Open source GIS - GRASS users conference 2002

Trento, Italy, 11-13 September 2002

GRASS logo

Abstracts


Abstract n.1


Corresponding author
SAIDI Ahmed, Centre National des Techniques Spatiales CNTS., 01, Avenue de la Palestine BP 13 31200 Arzew ALGERIA, ALGERIA
Other authors
DAOUD BRIKSI Hichem, Centre National des Techniques Spatiales
MISSOUMI Abdelkader, Centre National des Techniques Spatiales



The use of the GIS into the Forest Fire prediction. The Simulation Model


The development of control and management strategies for forest fire depends crucially on a deep knowledge of the forest fire phenomenon. One of the best ways for understanding forest fire, is to have a tool able to inform us before the beginning of the fire according to given environmental and climatic conditions. In this context, a fire modelling/simulation tool remains efficient for the prediction and management of forest fires. In particular, such a model allows a determination of zones susceptible to being ravaged by the fire in a specific time interval, with some degree of confidence. The objective of this paper is to present a study of the mechanisms by which a fire progresses. This study includes an elaboration of a development of a tool able to represent suitably the parameters of forest fire, its propagation and its behaviour in a given region. It becomes clear that in order to adequately characterise these mechanisms, with spatially distributed propagation parameters, the use of a GIS is unavoidable. GIS in combination with simulation techniques provides a sound basis for the appreciation and understanding of problematical forest fires. This resides in the power of GIS to represent all phenomenon presenting a geographical character. Such fire simulation models have two main attractions to operators in charge of management of forest fire (fire men, forests services, local collectives, etc.). First, they allow the consideration and verification of various forest control and management actions without having to suffer the disaster of a real fire. Hence they allow operators to define a long term coherent and homogeneous policy of prevention of forest fires. Such models also makes it conspicuously clear that the co-ordination of the intervention teams in the early stages of a fire is of paramount importance in developing an effective strategy for the control of the progression of forest fire.

Abstract n.2


Corresponding author
Hofierka Jaroslav, GeoModel s.r.o., M. Marecka 3, 841 07 Bratislava, Slovak Republic
Other authors
Suri Marcel, European Commission Joint Research Centre




The solar radiation model for Open source GIS: implementation and applications


Accurate solar radiation data are desired for various applications (environmental science, ecology, pohotovoltaics, climatology, etc.). Besides the earth's latitude, terrain is the major factor modifying the spatial and time distribution of solar radiation creating a strong spatial variability. Spatial solar radiation models integrated within a geographic information system provide a cost-efficient means for solar radiation modelling on large areas. The paper presents a new, substantially improved solar radiation model r.sun implemented in GRASS GIS. The model estimates beam, diffuse and reflected components of the global irradiance/irradiation on a horizontal and inclined surfaces. The r.sun works in two modes. For the instant time it calculates a solar incident angle and solar irradiance. In the second mode the daily sums of solar irradiation can be computed by the integration of the irradiance values calculated within a defined time step between the sunrise and sunset. The model architecture enables to compute the clear-sky values, i.e. not considering the cloudiness. The overcast conditions can be considered using the coefficients describing a ratio between the clear-sky and overcast values of the beam and diffuse radiation components. The two model applications are presented. The first one summarizes the estimation of global irradiation available to the photovoltaic systems with panels in a horizontal and inclined positions for the Central and Eastern European countries. The second application presents a calculation of the global irradiation for ecological studies in Slovakia.

Abstract n.3


Corresponding author
Preda Mauro , Università Cattolica del Sacro Cuore di Milano, L.go Gemelli 1, Italy
Other authors





Economia spaziale : modelli applicati e simulati con GRASS


Con il presente contributo si vuole dimostrare come alcuni modelli teorici elaborati nell'ambito della geografia economica (o economia spaziale), con elevato grado di astrattezza possano ricevere, in generale, nuovo impulso se introdotti in ambito GIS. Il discorso di amplia a comprendere le inevitabili sovrapposizioni disciplinari e le zone di raccordo tra due le diverse discipline, appunto. GRASS, in questo contesto non è solo un prodotto GIS ma anche, e soprattutto, come "linguaggio". GRASS come "linguaggio" aperto ha la capacità di esprimere ogni concetto, anche i più astratti.

Abstract n.4


Corresponding author
HESS Sigrid, University of Mainz, Institute of Geography, 55099 Mainz, Germany
Other authors





GRASS ON THE WEB


As HTTP is now the leading data transfer protocol, access to geocoded data through a standard browser has to be considered. The question arises to which extent GRASS is able to support Web interface. This kind of application relies on server-side CGI (common gateway interface) grids. Hopefull availabilty of web oriented interpreters facilitates devolpement of such applications. In our presentation we will propose a design based on a combination of PHP, GRASS scripts and other Linux standard tools. Should performance be a serious issue, actual processing could be deferred to a queue and results subsequently sent by mail. On some sample GIS request, we will show, that a simple Web interface for GRASS is easy to programme. The interest of a Java based viewer will be discussed.

Abstract n.5


Corresponding author
Acma Bulent, Anadolu University, Anadolu University, Department of Economics, Turkey
Other authors





Strategies for a Regional Innovative Knowledge Society:the Southeastern Anatolia


Strategies for a Regional Innovative Knowledge Society: the Southeastern Anatolia Region and the Southeastern Anatolia Project(GAP) of Turkiye As a Case Study Innovation and knowledge are fundamental to the economic development, growth, and future competitiveness of our Regions. Central to strategies of enabling this must be partnership and networking, both local and global - public and private, with action leading to a solid foundation for innovation. As competition is the ultimate driver of innovation, this calls for more entrepreneurial dynamism and greater mobility of knowledge based on an effective use of Information and Communication Technology (ICT) to enable increased interconnectivity between knowledge workers through virtual networking. In a Knowledge Society it is humans and human capital, their vision and abilities, which are most important to the positive development of our regions, rather than the predominant physical resources and issues of the past. Empowering our people to apply their intelligence, experience and imagination to leverage data and information into a Knowledge Society, requires dynamic regional leadership, a "can do" attitude and a holistic approach to regional development. A Regional Innovative Knowledge Society requires alternate nodes of development in various local contexts and new mechanisms such as awareness creation, access to telecommunications, education/training, finance, academic/industrial interactions and sectoral networks. This process is recognised at European level and is already happening at local level, as illustrated in Turkiye's Southeastern Anatolia Region and elsewhere. In the first section of this study, economic innovations depending on knowledge will be presented. In the second section the efforts of Europe Union about innovations will be investigated. In the third section strategies for regional innovation knowledge society will be examined. In the last section the strategies for converting the Southeastern Anatolia Region into an innovative knowledge society will be analyzed. JEL classification: O 31-32, R 11, R 58 Keywords: Innovation and Invention: Processes and Incentives, Management of Technological Innovation and R&D, The Southeastern Anatolia Region and The Southeastern Anatolia Project(GAP), Analysis of Growth, Development, and Changes, Regional Development Policy

Abstract n.6


Corresponding author
García Galiano Sandra, Universidad Politécnica de Cartagena. EUIT Civil. , Paseo Alfonso XIII, 52, Spain
Other authors





A GIS GRASS-embedded decision support framework for flood forecasting


In this paper a real-time operative decision support computer system named Shyska is presented. This applies the potentiality of GIS to the processing of information to different spatiotemporal scales. Shyska environment, developed with GIS GRASS-embedded functions, (a) Combines efficiently information from different sources presenting considerable time-space variability, both supplied by SAIH systems and from remote sensing (rainfall fields products of radar-satellite technology). (b) Integrates spatially distributed and hybrid hydrologic models, topographically based, oriented to real-time simulation and forecasting. Automatically extracts from the DEM the relevant parameters for formulating the hydrologic models used. Its final aim is to assist Hydrological Information Automatic Systems (SAIH systems in Spain), facilitating information management and use in real-time when alert and flash flood situations so typical of Mediterranean environments occur. Shyska has been applied to semiarid basins in the southeast of Spain, presenting satisfactory results.

Abstract n.7


Corresponding author
Bergès Jean Claude, PRODIG Université Paris 1, 191 rue St Jacques - 75005 Paris, France
Other authors





Support of WMO binary formats (BUFR and GRIB)


World Meteorological Organization monitors the largest network of environmental information. Although WMO data are freely available for scientific or educationnal use, few softwares can handle all the associated formats. Yet a smooth integration of meteorological information into a GIS would be a valuable tool for climate impact assessment. GRASS support of two WMO binary format (GRIB and BUFR) is presented. GRIB is devoted to transfer of gridded data, forecast model or satellite images whereas BUFR is object oriented. The decoders which have been developed are described with a focus on extension and update issues. But decoding format data is not enough for an efficient processing, meteorological objects need to be properly supported. Unit conversions and specific graphic representations have to be integrated. But more generally some thermodynamic computations should be possible. Relative importance of these developments is discussed.

Abstract n.8


Corresponding author
Ahmed Iftikhar, United Nations Volunteer (UNV) Program/Government of Montserrat (West Indies), Physical Planning Unit, Ministry of Agriculture, Land, Housing and Env. Brades, Montserrat (West Indies)
Other authors





GIS Application for Land Planning and Management of Montserrat (West Indies)


Montserrat, a small Caribbean Island, has severely suffered in recent past by natural disasters. Particularly major eruption of June 1997 completely destructed nearly two-third areas of the Island including airport, seaport and the capital, Plymouth. Devastation at such large scale had inauspicious impact on economic, social, environment and institutional structures, which resulted an immediate migration of nearly 62% population to U.K, USA, and nearby islands. Government, in collaboration with other international organizations like DFID and UNDP, is in process of rebuilding the Montserrat. Effective planning of the available limited land resources has become inevitable and a major concern to the government for a socially and economically developed Montserrat. Physical Planning Unit of Montserrat has been developing an intelligent Land Information System (LIS) using geospatial approach. Computerized ‘National Cadastre’ and ‘Land Registry’ systems, which will support landuse planning, infrastructural development and agricultural modernization, are being developed at first instant. Both, numerical as well as spatial (thematic layers) data sets are being developed using different GIS softwares. Land parcel boundaries and topographic maps would be revised using aerial survey data. Fresh GPS surveys will help in boundary verification of the parcels and other land features. This GIS/LIS system will provide bases for a national data warehouse to support infrastructural development activities of Montserrat.

