The main RPDS screen showing the screen areas: (a) Main toolbar, (b) Feature map / Classification panel, (c) Status panel and (d) Tabbed function panels. 

Contexts in source publication

Context 1

... are five Ta bbed fu nction panels that can be selected by dicking on the appropriate tab: Te mplate, Report, Indicators, Sites or New Sites (See Figure ...

Context 2

... general structure of the toolkit is presented in figure 1. Some elements of the vector support are shown in the figure but the support is currently fragmented. This is partly due to the different vector models and support there are in different subsystems. For example PGPLOT supports visualizing vector data. The vector model in grid library supports this. Some support, including store and retrieve functions for the RDBMS is also implemented for this vector model in the Grid module. The most common vector format is probably the ArcView Shapefile for which there is also an open source interface (Warmerdam, 1999). PostgreSQL, a major open source RDBMS, has also some built-in vector support of its own. Finally, there is a new open source proj ect for developing a vector database format and tools (Warmerdam ...

Context 3

... design of the WILDSP ACE DSS has benefited from significant input from scientists, biologists, other end-users, system developers, modellers and Geographie Information Systems (GIS) specialists. In fact, the system's blueprint came from scientists and biologists who understand what is most required. Figure 1 depicts a schematic of the system concept. At a glance, one can see that this system offers a generic framework to integrate data, text, maps, obj ects, images, sounds and knowledge input with user-friendly tools, including database management systems, mapping systems, graphics and analytical functions to produce output for interpretation, integration, further analysis and recommendation. The information and tools can be shared. For example, an analytical tool developed for a particular proj ect can easily be adapted by another proj ect, should the same approach be ...

Context 4

... /\,/ Figure 1. The River Run Internet GIS. One of the maps available is the "Base Map" which shows the d,istribution of the water quality stations in the Lower Cape Fear River. Users can select data layers to draw and identify the attributes of each data layer. Notice the locations of the sampling stations and the abundance of agricultural hog farms. The county boundaries, although not visible on the scrolling legend, are outlined as well as labeled on the map . Figure 3. Spatial pattem of dissolved oxygen in March 1999. At this time ofyear the levels of dissolved oxygen are much greater due to the high levels of rainfall typical for the Spring ...

Context 5

... S)'stem Figure 1. A possible integration approacb (actual screen shots from tbe WuNDa navigator and WuNDa GIS) DSS's have been integrated with models before, usually by wrapping the models and calling them from the DSS. Therefore, one approach to integrate DSS's, models and distributed data/metadata management tools could be to integrate existing models through a DSS with the data management and GIS component ( figure 1). The DSS would interface with the models on one sied and with the data management and the GIS component on the other side. This approach would be the proj ect driven approach mentioned earlier, which does not leave any resources to think about long-term architectural ...

Context 6

... S)'stem Figure 1. A possible integration approacb (actual screen shots from tbe WuNDa navigator and WuNDa GIS) DSS's have been integrated with models before, usually by wrapping the models and calling them from the DSS. Therefore, one approach to integrate DSS's, models and distributed data/metadata management tools could be to integrate existing models through a DSS with the data management and GIS component ( figure 1). The DSS would interface with the models on one sied and with the data management and the GIS component on the other side. This approach would be the proj ect driven approach mentioned earlier, which does not leave any resources to think about long-term architectural ...

Context 7

... Uwedat-Navigator (UWENAV, see Figure 1) is a tool which allows the user to parameterize the monitoring network. lt is clearly stated, that before any data can flow into the database, the monitoring network has to be defined. All this information (i.e. structural metadata, [Denzer et al.,1993]) needed, like operator of the network, related hosts and monitoring stations as well as pollutant parameters (see chapter Figure 4) is configured with this ...

