Resource assessment techniques for Continuous Cover Forests systems

Abstract

From a statistical point of view continuous cover forest (CCF) systems are heterogeneous populations, whose attributes show high variability and diversity. While information needs for homogeneous, even-aged, single species stands can easily be satisfied by providing information on statistical key parameters, CCF systems render information on non-timber attributes describing forest structures necessary. Beside statistical point estimates information on spatial patterns is crucial for managing CCF systems and describe their ecological situation.

This paper summarizes selected assessment methods for CCF systems. The Winkelmaß is an index of aggregation that can easily be applied for the assessment of spatial patterns. Adaptive cluster sampling is a field based assessment scheme focusing on rare species. Remote sensing techniques are ideal tools for providing geo-referenced, wall-to-wall information but only a limited set of attributes can be derived from currently available image products. Hyperspectral remote sensing data offer the possibility to handle mixed pixels, which are due to their heterogeneous structures frequent in CCF systems. Combining remote sensing and ground data by means of the kNN-method extends the set of attributes that can be mapped. Geo-statistical methods allow to provide mapped information for attributes which are assessed by field samples only.

Part of the results presented in this paper were obtained within projects funded by the German Air and Space Center DLR, Bonn and Dornier Satellitensysteme GmbH, Friedrichshafen. I want to express my thanks to Bernhard Kenter, Dr. Matthias Scheuber, Wolfgang Stiimer, Daniel Thiele and Helge Ziese, all TU-Dresden, Chair of Forest Biometry and Computer Sciences, and to Dr. Thomas Häussler and Markus Lautner, GAF, Munich, for helpful input and comments.