Abstract n.9


Corresponding author
ciriaco saul, Miramare Natural Marine Reserve, viale Miramare 349 34014 Trieste, ITALY
Other authors
Bonacito Clizia , Miramare Natural Marine Reserve
Costantini Marco , University of Trieste Department of Biology
Spoto Maurizio, viale Miramare 349 34014 Trieste


Sea-bed classification and sea-bottom mapping with GRASS in the R.N.M.M.


Bonacito Clizia* , Ciriaco Saul*, Costantini Marco** & Spoto Maurizio* *WWF Miramare Natural Marine Reserve info@riservamarinamiramare.it ** University of Trieste Department of Biology The W.W.F-Miramare Marine Reserve is a protected area of 120 hectares in the Gulf of Trieste (Northern Adriatic sea) The bottom is mainly soft and characterised by a sequence from rocky to mud gradually mixed. Recent research underlined that the presence of Lithognathus mormyrus, an important fish specie under the ecological and economic point of view, seems to be strictly linked to some particular substrata. GRASS, a DGPS (Omnistar Differential Correction), and the scientific echosounder Biosonics DT6000 Split Beam were used during the project to collect acoustic echoes from the bottom, in order to have an idea of the distribution of the substrata and to correlate, in the future, fish aggregation and bottom types. The data were analysed by the bottom analyser software VBT, developed by Biosonics Inch.

Abstract n.10


Corresponding author
Jha Raghunath , Lecturer, Institute of Engineering, Pulchowk, GPO-8975, EPC-1970, Kathmandu, Nepal
Other authors





POTENTAIL EROSION MAP FOR BAGMATI BASIN USING GRASS


Nepal varies from latitude 26022’ to 30027’ and longitude 80004’ to 88012’ and it occupies large part of central Himalayas and its foot hills. Total area is 147,181 sq. km and about 80% of the land is mountainous with rugged topography. Elevation varies from 70 meter to 8848 meter above mean sea level and increase from south to north. Nepal is divided into five physiographic regions from south to north namely the high Himal, high mountains, middle mountains, Siwalik and Terai. The most vulnerable area is middle mountains and Siwalik in both terms of natural and man made induced erosion causing sever damage of lives and properties. Bagmati basin is a basin starts from Middle Mountain, passes though Siwalik, and Terai. The Bagmati River originates from the Shivapuri Mountain which is situated to the north of the Kathmandu valley. The river flows in a south westerly direction across the valley and passing through a gorge called Chobhar gorge. From here the river continues flowing towards south east and then towards south having joined by several tributaries and ultimately passing trough the Terai of Nepal and flowing into the river Ganges in India. The total catchment area at the end of the Siwalik range is nearly 4000 sqkm. The topography comprises narrow valleys and steep slopes including fragile Siwaliks. The monsoon rain makes the whole watershed area prone to soil erosion, gulley formation and landslides in the hills and flooding in the valleys and the flood plain. In this research authors have used GRASS software to prepare a potential erosion map in the Bagmati basin using Revised Universal Soil Loss Equation (RUSLE). The LS factor of this equation is evaluated by Watershed function of the GRASS. The rainfall, soil factor and crop factors are taken from different research. Finally using the mapcalc function of the GRASS, erosion from each cell is calculated and re-classified into five groups (very low, low, moderate, high and very high). This map will be helpful to the planners to emphasizes their effort to minimize the soil erosion.

Abstract n.11


Corresponding author
Cebecauer Tomas, GeoModel s.r.o., M. Marecka 3, 841 07 Bratislava, Slovak Republic
Other authors
Hofierka Jaroslav, GeoModel s.r.o.
Suri Marcel, GeoModel s.r.o.



Processing digital elevation models by Regularized Spline with Tension


Whole title Processing digital elevation models by Regularized Spline with Tension method: tuning interpolation parameters for different datasets Regularized Spline with Tension (RST) belongs to the most accurate available interpolation methods. This method is implemented in GRASS as s.surf.rst command. Using a set of interpolation parameters, users can effectivelly control the quality of interpolation. However, to obtain good interpolation results, the user has to figure out what combination of controlling parameters gives the most appropriate results. The paper presents an overview of common problems selecting s.surf.rst parameters with different types of datasets. It is shown that some specific types of elevation data require data pre-processing which may substantially improve interpolation results. This pre-processing contains preparation of supplemental elevation points and selecting data subsets assigned with different variable smoothing parameter ensuring the best interpolation accuracy in critical surface areas (e.g. faults, valleys, ridges, etc.). Application examples include elevation data taken from contours, photogrammetric measurements and supplemental points at various scales and sizes of datasets. All types of interpolation errors and artefacts are well documented. The segmented processing of huge datasets (containing over 10 millions of elevation points) is also presented using the example of the Slovak national DEM.

Abstract n.12


Corresponding author
Paudits Peter, Geological Survey of Slovak Republic, Mlynska Dolina 1, Bratislava, 817 04, Slovak Republic
Other authors
Bednarik Martin, Faculty of Natural Sciences, Comenius University, Mlynska Dolina , Bratislava




USING GRASS IN EVALUATION OF LANDSLIDE SUSCEPTIBILITY IN HANDLOVSKÁ KOTLINA BASIN


The contribution shows the possibilities of GRASS in evaluation of the model territory - Handlovská kotlina basin (Central Slovakia), which is seriously affected by the various types of slope deformations (landslides, block rifts and fields and rock falls). The following parameters and factors of the environment were considered as a most important in a process of slope movement: lithology, the dip of slopes and average rainfall during the last years. The simplified information about these parameters were an input data for GRASS environment (in raster form as parametric maps). In the first step of operation, input parametric maps were reclassified into several categories using r.reclass (following the r.resample) or r.mapcalc commands. Then each category was statistically evaluated using the r.stats routines – comparison of certain category with unstable area in landslide inventory map. Based on r.stats results, new value was joined to each existing category in input parametric maps. This new value represented degree of potential hazard. The final evaluation was realised by two different ways. The first one (bivariate statistical analysis) is based on the simple addition (using r.mapcalc) of values in each reclassified parametric map. The results have been extrapolated to the surrounding area. The similar result was obtained by the second method (multivariate statistical analysis). In this method the all input maps were compared with each other using multivariate capabilities of r.stats command. Attained results from this comprehensive comparison were re-sorted according the covered area in landslide inventory map and whole evaluated area was reclassified into various degree of slope movement endangering. The results of both methods represent the prognosis maps of the slope deformation hazard assessment.

Abstract n.13


Corresponding author
Tipdecho Taravudh, High Performance Computing Research and Development Division , National Electronics and Computer Technology Center, Bangkok, Thailand
Other authors





Automatic Image Registration on 3D-to-2D Images


By ever increasing the number of sensors, the requirements of integrations among different sensors have been arisen, once is due to more gained benefits considering on differences of spectral, time, and so on. With reference to primitive methods of images registration, a 2D-to-2D image is commonly mentioned for the process of registration regarding simply on affine transformation, while 3D-to-2D image is namely used as orthorectification. Nevertheless, the specific method for 3D-to-2D image registration is still under development focusing on avoiding complexities of finding out sensor parameters particularly when integrated sensors as SPOT and Airborne laser data are considered, meanwhile automatic processing has also been requested. Nowadays, even the single and multiple automatic feature extraction are efficient tools, which perform good consequence. Some blunders are still occurrence such as the difficulties finding correctly matched features, where the heterogeneous area is targeted, and delaying of processing time. This paper introduces, therefore, improved model for automatic image registration with respect to 3D-to-2D image based on edges extraction and corners matching. Polynomial transformation was mentioned as major tools for transforming all tie points from slave to master image. Using selected tie points with gradually increasing its quantity, finally the RMS residual was approximately two pixels regarding systematic error and an efficiency index was about 0.99 at certain position of selected tie points. A method of processing based least square principal served for calculation of transformation. The Sobel and A’trous algorithms were both filters for edges extraction. There were subsequently closing algorithm and threshold processing to be applied for images. And then, corners of buildings were defined by using template-window searching that was relied on sixteen patterns. Finally, it showed successful with accepted residual particularly on the roofs of buildings.

Abstract n.14


Corresponding author
Socharoentum Monsak, National Electronics and Computer Technology Center (NECTEC), 25/30 Wuttakat Rd. Thaladplu Thonburi Bangkok 10600, Thailand
Other authors





Activities of GRASS in Thailand


GRASS has been initially promoted since 1998 by NECTEC (National Electronic and Computer Technology Center) in Thailand. The first activity is setting up a GRASS mirror site and follow by joining in GRASS development team. Even through these activities are initiated, the user friendly problem in Linux and GRASS were a big obstruction for users's starting. NECTEC has been supporting GRASS community in Thailand by launching a few projects like training, distributing cd-rom, maintaining mirror site and advertising. These projects increase GRASS users in Thailand to 30-40 users separating in many academics and organizations. However, GRASS users are still small. Presently, NECTEC has a few continuous projects to support GRASS such as GRASS training, developing thai label on map and installation programme in GUI style ,which we hope to give more advantage to thai GRASS users.