Context 8

... February 2000 a proj ect was initiated to develop a web site for visualization of water quality sampling data along the Lower Cape Fear River basin (Figure 1). The River Run website (www.uncwil.edu/riverrun) was developed for educators, university scientists, environmental regulators and the general public for environmental information dissemination. Therefore, the scientific premise for developing the River Run web site is the devlopement of an Environmental Decision Support System (E-DSS). By definition, an E-DSS is an "information system containing at least one component whose purpose is to support human decision making about an environmental issue" (Swayne, et al. 1999, p. 260). The River Run website is an excellent teaching tool which sparks the intellectual curiosity and imagination of its users. lt is particularly appropriate for undergraduate students enrolled in limnology and environmental studies ...

Context 9

... fundamental component of the modelling system is a raster based representation of spatial layers. The basic system layer is a template of the area of interest, and is represented in a rectangular grid of text numbers that extends beyond the boundaries of the active area. Cells that are active are denoted by value " l " , while inactive cells are denoted by value "O". Construction of this basic template is one of first activities undertaken when developing a new model. Figure 1 shows the opening six lines from an example map template. 0000000000000000000000000 0000000000000000000000000 0000000000000000110000000 0000000011111111111000000 0000000111111111111100000 0000000111111111110000000 Figure 1. Map ...

Context 10

... fundamental component of the modelling system is a raster based representation of spatial layers. The basic system layer is a template of the area of interest, and is represented in a rectangular grid of text numbers that extends beyond the boundaries of the active area. Cells that are active are denoted by value " l " , while inactive cells are denoted by value "O". Construction of this basic template is one of first activities undertaken when developing a new model. Figure 1 shows the opening six lines from an example map template. 0000000000000000000000000 0000000000000000000000000 0000000000000000110000000 0000000011111111111000000 0000000111111111111100000 0000000111111111110000000 Figure 1. Map ...

Context 11

... are several tools available in the UWEDAT system (see Figure 1) . We will briefly discuss the three main tools in this ...

Context 12

... EPORTS • MULTI MEDIA Figure 1. Schematic of the WILDSP ACE TM ...

Context 13

... rail chicks are precocial and leave the nest soon after hatching (Applegarth 1938). One adult tends the newly hatched chicks while the other parent continues incubation until all eggs have hatched (Applegarth 193 8, Meanley 1985). The life cycle of a clapper rail (Figure 1) is described in 4 ...

Context 14

... familiar with efficient data management and storage techniques will recognise immediately that the data formats presented in Figure 1 and Figure 2 are far from the most efficient available. For example, the row and column information presented in the first two values of each row in Figure 2 could be made redundant by adopting a protocol of ordering data from the top, left active cell to the bottom, right active cell on Figure ...

Context 15

... familiar with efficient data management and storage techniques will recognise immediately that the data formats presented in Figure 1 and Figure 2 are far from the most efficient available. For example, the row and column information presented in the first two values of each row in Figure 2 could be made redundant by adopting a protocol of ordering data from the top, left active cell to the bottom, right active cell on Figure ...

Context 16

... shell reads this template to develop a Cartesian coordinate system that serves for the representation and storage of other data layers within the model. Figure 2 provides layer information for the active cells identified in Figure 1. In this example, the first two numbers represent the row (R) and column (C) of the selected variable, along with the parameter (P) values associated with the various layers for that grid point, vis. (R, C, P1, P 2 , P3, P4, P5). In this case, the five parameter layers are land use, sub-area, upland erosion risk, stream network, and representative elevation. Figure 2. Parameter ...

Context 17

... we do not believe that these approaches adequately represent catchment processes. For example, the ecosystem approach is largely based on energy flows, biogeochemical cycling and biodiversity (Hamblin 1999). Catchments, however, function differently and are best defined on the basis of flow of mass (i.e. water). They are cascading systems with many different spatial components compared with ecosystems. Accordingly we have developed our own catchment systems framework, shown in Figure ...

