remaining trees and site as well as on the below ground carbon and nitrogen cycle is much more difficult to assess due to the difficulty of obtaining reliable data. An alternative that eliminates the need for detailed site-specific data is to use existing models that were explicitly designed for studying biogeochemical processes. In this paper we test and evaluate the applicability of a species-specific mechanistic biogeochemical model to assess the impact of different harvesting scenarios after thinning. Data are available from 36 Norway spruce stands covering three regions each with three harvesting scenarios after thinning and four replications. The different harvesting scenarios applied are: (1) whole tree harvest, (2) whole tree harvest after one vegetation period, i.e. without needles, and (3) commercial stem wood harvest. The modeling results suggest that the variant 3 exhibited the highest growth efficiency, a measure of net carbon uptake or growth, versus all other harvesting methods. This is confirmed by the available field observations. Nitrogen, an indicator of nutrient supply, behaved similarly. The important exception was nitrogen plant uptake, which was lower immediately after thinning on the sites where the lowest amount of biomass was removed. Significant relationships exist between observed versus modeled stand volume and volume growth suggesting that mechanistic modeling is a suitable diagnostic tool for analyzing the impacts of different management practices and thus, such models can be efficiently used to enhance silvicultural decision-making. # 2004 Published by Elsevier B.V.
Agricultural and Forest Meteorology, 2003
Based on daily weather records of minimum and maximum air temperatures as well as precipitation f... more Based on daily weather records of minimum and maximum air temperatures as well as precipitation from more than 250 stations between 1960 and 1998 across Austria we interpolate and validate daily climate interpolations using DAYMET. The current version of DAYMET interpolates on a systematic grid daily minimum and maximum temperature and precipitation from surrounding stations based on the principles of a weighted Gaussian filter. In addition, it calculates missing daily solar radiation (Srad) and humidity as it can be derived from temperature and precipitation data. In this study we calibrated DAYMET using the Austrian climate data base and developed a DAYMET point version, which allows us to interpolate daily weather for any location within the country, as it is needed to link existing field observations with missing weather data. We validated this procedure by using an independent data set of 23 stations located across the country. Our results can be summarized as follows: the sensitivity study using the full data set of about 3 million values of daily air temperature (minimum and maximum) as well as precipitation data indicated no regional or elevation related trends or biases. The only exception are daily precipitation predictions in very high altitudes (>1800 m), where model predictions diverge from observations probably due to an increase in the error of recording precipitation rates. The independent model validation using 23 stations consisting of minimum and maximum air temperature, precipitation, solar radiation observations as well as humidity data indicated no trends or bias. The mean error and the prediction interval, an indicator of the expected error range for future applications of the model, suggest no bias. Finally an assessment for two selected stations in Austria, Schmittenhöhe and Großenzersdorf, indicated a good coincidence between model predictions and observations using DAYMET.
Uploads
Papers by Katarina Merganicova