Abstract n.15


Corresponding author
Biagi Ludovico, Politecnico di Milano - Campus Como, Via Valleggio 11 - Como, Italia
Other authors
Brovelli Maria Antonia, Politecnico di Milano - Campus Como
Negretti Marco, Politecnico di Milano - Campus Como



Environmental thematic map and easy probabilistic estimation of a threshold exceed


Environmental observations are usually sampled in irregularly spaced points even though the information we need must be, in most cases, continuous and representative of continuous phenomena. To create the thematic map, the stochastic approach is often the best interpolating method to apply. One of the advantages of the application of this theory consists in making available not only a ?prediction? map (thematic map), but also an ?error prediction? map (accuracy map); starting with these two information layers a procedure can be implemented to verify if in some areas the predicted data exceed a threshold value at certain degree of probability. The methodology adopted, even though simplified from the theoretical point of view, gives in detecting risk areas, information that is more accurate than that deriving from usual deterministic approaches. The procedure has been developed by creating GRASS commands calling upon and integrating the statistic GSTAT library and commands which allow the computing of probabilistic risk areas.

Abstract n.16


Corresponding author
Brovelli Maria Antonia, Politecnico di Milano - Campus Como, Via Valleggio 11 - COMO, Italia
Other authors
Cannata Massimiliano, Politecnico di Milano - Campus Como
Longoni Ulisse Matteo, Politecnico di Milano - Campus Como



Managing and processing LIDAR data within GRASS


LIDAR (Light Detection and Ranging) is one of the most recent technologies in surveying and mapping. The LIDAR is based on the combination of three different data collection tools: a scanning laser mounted on an aircraft, a Global Positioning System (GPS) used in phase differential kinematic modality to provide the sensor position and an Inertial Measuring Unit (IMU) to provide the orientation (pitch, roll and yaw angular value). The laser sends out to the ground an infrared signal, which is reflected back to the sensor. The time employed by the signal, given the aircraft position and orientation, allows us to compute the earth point elevation. In standard conditions, taking into account the flight (speed 200- 250 km/hour, altitudes 500-2000m) and sensor characteristics (scan angle ?10-20 degrees, emission rate 2000 ? 50000 pulses per second), the earth elevations are collected within a density of one point every 0.5-3 meters. The technology allows us therefore to obtain very accurate (5-20 cm) Digital Surface Models (DSM). For many applications the Digital Terrain Model (DTM) is needed: we have to automatically detect and discard from the previous DSM all the features (buildings, trees, ?) present on the terrain. The elaborate procedure has been implemented within GRASS.

Abstract n.17


Corresponding author
Brovelli Maria Antonia, Politecnico di Milano - CAmpus Como, Via Valleggio 11 - Como, Italia
Other authors
Magni Diego , Politecnico di Milano - Campus Como




An archaeological web GIS application based on Mapserver and PostGIS


The availability of information of the cultural heritage of a territory can lead to an improvement in its exploitation and preservation and to a greater fruition by the community. To this aim a web-GIS of a natural and archaeological park in the Como area (in the northern part of Italy) has been designed and implemented. The main topic related to the project has been to create a tool easy to access and suitable at least for two kinds of users: domain experts, such as archaeologists, Artistic Heritage Offices and museums, and laypersons interested in information on the cultural assets of the area for educational and/or tourist purposes. Territorial and environmental information (physical, environmental, political-administrative, economical and utilities, road maps) has been integrated in order to provide, as much as possible, a complete and exhaustive frame where the archaeological entities are located. Views, queries and simple spatial analyses have been provided, taking into account the characteristics of the target users. The whole system has been implemented in html language exploiting Mapserver for the GIS functionality and the PostGIS Database Management System to the data storage.

Abstract n.18


Corresponding author
Shevlakov Alexander, Motivation free software consulting company, technical director, 142092 Troitsk, Moscow, Pushkovykh St, 9, r.106, Russia
Other authors





GRASS5.0 vector layer management and operation enhancement by using RDBMS Postgre


In addition to existing GRASS5.0 vector modules, the author supports a relatively non-expensive method of trans-application data exchange between GRASS and RDBMS PostgreSQL which internally supports vector formats as polygon, path, point, circle and possibly as many other geometry types as one may need. By using the new GRASS data interchange module v.to.pg, vector data can be loaded directly to a DBMS table as lists of vertices and points. The further operations include spatial queries with both built-in Postgres operators and new functions provided as dynamically loaded library written by author and other developers. The DBMS operations results can then be processed in GRASS after re-exporting the data. The scope of possible applications of this method includes time-dependent vector modeling such as buffering, wave-front models with complex 2-d topology computations.

Abstract n.19


Corresponding author
MASUMOTO Shinji, Department of Geosciences, Graduate School of Science, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558, Japan, JAPAN
Other authors
RAGHAVAN Venkatesh, Media Center of Osaka City University
NEMOTO Tatsuya, Department of Geosciences, Graduate School of Science, Osaka City University
SHIONO Kiyoji, Department of Geosciences, Graduate School of Science, Osaka City University


Construction and Visualization of Three Dimensional Geologic Model Using GRASS GIS


Recently, the need of the geologic information has been rising in the field such as environmental geology, disaster mitigation, and urban geological applications. For these fields, it is effective to provide geologic information as a 3D model that can be generated and visualized in general purpose GIS software. The present work aims at implementing methodology and algorithms for 3D modeling and visualization of geologic model using the Open Source GRASS GIS environment. 3D geologic model is constructed from the boundary surfaces of geologic units and the logical model of geologic structure. The algorithms for construction and visualization of model are based on the geologic function g. The geologic function g assigns a unique geologic unit to every point in the objective 3D space. The boundary surface that divides the objective space into two subspaces, was estimated using data from field survey. The logical model showing the hierarchical relationship between these boundaries surfaces and geologic units can be automatically generated based on the stratigraphic sequence and knowledge of geologic structures. Based on these algorithms, 3D geologic model can be constructed virtually on GRASS. Applying this model, various geologic surface and section models can be visualized in GRASS environment. Further, gNvizh was used for dynamic visualization of geologic cross-sections and generation of animated image sequences.

Abstract n.20


Corresponding author
Greenwood Richard, Greenwood Mapping, Inc., Box 461, Wilson, WY 83014, U.S.A.
Other authors





Using Mapserver to Integrate Local Government Spatial Data


Local government has the responsibility for various land management and regulation functions including land ownership registration, tax assessment, and zoning. Many of the information technologies that have come into use to support these tasks have led to departmental isolation. In many cases, each department has its own autonomous dataset which is not integrated with the datasets of parallel departments, despite overlapping spatial extents. Both GIS, and web-based technologies are well suited to aggregating disparate datasets. This is a discussion and demonstration of a Mapserver application developed for Teton County, Wyoming; a resort community in the Yellowstone National Park region of the United States. The application integrates spatial data from several county and municipal governmental departments that were previously un-related. In addition to vector and attribute data, a significant aspect of the system is the integration of raster imagery from various sources. Both color and grayscale digital ortho-imagery, in different projections, are seamlessly incorporated into the Mapserver application. Lessons learned in the use of ECW compressed imagery and Proj.4 image reprojection with Mapserver will also be discussed.

Abstract n.21


Corresponding author
Fryer John G., Department of Civil, Surveying and Environmental Engineering, University of Newcastle, Callaghan, New South Wales 2308, Australia
Other authors
Scarmana Gabriel, University of Newcastle, Callaghan, New South Wales 2308




IMPROVING D.T.M. GENERATION BY ENHANCING THE RESOLUTION OF REMOTELY SENSED DIGITAL


IMPROVING D.T.M. GENERATION BY ENHANCING THE RESOLUTION OF REMOTELY SENSED DIGITAL IMAGES ABSTRACT Close range and satellite-based photogrammetry allows the determination of the size and shape of objects from measurements made on remotely sensed images. The advent of digital technology has produced opportunities for new and diverse applications of this discipline to be undertaken which were not feasible with traditional photogrammetric techniques. Automated generation of D. T. Ms. (Digital Terrain Models) is one of such applications. Digital photogrammetry is sometimes limited by the cost of acquiring digital imagery at appropriate resolutions. Low-resolution imagery is relatively inexpensive to acquire, but may not provide the accuracy required, especially in subsequent processing to derive a D. T. M. Hence, the purpose of image enhancement is to improve the quality of a set of a lower resolution images so that the enhanced composite becomes more suitable for an application than the original images. This paper presents an algorithm which is device independent and can increase the spatial resolution of an undersampled, and thus aliased, image sequence. It is assumed that these low-resolution images are degraded, noisy, and displaced by sub-pixels shifts or translations with respect to a reference frame. Global translations between random frames in the sequence provide information that can be used to reduce some of the aliasing present in the frames. Translations vectors for each frame are estimated via area-based image matching algorithms. The proposed method is of moderate computational complexity and has proved to be robust under noisy circumstances. This algorithm is illustrated with applications which show its implementation using Fourier theory to model the greyscale surface of the enhanced image. In order to prove the algorithm's effectiveness as a photogrammetric tool, it was used in a series of three-dimensional tests using images of objects of known geometry. Stereoscopy sets of left and right images were taken of these objects, D.T.M. were created using both the original images and images enhanced by the algorithm, and these indicated its potential for use for photogrammetric surface modelling.