Context 18

... reason for the use of such a system lies beyond the realms of simple efficiency, and is more related to the requirements of the uses and users listed previously. For example, development of the template presented in Figure 1 can be undertaken by anyone with a text editor, and requires no particular spatial data management software, such as a GIS. In this way, establishment of a template for an environmental management model is, and is seen to be, readily accessible to all stakeholders. Similarly, the data shown in Figure 2 could be stored more efficiently, but this would potentially be at the cost of simplicity and comprehension by all involved parties. Also, anyone can open this file and quickly find (and edit) a value for a particular cell. Of course, within the modelling shell there are generic tools that allow the easy editing of spatial data once the template has been established, but the principles of simplicity and accessibility are essential for reducing the divisions between "experts" and stakeholders. lt is also relatively simple to convert other formats (such as from a GIS) to this format if need ...

Context 19

... ---. .. .. .. . �... .. .. .._ __ Figure 1. Location ofthe study area near Wagga Wagga, NSW. Seasonally wet and waterlogged soil mapping from the Kyeamba catchment (shaded) was used to obtain the threshold for classifying the UPNESS index into runoff sinks or no sinks, and this training set was used for the remainder of the study ...

Context 20

... fuzzy UPNESS index (0.0 -1.0) was cut into dry or wet classes at a value determined from the training area (the Kyeamba catchment of approx 1000 km2, see Fig. 1), that is, where the area ofUPNESS best matched the seasonally wet to waterlogged and sodic to saline soils from a 1 :25,000 scale soil landscape map (Chen and McKane, 1996, Summ erell et al., in prep ). Two classes are produced, the position in the landscape where these soils are expected to occur and where they are not. This wet soil class for each major catchment was expressed as the % ofthe area ofthe catchment. This we call UPNESS% and is interpreted as a reflection of sub-surface water accumulation, resulting in salt mobilisation. High UPNESS% values are inferred to mean high salinity ...

Context 21

... the third step in our indicator selection is to ensure that we are adequately representing the issues of concem in each catchment. In order to confirm our understanding of catchment condition issues in all parts of Australia's intensive land-use zone, we have conducted a survey of all catchment management coordinators in the country, asking for their summary of issues and issue severity in their catchments. W e also use these data to determine whether there are regional differences in issues which need to be tak:en into account in our indicator selection or in setting threshold values. lt would be unfair to award a good report card to a catchment for an ! ATM:lSPHE RE 1 Figure 1. A catchment systems model Figure 2. The A++ interface issue which cannot exist in its boundaries. If we determine that regional differences in catchment issues are significant, regionally-specific indicator sets can easily be incorporated into our catchment condition software system, ...

Context 22

... figure shows that the starting point of the catchment system is the materials in the catchment. These may be in-situ, such as soil, or imported, such as sediment input from upstream or fertiliser added to crops. The materials can then be modified, either through in-situ ecosystem scale processes like biogeochemical cycling, or catchment cascading processes like erosion or nutrient fluxes. Cascading processes require a transport mechanism ( analogous to power) and a conduit for transport (analogous to resistance). In the case of erosion, the mechanism could be water volume and velocity, and the conduit a gully or channel. The quality of water exported from a catchment should be an indicator of all these processes. As shown on the left of Figure 1, human action can modify any one of the boxes in the system, provoking changes in catchment ...

Context 23

... presented mesh generation procedure is applied to a part of the Delaware Estuary shown in figure 1. A sub-region of the estuary is shown in more detail in figure 2. In land regions complementary points are added to the set of measurement points to av oid numerical problems in the approximation process. The nodes of the quadrilateral finite elements are created on the iso-parametric lines of the u and v parameters. This leads to elements outside of the estuary region. As a result, these surplus elements are deleted. To achieve a smooth boundary polygon the transitional elements between water and land are cut and the remaining parts of the elements are ...

Context 24

... ). The number associated with these taxa indicates the level of certainty of the observation, the closer it is to zero the greater the degree of certainty. The theory underlying this model was first published in Walley et al., (1992), but is more fully described in . The window appears on the left of the screen (i.e. in place of the SOM window in Figure 1) in conjunction with biological sample data on the right. The season displayed is synchronised with the biological data and can be switched between spring and autumn by clicking the appropriate sample number on the biological data ...

