This paper presents an application of Support Vector Regression method for the prediction of an environmental variable such as the level of natural radioactivity. The basics of the method are described, and some practical considerations are presented, including the meaning of the methods parameters and their influence on the model. The use of the prior data is discussed. It is shown how to include the information on the variance of the measurements into the model. The use of cross-validation for tuning the parameters of the algorithm is presented. Some ideas for detecting the unusual training samples with SVR are discussed. Generally, the case study illustrates the usefulness of the considered approach for automated spatial mapping tasks in the presence of prior data.

Climate change calls for strategic planning that builds resilience in vulnerable areas to manage the associated risks. This paper discusses how adaptive various communities and industries are to climate change in the North West of Victoria (also known as the Victorian wheatbelt), Australia. Indicators of adaptive ability for communities and industries, and the importance of key drivers like government policies, expert advice and empirical evidence of developing this capacity are identified. It also incorporates input from key regional groups as well as current knowledge on adaptability of regional communities to climate change across three major themes: socio-cultural, economic, institutional/infrastructure. Each of these major themes has associated indicators, which in turn have an individual suite of measures, albeit all contributing to the overall adaptive capacity and spatial variability of these capacities. A Geographic Information System is used to collect and analyse the data and spatially represent the indicators and indices. Workshop participants used their expert-judgment to assess and weight indicators, measures and themes. The stakeholders participatory assessment, the quantification of diversified data and interests and the importance of multiple policy outcomes make the findings locally relevant. We find that capacity and preparedness to adapt to climate change varies substantially across communities and different parts of the grains industry.

Regularized Spline with Tension was used to interpolate two data sets representing radioactivity measurements at 200 locations. A cross-validation analysis showed that the size of the training data sets was too low to find optimal parameters using the cross-validation procedure. The resulting surfaces were strongly smoothed and less realistic than expected. Therefore empirical interpolation parameters were used to interpolate the data. Despite the fact that this empirical selection did not produced interpolation results with a lower overall predictive error, it preserved better local fluctuations and anomalies of the phenomenon. The detection of these features is important in radioactivity monitoring and emergency situations. The poor reliability of cross-validation was also confirmed by evaluation data set. It was concluded that the optimization of interpolation parameters cannot rely on cross-validation when the modeled phenomenon is not sufficiently sampled. The sampling density should be sufficient to represent spatial variations of the phenomenon and, at the same time, allow the optimization of interpolation parameters using automated procedures.

This paper presents an approach, based on universal kriging, for automatic mapping of monitoring data. The performance of the mapping approach is tested on two datasets containing daily mean gamma dose rates in Germany reported by means of the national automatic monitoring network (IMIS). In the second dataset an accidental release of radioactivity in the environment was simulated in the South-Western corner of the monitored area. The approach has a tendency to smooth the actual data values, and therefore it underestimates extreme values, as seen in the second dataset. However, it is capable of identifying a release of radioactivity provided that the number of sampling locations is sufficiently high. Consequently, we believe that a combination of applying the presented mapping approach and the physical knowledge of the transport processes of radioactivity should be used to predict the extreme values.

This study describes a method used to improve ore grade estimation in a limestone deposit in India. Ore grade estimation for the limestone deposit was complicated by the complex lithological structure of the deposit. The erratic nature of the deposit and the unavailability of adequate samples for each of the lithogical units made standard geostatistical methods of capturing the spatial variation of the deposit inadequate. This paper describes an attempt to improve the ore grade estimation through the use of a feed forward neural network (NN) model. The NN model incorporated the spatial location as well as the lithological information for modeling of the ore body. The network was made up of three layers: an input, an output and a hidden layer. The input layer consisted of three spatial coordinates (x, y and z) and nine lithotypes. The output layer comprised all the grade attributes of limestone ore including silica (SiO2), alumina (Al2O3), calcium oxide (CaO) and ferrous oxide (Fe2O3). To justify the use of the NN in the deposit, a comparative evaluation between the NN method and the ordinary kriging was performed. This evaluation demonstrated that the NN model decisively outperformed the kriging model. After the superiority of the NN model had been established, it was used to predict the grades at an unknown grid location. Prior to constructing the grade maps, lithological maps of the deposit at the unknown grid were prepared. These lithological maps were generated using indicator kriging. The authors conclude by suggesting that the method described in this paper could be used for grade-control planning in ore deposits.

This paper investigates the performance of two neural network (NN) methods viz. a radial basis function network (RBFN) and a multilayer feed forward network (MFFN) to predict the radioactivity levels at a given test site. A comparative evaluation of the two networks is done using Root mean square error (RMSE), Pearsons r, Mean error (ME) and Mean Absolute error (MAE). It was found that the RBFN performed marginally better compared to the other method

The modern Lule harbour-side dates back to 1649 when the old city was abandoned because its harbour-side and approaches had become too shallow to be useful. This shallowing, due to glacio-isostatic rebound, affects the new town also, but the results have been mitigated by coastal engineering. As a result of uplift and engineering, former harbour-side land is now far enough from the present shoreline for any maritime artifacts that might lie beneath them to be unsuspected.We report the outcome of a successful Monash University - Lule University project designed to identify such former harbour-side land parcels by digitizing and georeferencing a series of historical maps dating from the mid eighteenth century. The method has wide application. Typically, the spatial database built using the approach we have exemplified will be readily incorporated into official spatial datasets to be used in decision support. Thus, planning authorities charged with both re-development permit application appraisal and protection of buried heritage items can identify land parcels that would be likely to have buried artifacts, and places within the parcel that are most likely to produce surprises. Land re-developers would be pleased to have access to such information because of the extra scope they can derive for coping with the disruptions to excavation and building works that arise when heritage items are unearthed.

Victor A. Sposito Section Leader, Spatial Sciences PIRVic, Landscape Systems Science Department of Primary Industries, Victoria. AUSTRALIA victor.sposito@dpi.vic.gov.au Ray Wyatt Associate Professor School of Anthropology, Geography and Environmental Studies (SAGES) University of Melbourne Christopher J. Pettit Statewide Leader, Spatial Sciences PIRVic Landscape Systems Science Department of Primary Industries, Victoria. AUSTRALIA

In late June 1998, a damaging flood event occurred at Lakes Entrance (Victoria, Australia), which was partly due to retardation of floodwater flow through the Gippsland Lakes artificial entrance by the level of flood-tide delta sediment accretion in the Entrance and Reeves Channels. Analyses of digitised three-dimensional hydrographic datasets allowed the extent of pre and post-flood event bathymetric change to be visualised and also quantified.Despite the introduction (post-1998 floods) of engineered sediment management regimes in the Entrance and Reeves Channels, flood-tide delta accretion has continued. This sediment accretion potentially increases the vulnerability of infrastructure in low-lying areas of the Lakes Entrance township to any future Gippsland Lakes flood events.