Abstract n.22


Corresponding author
Zatelli Paolo, Dept. of Civil and Enviromental Engeneering, University of Trento, via Mesiano, 77, Italy
Other authors
Antonello Andrea, Dept. of Civil and Enviromental Engeneering, University of Trento




New GRASS modules for Multiresolution Analysis with wavelets


The recognition and the study of several spatial phenomena strongly depend on the resolution at which the analysis is carried out. It is therefore very important the availability of an efficient (from both a mathematical and a computational point of view) way to represent the same phenomena at different resolutions. Wavelets, families of base functions generated by dilation and translation of a mother function, provide a good balance of localisation in space and localisation in the frequency domain, unlike the Fourier transform (where the localisation in space is lost) and spline (where the localisation in frequency is lost). Wavelets can be used in a Multiresolution Analysis (MRA), where the signal and the difference from the previous resolution level are orthogonal, to provide an efficient tool for investigating resolution dependent phenomena. Four new different GRASS modules have been created for the wavelets analysis and synthesis with orthogonal and bio-orthogonal wavelets bases. A bidimensional signal is processed into four sub-images, one showing the original image at coarser resolution (de-noised), while the other three represent the difference (noise) between the actual level of decomposition and the previous level (in x, y and diagonal direction). A first application to a geomorphologic problem of shape recognition, exploiting the advantages of the multiresolution technique, is presented.

Abstract n.23


Corresponding author
Zatelli Paolo, Dept. of Civil and Enviromental Engeneering, University of Trento, via Mesiano, 77, Italy
Other authors
Ciolli Marco, Dept. of Civil and Enviromental Engeneering, University of Trento
Rea Roberto, Dept. of Civil and Enviromental Engeneering, University of Trento
Zardi Dino, Dept. of Civil and Enviromental Engeneering, University of Trento


Modelling of evaporation processes over tilted slopes by means of 3D GRASS raster


Models of atmospheric processes occurring close to the ground and GIS display a point of contact since most of the input variables of lower atmosphere phenomena models are distributed on the land surface. On the other hand, the intrinsic 3D character of such phenomena has so far prevented the development of atmospheric models within a GIS environment, since no tools for the management and elaboration of 3D data were available. The recent introduction of the 3D raster management modules in GRASS has made possible to implement a model for the evaluation of wind profiles and humidity distribution along a sloping valley. This model exploits the power of the 3D GRASS map algebra (r3.mapcalc) to implement a simplified model generalizing Prandtl (1942) model for sloping valleys winds to take into account humidity and evaporation processes on the soil. Two applications to both an ideal valley and a real one, with validation of the model results, are presented.

Abstract n.24


Corresponding author
Zatelli Paolo, Dept. of Civil and Enviromental Engeneering, University of Trento, via Mesiano, 77, Italy
Other authors
Ciolli Marco, Dept. of Civil and Enviromental Engeneering, University of Trento
Vitti Alfonso, Dept. of Civil and Enviromental Engeneering, University of Trento
Zardi Dino, Dept. of Civil and Enviromental Engeneering, University of Trento


2D/3D GRASS MODULES USE AND DEVELOPMENT FOR ATMOSPHERIC MODELING


3D GRASS modules for environmental data management and analysis allow the implementation of algorithms for treatment of variables related to physical phenomena closely depending on interaction with ground and on landscape characteristics. It is possible to carry out an integrated and complementary use of 2D and 3D modules in GRASS. The development of slope winds follows ground heating and cooling which depends on the energy balance at the ground level. A simplified mathematical model to simulate slope winds developing along sloping valley side-walls has been developed by Prandtl in 1942. The model provides an analytical solution depending on the coordinate "n" of a point in the 3D volume measured along the normal to the terrain surface. A new module, r3.isosurf, has been developed to evaluate this coordinate in the entire 3D domain. This module can optionally generate a 3D raster where each cell is assigned the value of the cell on the surface belonging to the same normal vector. In this way it is possible to transfer 2D into a 3D algebra. The energy balance depends on astronomical as well as geometric and topographic factors, such as sun azimuth angle, which can be obtained using GRASS. In this paper a model for the simulation of slope winds in GRASS is presented. The teorical background, extending Prandtl (1942) solution, is outlined as well as the development and the use of mixed 2D/3D modules in GRASS. The application to a real case of an area near Trento provides results which are in accordance with physical reasoning, data obtained from both ground based and airborne measurements.

Abstract n.25


Corresponding author
Bezzi Marco, Università degli Studi di Trento, Via Mesiano 77, Italy
Other authors
Comunello Giovanni, Professional
Cherubini Paolo, WSL - Birmensdorf
Ciolli Marco, Università degli Studi di Trento
Cantiani Maria Giulia, Università degli Studi di Trento

GIS AND DENDROCHRONOLOGICAL TECHNIQUES FOR AVALANCHE HAZARD MAPPING WITH GRASS


Snow avalanches are one of main natural hazard for human activities and infrastructures in the Alpine regions. In the last years the opening up of the mountain for tourism has increased the risk potential of this natural hazard. Information about the frequency and the extensions of these phenomena is a prerequisite for environmental management in the mountain regions. By integrating dendrochronological and GIS techniques it is possible to reconstruct the frequency and the extensions of these past events. The aim of this work is to reconstruct the past avalanche events which occurred in two avalanche paths in the Pejo Valley (Northern Italy). Dendrochronology has been used to date the past snow avalanches for the last 100 years; GRASS-GIS has been used to obtain an avalanche risk map based on morphological and vegetation features and to create maps of different avalanche paths through the years. All the results were compared with historical data available from written records, for the period 1970-2000, chosen like period of calibration. The good correlation shows the efficacy of this methodology for avalanches forecast and for the creation of new possible location avalanche map (CLPV).

Abstract n.26


Corresponding author
Ciolli Marco, Facoltà di Ingegneria Università di Trento, Via Mesiano 77, Italy
Other authors
Deola Claudio, Facoltà di Ingegneria Università di Trento
Tamanini Tomas, Facoltà di Ingegneria Università di Trento



LAND USE CHANGE MAPS IN A DEVELOPING COUNTRY. APPLICATION OF GRASS GIS IN CAIA DIS


The lack of cartography is one of the main problems for the organizations which are working in developing countries. An application of GRASS GIS in the Caia District (Mozambique) is presented. Combining GRASS GIS satellite image analysis capability with field work and GPS point position acquisition, it has been possible to produce land use maps, which can be useful to give a relevant first picture of the landscape and to understand some critical problems of the area. Two series of landsat satellite images taken in 1999 and in 2001 have been analysed. Some of the most interesting landscape changes due to Zambesi river flooding of the year 1999 have been underlined in a dramatic way by the satellite image analysis carried out with GRASS. By means of GRASS the impressive river path change and the expansion of Caia city due to the exodus have been mapped and quantified. However, satellite image resolution (30 meters) does not allow a detailed observation of the different crops types. In this area, agriculture is very disperse and different kinds of crops are grown in small and very fractioned fields. A future development of the research will be addressed to carry out an analysis of Ikonos images (1 meter resolution) to produce more detailed land use maps. This first approach has underlined that an Open source GIS like GRASS can be very useful in developing countries for practical application and the whole methodology can be easily and cheaply transferred to the local organizations.

Abstract n.27


Corresponding author
Cemin Andrea, Facoltà di Ingegneria Università di Trento, Via Mesiano 77, Italy
Other authors
Ciolli Marco, Facoltà di Ingegneria Università di Trento
Nave Domenico, Facoltà di Ingegneria Università di Trento



MODELLING EMISSION AND DISPERSION OF ROAD TRAFFIC POLLUTANT FOR THE TOWN OF TRENTO


For every town, not only in developed but also in developing countries, air pollution problems are of great concern. There are many pollution sources, but road traffic, in particular, may lead to high air concentration levels of substances dangerous for human health like nitrogen oxides and fine particulates. Where traffic data are available, models which can evaluate pollution degree starting from emission data are particularly interesting. Aim of this work is to develop pollution models in GRASS starting from emission and traffic data. To compare the results of GIS elaborations, data measured by some pollution control stations and a bio-monitoring study on lichens are available. Both 2d and 3d GRASS capabilities have been used to implement the pollution model and to represent spatial data distribution. Urban situations are characterized by a complex and non-homogeneous flow geometry, so two models have been tested: Gaussian plume model and Canyon model. The results have been checked against both measured data and bio-monitoring results. GRASS has proved to be effective both as a calculation tool and as a mapper of 2d and 3d results. Further development of the work will consist in the implementation of more complex emission to concentration models.

Abstract n.28


Corresponding author
Bivand Roger, Norwegian School of Economics and Business Administration, Breiviksveien 40, N-5045 Bergen, Norway
Other authors





GRASS/R interface workshop and demonstration


This proposed three hour practical session needs access to a teaching room with a number of machines with GRASS 5.0 installed, preferably also with pre-installed locations, and session logins for participants. Maybe it would be best organized on Saturday 14 September. The workshop will show some of the possible applications of R on spatial data, then show how to install R, how to install the interface, and how to use it for raster and sites data. The location to be used will probably be the Burrough/McDonnell/gstat Maas Bank soil pollution data set, but this may be supplemented by point-pattern data. Input from the local organisers is crucial, and I hope you will contact me about this soon. I plan to be in Trento early to help set it up before the conference begins.

Abstract n.29


Corresponding author
Bivand Roger, Norwegian School of Economics and Business Administration, Breiviksveien 40, N-5045 Bergen, Norway
Other authors





Development of the GRASS/R interface - GIS and statistical data analysis


This presentation is a follow-up from: Bivand, R., Neteler, M. (2000): Open Source geocomputation: using the R data analysis language integrated with GRASS GIS and PostgreSQL data base systems. It covers subsequent changes in R, in particular memory management, developments in the interface itself related to GRASS 5.0, and foreshadows work on a vector interface based on the 5.1 technology. Work on a map() function within R will also be presented in this context, and opportunities for further collaboration with the R community described. The presentation constitutes an introduction to the proposed GRASS/R interface workshop, but can exist independently. One of the goals is to elicit user feedback on the usefulness of interface work, and how it should be encapsulated, based on the fact that every platform that can run GRASS can also run R. This implies that many data analysis tasks can potentially be passed off to R, for example in shell scripts.