Context 25

... ural Network (S OM) Classifications nl The window that is displayed by this button is shown on the left side of Figure 1. lt is � always displayed in conjunction with the biological data, but unlike the Bayesian classification, the SOM classification is an annual one, based upon the combined spring and autumn samples. Thus, switching between spring and auturnn sample numbers in the biological data window does not change the information given in the SOM window. This classification system is based on a square map, divided by a set of 20x20 grid lines (not shown on Figure 1 ). Each grid intersection represents a particular type of pattem in the combined biological and environmental data. The pattems are allocated to the grid intersection in an ordered way, such that similar pattems are close neighbours and dissimilar pattems are well separated. A fe ature map of any one of 96 attributes of the data (e.g. number of families) can be plotted on the map space, simply by selecting the attribute from the two scrollable lists (one containing the 76 taxa, the other the 20 environmental/river quality variables). The grid intersection (i.e. particular pattern) to which the current sample has been classified is indicated by a black circle on the map. A SOM based GQA river quality classification can be derived by selecting GQA-Neural Net and noting the class corresponding to the position of the black circle. The feature maps can also be used to check if the taxa found in a sample were in keeping with those predicted by the SOM model. A detailed description of this model and its underlying theory can be found in Walley et al., (1998 & 2000). A SOM viewer program that permits the comparison of any two SOM feature maps is available through the CIES web site (CIES, ...

Context 26

... ural Network (S OM) Classifications nl The window that is displayed by this button is shown on the left side of Figure 1. lt is � always displayed in conjunction with the biological data, but unlike the Bayesian classification, the SOM classification is an annual one, based upon the combined spring and autumn samples. Thus, switching between spring and auturnn sample numbers in the biological data window does not change the information given in the SOM window. This classification system is based on a square map, divided by a set of 20x20 grid lines (not shown on Figure 1 ). Each grid intersection represents a particular type of pattem in the combined biological and environmental data. The pattems are allocated to the grid intersection in an ordered way, such that similar pattems are close neighbours and dissimilar pattems are well separated. A fe ature map of any one of 96 attributes of the data (e.g. number of families) can be plotted on the map space, simply by selecting the attribute from the two scrollable lists (one containing the 76 taxa, the other the 20 environmental/river quality variables). The grid intersection (i.e. particular pattern) to which the current sample has been classified is indicated by a black circle on the map. A SOM based GQA river quality classification can be derived by selecting GQA-Neural Net and noting the class corresponding to the position of the black circle. The feature maps can also be used to check if the taxa found in a sample were in keeping with those predicted by the SOM model. A detailed description of this model and its underlying theory can be found in Walley et al., (1998 & 2000). A SOM viewer program that permits the comparison of any two SOM feature maps is available through the CIES web site (CIES, ...

Context 27

... system was designed for use by practising river ecologists and river quality managers having a relatively low level of computer literacy. The interface was designed to be user-friendly but fairly flexible, allowing the user to view different sets of related information alongside each other. This was achieved by dividing the basic working screen into five parts, as illustrated by the typical screen shown in Figure 1. Along the top of the screen is the button bar that appears on all screens and enables the user to select a different site or different type of information. Once a site has been selected, a status bar just below the button bar gives details of the site's location. Along the bottom of the screen an instruction bar suggests further actions to the user. The remainder of the screen is used for the display of information windows, normally one on the left and one on the right, as illustrated in this example, but in one particular case (i.e. chemical data), a single window occupies the füll width. ...

