| 2004 
  
ectonic 
Syrian 
sium on 
67. San 
et de la 
195, P. 
n, A.N. 
ab Rift 
story of 
5, 362- 
ment of 
0. 
i£ fuer 
indings, 
iblished 
aille du 
lu Fosse 
ibal As- 
grands 
ences de 
nd J.-J. 
Lebanon 
logische 
allad, Y. 
st Syria. 
  
  
TIME FOR CHANGE - QUANTIFYING LANDSLIDE EVOLUTION USING 
HISTORICAL AERIAL PHOTOGRAPHS AND MODERN PHOTOGRAMMETRIC 
METHODS 
J. Walstra * *, J.H. Chandler *, N. Dixon *, T.A. Dijkstra * 
“ Dept. of Civil and Building Engineering, Loughborough University, Leicestershire LE11 3TU, UK — 
J.Walstra@lboro.ac.uk 
Commission IV, WG IV/6 
KEY WORDS: geomorphology, DEM/DTM, extraction, modelling, visualization, quality, change detection 
ABSTRACT: 
A sequence of historical aerial photographs captures morphological change, which can only be unlocked by using appropriate 
photogrammetric methods. There are several challenges: initially it is necessary to trace and acquire suitable imagery in an 
appropriate format; typically there is a lack of precise photo-control available at the time of photography and similarly, it is rare to 
have access to the original camera calibration certificate. Furthermore, it is essential to assess the quality of the extracted 
morphological data. Once these challenges have been overcome, such morphological change data can be used to develop landslide 
evolutionary models, which can be related to localised climate change. This paper describes the use of modern digital 
photogrammetric methods to extract high-resolution digital elevation models (DEMs) and orthophotographs using historical 
photography. Initial results from a case study at Mam Tor (Derbyshire, UK) show the potential of the methods to quantify past 
landslide movements. Photographs from four different epochs (1953, 1971, 1973 and 1990), of varying scale and quality, were 
processed. Difficulties caused by limited ground control and limited camera calibration data were solved by differential GPS and 
self-calibrating bundle adjustment methods. The quality of the data was at an adequate level to extract ground movements that were 
consistent with observations by other authors. A more detailed analysis of the displacements in both vertical and horizontal 
directions is expected to increase the understanding of the mechanics of the landslide. This is an ongoing research project in which 
these dated spatial data will be used to relate landslide evolution to climate change. 
1. INTRODUCTION (Adams & Chandler, 2002). However, most of these methods 
are usually not capable of revealing quantitative data about past 
It is generally acknowledged that global climate is changing movements, which restricts the length of these records. 
(ICCP, 2001). Future climate change will have great impacts on Historical aerial photographs are a resource of spatial data that 
human society, but prediction of these impacts is subject to can be unlocked using photogrammetric techniques, and hence 
large uncertainties. This paper is framed in a research project are capable of being used to extend these records. Systematic 
dealing with the consequences of climate change for active aerial surveys started after the Second World War, providing 
landslides in the UK. Understanding of the climatic controls on potential data covering a period of more than 50 years. The aim 
mass movements requires analysis of past climate and landslide of this paper is to present the digital techniques used to extract 
activity patterns. However, only few high quality data records morphological change from a sequence of aerial photos and 
are available that are able to give a continuous picture (Dikau & discuss their potential value for validating climate controlled 
Schrott, 1999). landslide models. 
As climate is highly variable in the short-term, a significant 
change can only be recognized over longer periods. Moreover, 3: EXTRACTING QUANTITATIVE 
geomorphologic processes are subject to extrinsic and intrinsic MORPHOLOGICAL DATA USING DIGITAL 
thresholds, resulting in complex responses to environmental PHOTOGRAMMETRIC TECHNIQUES 
changes (Schumm, 1979; Brunsden & Thornes, 1979). These 
complex reactions mask the causality between climate change Historical aerial photographs can provide detailed information 
and geomorphologic response and makes modelling difficult. of past morphological change (e.g. Chandler & Cooper, 1989; 
Usually, while extensive climate data records are available Chandler & Brunsden, 1995). Despite its many advantages, use 
(systematic data collection started in the UK in the 1660s; of analytical photogrammetric methods remain expensive, com- 
Hulme et al., 2002), obtaining accurate historical data about plex and require a significant amount of experience (Chandler, 
landslide movements and landform change is more challenging. 2001; Brunsden, 1993). Recent developments in computer 
There is sophisticated equipment for monitoring present-day technology have had a significant impact upon photogrammetry. 
landslide movements: inclinometers, tiltmeters, extensometers, Advances in automated digital photogrammetry now allow 
time domain reflectometry (TDR) (Kane & Beck. 2000), global high-resolution quantitative data to be extracted automatically 
positioning system (GPS), synthetic aperture data (SAR) (Walker, 1996; Brunsden & Chandler, 1996). Advantages of 
(Buckley & Mills, 2000), satellite and aerial imagery automated digital photogrammetry include (Chandler, 1999): 
(Brunsden, 1993) and light detection and ranging (LIDAR) 
475