Abstract n.30


Corresponding author
Sboarina Chiara, Centro di Ecologia Alpina, Viote del Monte Bondone, 38040 Sardagna (TN), Italy
Other authors





Development of a complete climate database using a new GRASS module


The aim of this work is the development of a gridded topographic and monthly mean climate database for one particular region, Trentino in the northeast of Italy. The available data are the daily minimum and maximum temperature and the daily total rainfall for a period of ten years (1st Jan 1990 - 31st Dec 1999) from 64 measurement stations while the complete database will include also monthly mean temperature, relative humidity and solar radiation. First the gaps in the time series of data were filled calculating the linear regression between the same variable of the nearest measurement stations. The monthly means or the cumulative values of the available variables were computed and then an interpolation method was used to create a gridded database of these variables. The mean monthly temperature was calculated in GRASS using r.mapcalc while for humidity and solar radiation an existing tested program was implemented as a new module of GRASS. This module, named r.clim, calculates monthly mean relative humidity and solar radiation from minimum and maximum monthly temperature, total monthly rainfall, elevation, slope, aspect, latitude and eastern and western orographic obstructions. The product of this work are 120 maps, one for each month of the ten years, for each variable.

Abstract n.31


Corresponding author
Mitchell Scott, Department of Geography, University of Toronto at Mississauga, 3359 Mississauga Road, Mississauga, Ontario L5L 1C6, Canada
Other authors
Csillag Ferko, Department of Geography, University of Toronto at Mississauga
Tague Christina, Department of Geography, San Diego State University



Advantages of open-source GIS to improve spatial environmental modelling


Spatial heterogeneity is crucial for both GIS and environmental models; for the first, it is more of a representation issue, for the second it is more of a functional issue. As a common element, it can be seen as an interface between them to guide research. In developing models, spatial heterogeneity influences conceptualization of environmental function; appropriate units to represent the system will vary depending on heterogeneity. Open source GIS is ideal because it allows everyone to introduce concepts "on top" of a base representation (whether the system involves watersheds or market areas, the underlying GIS layers may still be raster). The open source environment provided by GRASS is also very helpful since we can investigate other source code, and draw upon the knowledge of an enthusiastic development community. The statistics to handle spatial heterogeneity do not match 1:1 with GIS data models; significant expertise is required for the translation. To handle spatial heterogeneity for functional models, we have built sampling (random, regular, & stratified), partitioning, and model-construct routines in and around GRASS. These tools allow us to track and evaluate functionally similar representations and their associated uncertainties in various models. Samples of our applications are presented.

Abstract n.32


Corresponding author
Panatkool Apirak, Computing Research and Development , 112 Thailand Science Park, Paholyothin Rd., Klong 1, Pathumthani 12120, Thailand
Other authors
Laoveerakul Sitthichai , Computing Research and Development




Geographical Distributed System using Grid


Geographical data is typically generated and stored locally which is then processed by a limited number of specialized computations or services on that site. Nonetheless, to get a global view, one needs to gather information from various locations and process them by using different specific services. Since each location does not contain all the required services, we normally gather data from those locations to be processed on a central server that has all the services. This centralized server paradigm contains some drawbacks such as single point of failure, network congestion, data inconsistency and others. Furthermore, in some real time applications, e.g., traffic control, pollution modeling etc., these problems must be solved. To alleviate the situation, we propose a decentralized paradigm based on the computational grid model. On this model, data is distributed on the Internet and will be processed by new kind of services called the web services. We adopt the peer-to-peer (P2P) protocol to define the communication between any two adjacent parties. That is each site can assume either client or server role when communicating with the other site. By using this protocol, a web service can be moved to any site to process data thereby resolving the need to have a centralized server.

Abstract n.33


Corresponding author
Kaitala Seppo , Finnish Institute of Marine Research , O Box 33, FIN-00931 Helsinki, Finland, Finland
Other authors
Shavykin Anatoly, Murmansk Marine Biological Institute, Russia
Volkov Vladimir , Nansen International Enviromental and Remote Sensing Centre, St.Petersburg, Russia



Environmental GIS database for the White Sea


Into the White Sea GRASS-GIS database bathymetric, hydrological, hydrochemical and hydrobiological data is included for the years 1985 1986, 1989, 1990. Data was collected mainly once in a month during the ice free periods. Bathymery data with resolution 1 by 0.5 minutes was used to model the bottom surface of the White Sea. Russian coordinate system of Pulkova 42 was used. DCW vector data were used for the shoreline and the White Sea was divided to 7 traditional geographic areas. Raster layer of the bathymetry was divided with vector polygons to appropriate geographic areas and the area and volume were calculated. The hydrological and chemical parameters as water temperature, salinity, inorganic nutrients were included for the years 1985 1986, 1989, 1990. The change of the ratio of inorganic nutrients is used to evaluate the regulating factors of phytoplankton succession during the growth season. The river runoff of dissolved substances at the outlets of rivers and the distribution of organic pollutants in the sediments is used to estimate human impacts in the White Sea ecosystem. The database is used to validate numerical ecosystem modeling applications in purpose to evaluate possible effects of the climate change and growing human impact on the ecosystem.

Abstract n.34


Corresponding author
NEMOTO Tatsuya, Department of Geosciences, Graduate School of Science, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi-ku Osaka 558-8585, Japan
Other authors
RAGHAVAN Venkatesh, Media Center, Osaka City University
MASUMOTO Shinji, Department of Geosciences, Graduate School of Science, Osaka City University
SHIONO Kiyoji, Department of Geosciences, Graduate School of Science, Osaka City University


Development of SISGeM - An Online System for 3D Geological Modeling -


We have developed a prototype Spatial Information System for Geological Modeling (SISGeM). SISGeM has been successfully implemented by integrating the GIS and relational database system (RDBMS) with the Internet. SISGeM will allow better management of geological information. More importantly, it's capabilities of generating data for visualizing 3D geologic features can server as means to provide value-added geological information. SISGeM was implemented on a personal computer using the Linux operating system. The spatial data management is carried out using GRASS. NVIZ has been used for visualizing geologic cross-sections and generating 3D animations. Online access to GRASS functions has been enabled using the basic framework of GRASSLinks interface. Several additional modules were incorporated into web interface to tailor the system for its present needs. RDBMS manage data that is recorded in a field outcrop or derived by digitizing hardcopy geologic maps. In this presentation, we discuss about the spatial information system in light its ability to manage geological data and its visualization capabilities. Further, we present examples of geologic model generated using actual geologic data derived from geologic survey.

Abstract n.35


Corresponding author
YONEZAWA Go, Department of Geosciences, Graduate School of Science, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan
Other authors
NEMOTO Tatsuya, Department of Geosciences, Graduate School of Science, Osaka City University
MASUMOTO Shinji, Department of Geosciences, Graduate School of Science, Osaka City University
SHIONO Kiyoji, Department of Geosciences, Graduate School of Science, Osaka City University


3-D Geologic Modeling and Visualization of Faulted Structures


In order to draw the geologic map of geologic structure that has been formed through the various events, we need to establish a logic model of geologic structure that defines recursively a logical relation between geologic bodies and surfaces. We extended an algorithm that has been formed through a sequence of sedimentation and erosion to apply a faulting. We newly defined a faulting that a fault plane divided a three dimensional (3-D) region into two blocks, the existing geologic body and the open space. And, the geologic structure formed by a sedimentation and erosion is saved in each block as it is. When such an extended logical model is used, logical model of faulted geologic structures can be analyzed without changing flow of the usual computer processing includes the estimation of surfaces and the visualization of 3-D geologic structure. In this presentation, we present the result of visualization for 3-D faulted geologic structures using the visualization tool NVIZ that is attached to GRASS.

Abstract n.36


Corresponding author
Nieminen Juhana, University of Helsinki, Environmental Biology GIS Laboratory, P.O.Box 44 (Jyrangontie 2) FIN-00014 Helsinki University, Finland
Other authors
Kaitala Seppo, University of Helsinki, Environmental Biology GIS Laboratory
Kiviaho Katja, University of Helsinki, Environmental Biology GIS Laboratory



University of Helsinki Environmental Biology GIS Laboratory


The Arctic is amongst the most fragile environments of the world. The program of then Conservation of Arctic Flora and Fauna (CAFF) was established to address and advise on the special needs of Arctic ecosystems, species, and their habitats. CAFF is a high level intergovernmental forum of the Arctic Council supported by all eight Arctic countries. The CAFF program decided to produce a report on Arctic biodiversity status and conservation needs. Environmental Biology GIS Laboratory of the University of Helsinki had a key role in producing maps and graphics for the report. Rapid response in map projections development by the GRASS community made possible to utilize a low cost tool for analysis, manipulation and visualization of diverse data sets coming from the eight Arctic countries and several independent researchers around the world.

Abstract n.37


Corresponding author
Clerici Aldo, Università di Parma - Dipartimento di Scienze della Terra, Parco Area delle Scienze 178/A - 43100 Parma, Italy
Other authors





A GRASS GIS based Shell script for Landslide Susceptibility zonation.


Among the many GIS based methods for the construction of maps portaying the probability of future occurrence of landslides, the so called Conditional Analysis applied to the Unique Condition Units proved to be conceptually very simple, but operationally complex. In fact, in this method landslide susceptibility is simply expressed as landslide density in correspondence with different combinations of instability factors. On the other hand, the method requires a long and well defined sequence of operations and can be long, tedious and error prone. Furthermore, in the previous use of this method, some aspects, like the definition of landslide dimensions and the assessment of factor values, turned out to be serious pitfalls. For these reasons, a shell script mainly based on the GRASS GIS commands was created in which the upper edge of the landslide main scarp is considered as the morphological element representing the landslide and is used to define both the factor combination and landslide extension. The script, starting from a Landslide Inventory map and a number of factor maps, automatically carries out the whole procedure resulting in the construction of a Landslide Susceptibility map. All these characteristics make the procedure easy and fast, making Landslide Susceptibility zonation on a regional scale with high spatial resolution possible in a very short time.

