Using Unmanned Aerial Vehicles (UAV) to Quantify Spatial Gap Patterns in Forests

Gap distributions in forests reflect the spatial impact of man-made tree harvesting or naturally-induced patterns of tree death being caused by windthrow, inter-tree competition, disease or senescence. Gap sizes can vary from large (>100 m2) to small (<10 m2), and they may have contrasting spatial patterns, such as being aggregated or regularly distributed. However, very small gaps cannot easily be recorded with conventional aerial or satellite images, which calls for new and cost-effective methodologies of forest monitoring. Here, we used an unmanned aerial vehicle (UAV) and very high-resolution images to record the gaps in 10 temperate managed and unmanaged forests in two regions of Germany. All gaps were extracted for 1-ha study plots and subsequently analyzed with spatially-explicit statistics, such as the conventional pair correlation function (PCF), the polygon-based PCF and the mark correlation function. Gap-size frequency was dominated by small gaps of an area <5 m2, which were particularly frequent in unmanaged forests. We found that gap distances showed a variety of patterns. However, the polygon-based PCF was a better descriptor of patterns than the conventional PCF, because it showed randomness or aggregation for cases when the conventional PCF showed small-scale regularity; albeit, the latter was only a mathematical artifact. The mark correlation function revealed that gap areas were in half of the cases negatively correlated and in the other half independent. Negative size correlations may likely be the result of single-tree harvesting or of repeated gap formation, which both lead to nearby small gaps. Here, we emphasize the usefulness of UAV to record forest gaps of a very small size. These small gaps may originate from repeated gap-creating disturbances, and their spatial patterns should be monitored with spatially-explicit statistics at recurring intervals in order to further insights into forest dynamics.

[1]  Lael Parrott,et al.  Measures of structural complexity in digital images for monitoring the ecological signature of an old-growth forest ecosystem , 2008 .

[2]  Benoît St-Onge,et al.  Interactions of multiple disturbances in shaping boreal forest dynamics: a spatially explicit analysis using multi‐temporal lidar data and high‐resolution imagery , 2010 .

[3]  C. Ronnie Drever,et al.  Effects of a wide gradient of retained tree structure on understory light in coastal Douglas-fir forests , 2003 .

[4]  Hiroshi Tanaka,et al.  Fifteen years of canopy dynamics analyzed by aerial photographs in a temperate deciduous forest, Japan , 1997 .

[5]  F. Mark Danson,et al.  Testing the Application of Terrestrial Laser Scanning to Measure Forest Canopy Gap Fraction , 2013, Remote. Sens..

[6]  Akemi Itaya,et al.  Quantification of repeated gap formation events and their spatial patterns in three types of old-growth forests: Analysis of long-term canopy dynamics using aerial photographs and digital surface models , 2012 .

[7]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[8]  François A. Gougeon,et al.  Scale-dependent competition at the stand level assessed from crown areas , 2008 .

[9]  Jens Nieschulze,et al.  Implementing large-scale and long-term functional biodiversity research: The Biodiversity Exploratories , 2010 .

[10]  I. Király,et al.  The effect of light conditions on herbs, bryophytes and seedlings of temperate mixed forests in Őrség, Western Hungary , 2009, Plant Ecology.

[11]  Jacques Ranger,et al.  Effects of tree species on understory vegetation and environmental conditions in temperate forests , 2003 .

[12]  Arko Lucieer,et al.  Development of a UAV-LiDAR System with Application to Forest Inventory , 2012, Remote. Sens..

[13]  D. Stoyan,et al.  Statistical Analysis and Modelling of Spatial Point Patterns , 2008 .

[14]  D. Stoyan,et al.  Fractals, random shapes and point fields : methods of geometrical statistics , 1996 .

[15]  Thorsten Wiegand,et al.  Finding the Missing Link between Landscape Structure and Population Dynamics: A Spatially Explicit Perspective , 1999, The American Naturalist.

[16]  Zhe Jiang,et al.  Spatial Statistics , 2013 .

[17]  Elisabeth K. V. Kalko,et al.  High plant species richness indicates management-related disturbances rather than the conservation status of forests , 2013 .

[18]  Stephan Getzin,et al.  Assessing biodiversity in forests using very high‐resolution images and unmanned aerial vehicles , 2012 .

[19]  Jens Nieschulze,et al.  An assessment of forest canopy structure by LiDAR: Derivation and stability of canopy structure parameters across forest management types , 2012 .

[20]  Dietrich Stoyan,et al.  Marked Point Processes in Forest Statistics , 1992, Forest Science.

[21]  K. Moffett,et al.  Remote Sens , 2015 .

[22]  Tapio Linkosalo,et al.  Estimation of a spatial tree-influence model using iterative optimization , 1998 .

[23]  Yi Lin,et al.  Mini-UAV-Borne LIDAR for Fine-Scale Mapping , 2011, IEEE Geoscience and Remote Sensing Letters.

[24]  Joachim Saborowski,et al.  Adapting the pair-correlation function for analysing the spatial distribution of canopy gaps , 2009 .

[25]  R. Chazdon,et al.  Light gradient partitioning by tropical tree seedlings in the absence of canopy gaps , 2002, Oecologia.

[26]  Thorsten Wiegand,et al.  Extending point pattern analysis for objects of finite size and irregular shape , 2006 .

[27]  C. Hopkinson,et al.  Landscape-scale forest disturbance regimes in southern Peruvian Amazonia. , 2013, Ecological applications : a publication of the Ecological Society of America.

[28]  Daniel P. Bebber,et al.  Patterns of regeneration and ground vegetation associated with canopy gaps in a managed beechwood in southern England , 2006 .

[29]  Christian Messier,et al.  Use of a spatially explicit individual-tree model (SORTIE/BC) to explore the implications of patchiness in structurally complex forests , 2003 .

[30]  Heikki Kalle,et al.  Spatial pattern and species richness of boreonemoral forest understorey and its determinants—A comparison of differently managed forests , 2007 .

[31]  George Alan Blackburn,et al.  Quantifying the spatial properties of forest canopy gaps using LiDAR imagery and GIS , 2004 .

[32]  Benoît St-Onge,et al.  Spatially explicit characterization of boreal forest gap dynamics using multi-temporal lidar data , 2008 .

[33]  Karen Anderson,et al.  Lightweight unmanned aerial vehicles will revolutionize spatial ecology , 2013 .

[34]  Robert S. Nuske,et al.  Gap disturbance patterns in an old-growth sessile oak (Quercus petraea L.)–European beech (Fagus sylvatica L.) forest remnant in the Carpathian Mountains, Romania , 2013 .

[35]  S. Wich,et al.  Dawn of Drone Ecology: Low-Cost Autonomous Aerial Vehicles for Conservation , 2012 .

[36]  John Sessions,et al.  Eyes in the Sky: Remote Sensing Technology Development Using Small Unmanned Aircraft Systems , 2013 .

[37]  James R. Runkle,et al.  Treefalls Revisited: Gap Dynamics in the Southern Appalachians , 1987 .

[38]  Pu Mou,et al.  The effect of spatially variable overstory on the understory light environment of an open-canopied longleaf pine forest , 2002 .