Abstract n.38


Corresponding author
Reiter Bernhard, Intevation GmbH (GRASS Developer, FreeGIS Project Founder, FSF Europe Coreteam), Georgstraße 4, 49074 Osnabrück, Germany
Other authors





How GRASS' development reflects Free Software history and what to expect next.


In 1999 GRASS development took a successful turn and is growing ever since. Before there was a hole in which GRASS development and its user base effectively shrunk after GRASS lost the support of the US military. This roughly reflects the historic periods of Free Software. In the beginning everybody used Free Software in academia, followed by a trend towards proprietory software and since the early 1980s an explicit turn towards Free Software brought a new bloom which was noticed broadly at the end of the 90s. This paper will explain the basics of Free Software, the motives behind it and the common licensing methods. It will shed light on why we see so many advantages in it. The steps and reasoning behind involved applying this to GRASS are illustrated. After drawing the historic similiarities this gives the basis for an disscussion and outlook on GRASS' future.

Abstract n.39


Corresponding author
Wagner Jan-Oliver, Intevation GmbH (FreeGIS coordiator), Geogstrasse 4, 49074 Osnabrueck, Germany
Other authors





How Free GIS components grow solutions


GRASS is the visible cornerstone of available Free Software for geographic information processing. But many more components are known to exist. (The FreeGIS project lists about 150 entries in early 2002.) This paper will show the development perspective on how these components can grow together in the future to tackle all that tasks users will want to accomplish. It will cover selected software and solution examples from a number of subfields beside GRASS like formats (GDAL, OGR), databases (postgis), frontends (mapserver), and interactive exploration (thuban). Additionally there will be comparisons to other successful Free Software areas and how development and user community has grown there. For many practitioners of Geographic Information Systems developing software is a side issue. A lot of computer scientists themselves are not familiar with how Free Software development fosters the communication and involvement of users. Thus it will be shown how to use the FreeGIS resources in particular to benefit developers and users alike.

Abstract n.40


Corresponding author
Alameh Nadine, Global Science & Technology, 6411 ivy lane Suite 300, Greenbelt, MD 20770, USA
Other authors





Open Source OGC Web Mapping Servers and Clients: Experience and Lessons Learned


The MIT Computer Resource Laboratory has been recently involved in prototyping technologies that facilitate networked access to geospatial data. This paper is based on the experience of building two open source tools that originated within that laboratory: the MITOrthoServers and the MITOrthoTools. MITOrthoServers provide, via the OGC Web Mapping Service (WMS) interface, customized snippets of digital orthophotos using just-in-time geoprocessing. The MITOrthoTools are extensions to popular GIS packages that can communicate with any WMS-compliant web server to retrieve maps directly into a user’s mapping window. Providing these technologies as open source accompanied with easy-to-use documentation enabled developers to quickly move into the web services space by customizing the open source tools to fit their client’s needs. These technologies, currently available for download at freegis.org and www.ogcnetwork.org, are continuing to grow as a result of the collaborative open source development process. After providing an overview of the tools, this paper discusses the motivation behind following an open source approach. The paper also highlights the efforts to enhance, document, support and sustain the tools in a rapidly growing field. The paper concludes with lessons learned on the role of such open source efforts in promoting a future GIS Web Services Architecture.

Abstract n.41


Corresponding author
O'Donnell Scott, self, 1820 Winfield Dr., Lakewood, CO, 80215-2553, USA
Other authors





Weather Radar Enhanced Flash Flood Forecasting


Located in the Rocky Mountains of the United States, Colorado is subjected frequently to high intensity, short duration, summer season convective precipitation. Many mountainous and urbanized watersheds have rapid hydrologic responses to these storms and are prone to flash flooding. By coupling weather surveillance radar reflectivity with a distributed rainfall/runoff model, model results provide an early indication of flooding potential within the downstream watersheds of the intense storms. In real-time, weather radar reflectivity is translated into areally distributed rainfall estimates and are used for input to the GRASS rainfall/runoff model, r.hydro.CASC2D. The resulting output maps and files are analyzed to notify emergency managers of potentially dangerous flooding conditions caused by the thunderstorms. Storm locations are shown by displaying the radar reflectivity and rainfall intensity maps. The generated graphical displays provide understandable information regarding the location, timing, and flooding potential for the watersheds downstream of the severe storms. At locations selected for their vulnerability to flooding (road crossings, bridges, towns, etc.) any stream forecast exceeding Alarm or Alert levels trigger appropriate warning responses. The improved lead-times provide additional time to evacuate the affected areas.

Abstract n.42


Corresponding author
Zaman Md. Shahid Uz, University of the Ryukyus, Dept. of Information Engineering, 1, Senbaru, Nishihara 903-0213, Japan
Other authors
Miyagi Hayao, University of the Ryukyus
Chen Yen-Wei, University of the Ryukyus



GIS Oriented platform for solving real world logistic vehicle routing problem


Logistics optimization problems related with vehicle routing such as warehouse locating, track scheduling, customer order delivery, wastage pickup etc. are very interesting and important issue to date. Many systems/algorithms have been developed/proposed to optimize those logistics problems. But majority of them can not handle the real world spatial data directly, or those can handle are mostly dedicated for a particular problem. The system developed for a particular problem may not be suitable for others due to the variation of different nstraints. The constraints may include geographical, environmental and road traffic nature of the working region along with other constraints related with the particular problem. So user always need to modify/change the original routing algorithm in order to fulfill the purpose. In our study, we proposed a general purpose Geographical Information System (GIS) oriented vehicle routing system by combining GIS road map and Database Management System (DBMS) together so that developed optimization algorithm can interact with real world spatial data directly to solve different kind of vehicle routing problems. Finally, the platform can also be extended to build new Spatial Decision Support System (SDSS) for solving complex vehicle routing problems.

Abstract n.43


Corresponding author
Nieminen Juhana, Environmental Biology GIS Laboratory, University of Helsinki, P.O.Box 44 (Jyrangontie2) FIN-00014 Helsinki University, Finland
Other authors





Teaching GIS the GNU way


Geographical Information Systems (GIS) have become a standard in geographical, environmental, biological and forestry related sciences. Teaching new techniques, analysis methods and cutting the edge models may be an expensive task for educational institutions and as well as dedicated students. The Internet has boosted the availability of GNU software which have developed to such quality that it has become less of a challenge for University teachers, researchers and students to start using them as everyday tools. The Division of Environmental Biology at the University of Helsinki has been developing a student training program on information technology and geographical information using available GNU tools for the last five years with positive results. The curriculum is composed of basic Linux training which includes the introduction to essential GNU tools (e.g. Emacs, GIMP, Sketch, Octave & R) and GIS courses at basic and advanced levels using GRASS. The use of GNU software in education has become a potential alternative to proprietary software leaving the limited financial resources available for actual teaching and research. Keywords: GIS; education; GRASS; GNU

Abstract n.44


Corresponding author
Nieminen Juhana, Environmental Biology GIS Laboratory, University of Helsinki, P.O.Box 44 (Jyrangontie 2) FIN-00014 Helsinki University, Finland
Other authors





GIS-based habitat potentiality model: In search of natural hotspots


Decision makers and as well as the general public show an increasing interest in understanding the criterion for conserving and protecting endangered habitats under the EU habitats directive. Often a well structured visualization may be more convincing than a thick report explaining the same facts. The distribution and abundance of rare plant and animal communities influence the spatial arrangement of endangered habitats across landscapes. A habitat potentiality model has been created and tested to find the remaining potential endangered habitats using ecological criteria, GIS analysis and visualization to pin point the remaining hotspots. A pilot study was performed on a 1678km2 Vantaa River watershed area using heterogeneous digital and analog data. The method described may be applied to several scales of digital terrain data and used to delimit major habitat types from each other. The method combines the use of known GIS analysis techniques (e.g. overlay, weighting, multi-criterion evaluation, standardization) with empirical evaluation and expert knowledge on ecological factors. Keywords: modeling; GIS; ecological expert knowledge; Vantaa River; hotspot, endangered habitats

Abstract n.45


Corresponding author
Mitasova Helena, Dept. of Marine, Earth, and Atmospheric Sciences, North Carolina State University, 1125 Jordan Hall, NCSU Box 8208, Raleigh NC 27695, USA
Other authors
Drake Thomas, Dept. of Marine, Earth, and Atmospheric Sciences, North Carolina State University
Harmon Russel, Army Research Office, USA
Hofierka Jaroslav, Department of Geography and Geoecology, University of Presov, and MEAS NCSU, USA
McNinch Jesse, Virginia Institute of Marine Science, College of William and Mary

Spatio-temporal monitoring of evolving topography using LIDAR, RTKS and sonar data


Sustainable management of highly dynamic costal topography with fast moving sand dunes, eroding shoreline and anthropogenic changes to bathymetry and coastal terrain represent significant challenges for coastal management. Modern mapping technologies bring capabilities to monitor this dynamic environment with unprecedented spatial and temporal detail. GRASS GIS provides a wide range of tools to support such monitoring as well as analysis and creation of spatio-temporal, multiscale models of the monitored areas. The application of LIDAR (LIght Distance And Ranging - Laser altimetry), Real time kinematic GPS, Infrared Digital Orthophotography and side scan sonar data for monitoring and assessment of coastal dynamics will be demonstrated for two study areas of North Carolina coast: a) Jockey's Ridge, the largest sand dune on the East coast, and b) Bald Head Island near the mouth of the Cape Fear river. The spatial interpolation and topographic analysis of these different types of high resolution data, including the application of anisotropy, will be demonstrated and advantages and needed improvements to the s.surf.rst will be discussed. Visualization using multiple surfaces, cutting planes and spatially variable resolution will be presented using the latest version of nviz.

Abstract n.46


Corresponding author
Mitasova Helena, Dept. of Marine, Earth and Atmospheric Sciences, North Carolina State University, 1125 Jordan Hall, NCSU Box 8208, Raleigh, NC 27695-8208, USA
Other authors
Hofierka Jaroslav, North Carolina State University, Raleigh, USA and University of Presov, Slovakia
Mitas Lubos, Department of Physics, North Carolina State University, Raleigh, USA



GRASS and modeling landscape processes using duality between particles and fields


New, high resolution mapping technologies provide data for creating spatio-temporal models of landscapes undergoing substantial changes due to interactions of natural processes and socio-economic development. Availability of such data requires a new generation of models, capable of handling simulations at multiple scales including complex interactions typical for modeling at high level of detail. We present one of the promissing approaches to modeling of spatial processes in complex landscapes based on path sampling method. This method uses the duality between particle and field representation to solve the governing equations without the need to create complex meshes, making it easy to link the simulations with GIS. We demostrate the approach using the simulation of water and sediment flow for the Centennial Campus area undergoing significant change in land cover and topography due to the transformation of a natural area to urban and recreational land use. The use of GRASS5 for creating models of past, current and future landscapes, derivation of parameters for hydrologic and sediment transport simulations and visualization of resulting time series using 2d and 3d animations will be presented. We also discuss the issues related to the implementation of path sampling simulation tools for wide range of applications within GRASS.

Abstract n.47


Corresponding author
Bonk Radoslav, Ph.D student, Dept. of Physical Geography and Geoecology, Faculty of Natural Sciences, Mlynska dolina, 842 25, Slovakia
Other authors





Scale-dependent geomorphometric analysis for glacier mapping at Nanga Parbat,...


Scale-dependent geomorphometric analysis for glacier mapping at Nanga Parbat, Pakistan Nanga Parbat, the ninth largest mountain in the world, is the result of the complex interaction of climatic, tectonic, and surface processes. Glaciation has had a significant impact on the landscape, although the magnitude of alpine glacier erosion on the topographic evolution of the massif still remains unclear. Addressing this problem requires the ability to map alpine glaciers and determine the extent of previous glaciations using topographic information. Consequently, the objective of the research was to determine if topographic analysis can be used to accurately map modern-day glaciers. A digital elevation model was generated, and the geomorphometric parameters of slope angle, slope aspect, profile curvature and tangential curvature were computed and evaluated. Elemental terrain-form objects were generated on the basis of homogeneous first-order parameters. Spatial clumping was then used to identify elemental terrain-form objects. Object-oriented analysis was then used to compute twelve attributes for each object. These attributes were then used for classifying alpine glaciers. The results indicated that the slope parameter was most useful for identifying the influence of glaciation on the landscape. The curvature parameters produced highly variable results. Ultimately, the approach was not able to accurately delineate alpine glaciers on the landscape. Analysis indicated that topographic analysis must model more than one hierarchical level of the topography. The results indicated that attributes representing shape and the contextual relationships between objects must be included. Therefore, more research regarding automated mapping of glaciers is warranted, and more sophisticated analysis and modeling of the topography is required.

Abstract n.48


Corresponding author
Frigeri Alessandro, Dipartimento di Scienze della Terra - Geologia Strutturale e Geofisica , Università degli Studi di Perugia, P.zza Università 1, 06100 Perugia, Italy
Other authors
Federico Costanzo, Dipartimento di Scienze della Terra - Università degli Studi di Perugia
Minelli Giorgio, Dipartimento di Scienze della Terra - Università degli Studi di Perugia
Pauselli Cristina, Dipartimento di Scienze della Terra - Università degli Studi di Perugia
Caldarella Cristina,

Identifying wrinkle ridges structures from Mars MOLA and Viking mission data: usin


Identifying wrinkle ridges structures from Mars MOLA and Viking mission data: using Grass in planetary geology. The numerous missions to Mars have sent information about the red planet during last decades. Recently it has been shown that Mars has ice on its poles and that ice could be probably present all over the planet subsurface. The presence of ice on subsurface could drive geological structures not observable on the Earth. The most inferred ice-related morphological structures are the so called 'wrinkle ridges'. Grass GIS has been used to import and store in a common geospatial database Viking image data and the latest MOLA (Mars Orbiter Laser Altimeter) DEM. GRASS raster based analysis at different resolution has been used to identify wrinkle structures. The image modules are used to align Viking data with MOLA DEM. The spatial distribution of these structures is then analyzed to find out how they relate to other geological features.

Abstract n.49


Corresponding author
Frigeri Alessandro, Dipartimento di Scienze della Terra - Geologia Strutturale e Geofisica , Università degli Studi di Perugia, P.zza Università 1, 06100 Perugia, Italy
Other authors
Chang Yet-Chung, Institute of Oceanography, National Taiwan University, Institute of Oceanography.




Implementing the automatic extraction of ridge and valley axes using the PPA aglor


Implementing the automatic extraction of ridge and valley axes using the PPA aglorithm in Grass GIS.----- The profile recogniction and polygon breaking (PPA) algorithm permits to extract the ridge and valley axes automatically over wide areas maps. The algorithm simulates the human thinking in the line drawing process. First a process of profile recognition takes all the points close to the possible axes and connects them as a belt of closed polygons. The polygons are then broken into a continous line and a smoothing algorithm adjusts the line. The algorithm, originally coded in Fortran, has been adapted to be used for raster analysis in Grass. The module takes a raster map, typically a DEM, and outputs a vector file representing the ridge or the valley axes. Various testings show the features of the algorithm and the implementation into Grass Gis.

Abstract n.50


Corresponding author
Frigeri Alessandro, Dipartimento di Scienze della Terra - Geologia Strutturale e Geofisica , Università degli Studi di Perugia, P.zza Università 1, 06100 Perugia, Italy
Other authors
Minelli Giorgio, Dipartimento di Scienze della Terra, Università degli Studi di Perugia




Low cost PDA/Gps based field logging solution for GRASS data.


The need of electronic data to be inserted in geospatial databases lead us to develop a simple Personal Digital Assistant (PDA) based field log system to store punctual data to be imported directly in GRASS as a site-file. The application communicates with a GPS unit so positional data are automatically stored, allowing the user the user to input only categorical data. The system frame has been chosen so that the overall cost of a basic system is very low but no changes are required to use it with professional GPS units and different PDA systems. To test the system a Structural Geology survey has been made in Central Appennines and the pros/and contros of the system have been reported. The software is released under the terms of GPL license so that everyone could improve it or adapt it to his own field logging campaign.

Abstract n.51


Corresponding author
Candidi Tommasi Crudeli RafDouglas, Studio Candidi Tommasi, v. Torino, 90/3 - 33100 Udine, Italy
Other authors





The use of LIDaR technologies in Grass


The developement of Light Impulse Detection and Ranging (LIDaR) offers new horizons to the professional users of geographic systems. In this presentation a pratical use of LIDaR technologies in Grass is given: starting from the preliminary planning of a mission and basic operations with LIDaR datasets, some layout strategies and some ways to extract useful morphologic informations will be dealt with. Furthermore, the developement and potential problems of algorithms for DTM extraction will be discussed. As a last point the integration with satellite data for periodical updates will be focussed.

Abstract n.52


Corresponding author
Raghavan Venkatesh, Media Center, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan
Other authors
Masumoto Shinji, Faculty of Science, Osaka City University




Implementing an online spatial database using the GRASS GIS environment


Spatial databases offer a means to build and deliver geographically referenced information over computer networks. There are already several Open Source Software packages available that could to tailored to develop spatial data infrastructures. In the present paper we describe the salient features of an spatial database that was developed by integrating the GRASS GIS and PostgreSQL Object-Relational database into a Web based client/server environment. The system has been used to build and manage spatial databases. The system can facilitate easy and rapid collection and dissemination of spatial information. Since the system is independent of any proprietary software, it is easily and economically adapted in a distributed spatial database environment. Further, the system can serve as a platform for rapid implementation of spatial data infrastructures through collective participation and also serve as a means for standardizing data collection. Such efforts will help coordinate better strategies in environmental assessments, resource management and hazard mitigation through speedy sharing of spatial information amongst specialist groups and also the general public.

Abstract n.53


Corresponding author
Harahsheh Hussein, United Arab Emirates University, P.O.Box 17551 Al-Ain, United Arab Emirates
Other authors





Assessment and monitoring of desertification in north of Jordan


The use of Remote Sensing to provide environmental information on the earth’s surface, and GIS to manipulate and analyze information extracted from different sources of data, are actually important tools for the understanding of environmental hazards. This is applied especially to study desert conditions, desertification monitoring, assessment and mapping. This study applied remote sensing techniques to extract and evaluate the physical parameters related to land degradation and desertification such as water, vegetation, geomorphology, drainage system, soil moisture…etc. Also the study used multi-temporal analysis of Landsat TM data to analyze the status of vegetation, as well as the increasing of irrigated area during the period of 1983-1997. The results of analysis show clearly the net decrease of vegetation cover. We note also the presence of unpalatable species, which replaced the palatable ones. This situation is a direct result of the great deterioration of natural vegetation cover. GIS tool was used to analyze and combine different types of parameters (Land use, soil,…). As a result of our analysis of the GIS layers, we obtained maps of soil susceptibility to erosion, vegetation degradation and salinization. A final stage of this research was the creation of desertification map describing the different types of land degradation occurring in the study area.

Abstract n.54


Corresponding author
Blazek Radim, ITC-irst, Via Sommarive, 18, Italy
Other authors
Neteler Markus, ITC-irst




The new GRASS 5.1 vector architecture


The presentation describes the new GRASS 5.1 vector library architecture. This new architecture overcomes the vector limitations of GRASS 4.x-5.0 by extending the vector library with 3D capabilities and database management (DBMS) support. The new vector library also supports the integration of external GIS vector files in SHAPE or other vector formats without the necessity of importing them into GRASS. Besides internal storage of vector geometry and attributes the optional DBMS support targets to utilize PostGIS/PostgreSQL for data storage. This enables users to manage large datasets with simultaneous external write access to the database tables. Additionally an ODBC interface is implemented to establish connection to other, even proprietary database systems. The presentation will contain also live sample applications of a running PostGIS data storage session as accessed by several GRASS clients at the same time as required for traffic management or other systems.

Abstract n.55


Corresponding author
Worth Carl, USC Information Sciences Institute, 3811 N. Fairfax Dr. Suite 200; Arlington, VA 22203, USA
Other authors
Schott Brian, USC Information Sciences Institute




GIS Goes Mobile: Using GRASS on Handheld Computers


Mobile computing technology has the potential for obvious benefits to producers and consumers of GIS data. Today's handheld computers now have the compute resources necessary to run full-featured UNIX-like operating systems such as Linux, along with the X Window System. However, handheld computing environments still have significant constraints including limited system/storage memory and severely limited user-interface. We discuss the challenges we faced as we deployed the GRASS software on Compaq iPAQ handheld computers, (plus GPS), running the Familiar Linux distribution. We also discuss the contributions we have made to GRASS user-interfaces and packaging to make it more suitable for use on handheld computers.

Abstract n.56


Corresponding author
LUO Jiancheng, Dr., LREIS, Institute of Geography, 917 Building, Datun Road, Beijing, China, 100101, China
Other authors
Chen Qiuxiao, PhD Student
Zheng Jiang , PhD Student
Zhou Cheng Hu, Professor
Leung Yee, Professor

A Knowledge-Integrated Stepwise Optimization Model for Feature Mining in RS Image


The feature space is always separated by one-off assumption of the probability distribution functions (PDFs) for all features in order to model the mixture distribution. However, due to the inter-overlapping phenomenon among the points or the confusing influence from the around discrete points, it is very difficult to obtain the subtle and procedural structure of the mixture distributions of the feature space which influences the accuracy and interpretability of the results in the procedure of further analysis. Extending on the method of Gaussian mixture modeling and decomposition (GMDD), a new feature mining method named stepwise optimization model (SOM) with Genetic Algorithms (GA) is proposed in this study for the extraction of tree-like hierarchical structure of feature distributions in the mixture feature space. To better approximate the reality, integration of SOM-GA with symbolic geographical knowledge is thus essential in the feature mining of remote sensing images. A knowledge-integrated SOM-GA model that combines the power of SOM-GA and the logic reasoning of rule-based inference is proposed in the present study. In addition to conceptual and technical discussions of the model, our arguments are substantiated by a real-life application. It is shown by the results of practical applications of land-cover classification that the proposed method performs more accurately, more simply in structure, more easily in integration with symbolic knowledge and more interpretably. The study shows that such an integrated approach has great potential in remote sensing research.

Abstract n.57


Corresponding author
Menegon Stefano, ITC-irst, Via Sommarive, 18, Italy
Other authors
Fontanari Steno, ITC-irst, Italy
Dassau Otto, University of Hannover, Germany
Blazek Radim, ITC-irst, Italy
Neteler Markus, ITC-irst, Italy

Wildlife Management and Landscape Analysis in the GRASS GIS


Further authors: Stefano Merler, Cesare Furlanello -- In this paper we describe how GRASS GIS resources have been developed and integrated for centralized data archiving and predictive modeling in several wildlife management tasks in Trentino, Italian Alps. In particular, we will present the development of a multiscale site characterization based on the integrated use of orthophoto landuse classification, morphometrical analysis of DEM (altitude, slope, aspect and curvatures) and the quantitative analysis of landscape structure at different scales. The methodology has been applied at a mesoscale (6200 Kmq, 30 Gb ortophoto at 1 meter cell resolution) for the predictive modelling of deer-vehicle collisions and for developing guidelines for the improvement of black grouse habitat improvement, two projects for the Wildlife Management Service of Trentino. We devised an environment of GRASS and R tools (modules and interface scripts) to automate the database preparation, including variables as distance from urban areas and from waters, wildlife population density map, vector line analysis (road curvatures). The data are managed by database tools (PostgreSQL), allowing the development of computationally intensive predictive models. We present variable importance analysis and classification with bagging of tree-based classification models.

Abstract n.58


Corresponding author
Löwe Peter, University of Wuerzburg, , Germany
Other authors





A spatial decision support system for radar meteorologic data in South Africa


The south african weather service SAWB operates a radar station network, which is unique for southern africa. GRASS was introduced to the SAWB radar research center (now: METSYS) in 1999, where it is still used for meteorologic research. An automated import-tool for MDV-files was created to access to the radar data in GRASS. A postgresql database system is used to store data and quicklooks to allow multitemporal queries. The convective nature of precipitation in southern africa creates public demand for severe weather information systems. Such a system was set up in GRASS, using a rule based expert system. It isolates storm cells from the reflectivity fields and derives information about their development stages, which is stored using XML. This information is cast into Email or SMS messages for the interested public for areas of interest. Further, HTML-maps for the whole country are created, which serve as a continously updated front-end for a web-based weather information system. This service can also made accessible for people in remote locations by broadcasting it as a datastream from satellite through the worldspace digital radio system.

Abstract n.59


Corresponding author
Sguerso Domenico, University of Trento, Via Mesiano 77, Italy
Other authors





Reference frames: definition and management


Geographical Information System often required much heterogeneous kind of data, like cartography supports, GPS points, satellite images and so on, generally in different reference frame and/or in different cartography representation. The problem of different datum and representation is a classical problem in geodesy that software generally solves with dedicated tools. This problem has different sources, so different kinds of transformations have to be applied. Starting from reference frame and cartography definitions, this paper would like point out main relations with set-data and which external inputs are necessary for a correct planimetry and altimetry data management. Nevertheless, frequently transformation parameters are not available, so some different procedures with different approximations have to be accepted. Examples with Grass open Gis tools are studied.

Abstract n.60


Corresponding author
Santitamnont Phisan, Chulalongkorn University, Phayathai Rd., Patumwan, Bangkok 10330, Thailand
Other authors





Challenges of Using Open Source GIS - Experiences of Thailand


The article reports experiences in using open source GIS software for research and development at Chulalongkorn university, Bangkok Thailand. Open source software cuts off budgets, gives freedom to student and researcher to work and to support in-depth learning and training. Spin-off research from open source GIS is now accepted and used in the real practice. Several prototype and operational information systems are developed based on integration of various open source technology. Some interesting projects based open source will be briefly overviewed and demonstrated. A conclusion on benefits and pitfalls of using open source GIS software will drawn at the end.

Abstract n.61


Corresponding author
Toma Laura, Department of Computer Science, Duke University, Durham, NC 27708, USA
Other authors
Lars Arge, Department of Computer Science, Duke University
Mitasova Helena, Department of Marine, Earth and Atmospheric Sciences,North Carolina State University



Flow Computation on Massive Grid Terrains


Processing massive data sets presents significant challenges to GIS systems demanding efficient solutions. In our previous work we developed Terraflow, the first terrain flow analysis software designed and optimized for massive data, based on scalable flow routing algorithms. Terraflow greatly outperforms existing systems and is capable of solving problems of a scope and scale that are impractical with traditional algorithms. For instance, the approach allowed us to extract stream network from 6700x4300 grid in 38 minutes instead of over 17 days required to complete only 65% task using the tools currently available in GRASS.The project presented in this abstract is aimed at incorporating Terraflow into GRASS and augmenting it with additional functionality in order to create an integrated flow tracing and watershed analysis tool. To assess the possibility to replace the numerous existing GRASS flow routing modules we investigate their functionality, performance and scalability and compare them with Terraflow. The new GRASS module based on Terraflow will integrate functions for filling depressions, assigning flow directions, computing flow accumulation, and various other watershed related indices. It will provide the user with additional flexibility by choosing between D8 and D-infinity routing. Moreover, it will be based on efficient and scalable algorithms and will greatly extend the size of the grids that can be processed compared to the existing GRASS functions.

Abstract n.62


Corresponding author
Rigon Riccardo, Dept. of Civil and Enviromental Engeneering, University of Trento, via Mesiano, 77, Italy
Other authors
Antonello Andrea, Dept. of Civil and Enviromental Engeneering, University of Trento
Pisoni Silvano, Dept. of Civil and Enviromental Engeneering, University of Trento
Cozzini Andrea, Dept. of Civil and Enviromental Engeneering, University of Trento


Horton: a new set of tools for geomorphological analysis ported into GRASS


We implemented many new routines for the analysis of DEMs. These routines are based upon a suite of programs called Horton and complement those already present in GRASS. The availability of very accurate Digital Elevation models open new inroads to the quantitative gemorphological analysis and Horton tries to fill the gap between the possibility offered by recent theoretical studies and the every day practice. In this note we introduce the main features of the implemented programs and apply it to some sample basin. Among the main applications there are: a new kind of pits filler, an upslope catchement area and a Hack's length estimators, a width function and rescaled width function builder and an implementation of the SHALSTAB shallow landslide model with some extensions.