Context 28

... reasoning involves assessing the degree of belief in a certain proposition, using the axioms of probability for combination of beliefs. A graphical probability model involves representing the key variates in a physical system as a network, with the variates as nodes and the relationships as edges. Only the "significant" relationships are represented and the edge is directed from a variate representing a cause to one representing an effect. For more details, the interested reader is encouraged to consult [Russell and Norvig] . Key elements in this representation that make it useful are 1. sparsity (only significant causes) and adequacy of the vocabulary to represent the system being modelled. As described later, we also can incorporate decisions and utility measures to estimate the value of actions, as an intrinsic part of the model. lt is therefore possible to illustrate the probabilistic interaction of variables using a graph. Here we illustrate probabilistic models using Bayesian Networks [Heckerman] . The parameters of such models consist of conditional probability distributions that are associated with the arrows of in the graph and unconditional probability distributions associated with variables that have no incoming arrows. Such representations can also aid the modelling of decision processes. This is accomplished by introducing additional decision variables as nodes in the graph. We demonstrate the application of utility theory in such decision processes. Specifically, we examine modelling the cumulative effect of several decisions within a process model for sediment and pesticide delivery to receiving waters into which a watershed drains. Suppose, for example, we are modelling erosion and sediment transport in an agricultural drainage network. Within each homogeneous plot, a deterministic model ( e.g. GAMES, [Rudra et sl.]) predicts the generation of sediment within the plot, and a spatial network illustrates the pathway of sediment transported from above and to the plot below in elevation. For this discussion, we ignore the model mechanics. Then the sediment S(D) delivered downstream (downslope) is related to the sediment delivered from the immediate upstream plots LS(U) and sediment generated intemally, S(G) by S(D) = ex( LS(U) + S(G)). The factor ex is called the sediment cell delivery ratio (CDR) . The CDR is generated by the factors considered in the development of the GAMES model: slope, aspect ratio with respect to steepest descent, roughness, cropping practices, soil type, etc. (Figure 1 ). Within this plot or cell, a decision can be made about land use or cropping practice that will have an effect on the delivery ...

Context 29

... main RPDS screen is shown in Figure 1. lt comprises: a Title bar with the RPDS icon and title, plus minimisation and close buttons; a Ma in toolbar; a Feature map I Classification panel; a Status panel; Ta bbed fu nction panels. ...

Context 30

... 1 presents a schematic of the modelling approach, which consists of linking the watershed and hydrodynamic model results with the lake water quality box model. The integration step is implemented by using the RAISON Decision Support System (Lam et al. , 1994). RAISON is a generic system that offers input facilities to accept various types of data and maps which are then stored its own intemal fully-linked database, map subsystem and graphic components. In this paper, we will discuss only the linkage between the land-based hydrological model and the lake water quality box model using the results from the hydrodynamic runs. Figure 1. Modeling approach for the Lake Malawi case study There are three ways to incorporate models into RAISON. Tue first way is to run the model separately from RAISON and feed the model results into RAISON as input. This is the method used to incorporate the results from the hydrodynamic model (DELFT3D) in this proj ect. The computed results such as transport and temperature are only required by the lake water quality box model. The second way to incorporate a model is to have it linked to RAISON, e.g., making use of the RAISON databases, maps and graphical facilities. The example for this method is the watershed model (AGNP S) which is an external executable program but with the input and output data linked to the database, map and graphical components in RAISON. This linkage to RAISON enables the model to communicate with the water quality box model directly in order to generate results iteratively for scenario testing of land use effects on lake water quality. The third way to use a model in RAISON is to rewrite the code for the model in a programm ing language ( e.g. Visual Basic) which makes it possible to link the model directly to RAISON. The water quality box model (NWRI Box Model) is such an example, because the original code for this model needs to be modified to accept input from the AGNPS and DELFT3D models. ...

Context 31

... schematic representation of the NZDIS system architecture is shown in Figure 1. All of the solid, directed lines shown in the figure represent agent messages expressed in the Foundation for Intelligent Physical Agents Agent Communication Language (FIPA ACL) [FIPA, 2000]. Not shown in this figure is the System Facilitator which maintains a directory of all agents in the system and with which each agent must register when it is incorporated into the system. A query from the user agent is passed to the Query Processing Subsystem, which endeavors to query multiple data sources in order to satisfy the user's query. In the following section, the operations of the Query Processing Subsystem are described in greater detail. A query from the user agent is passed to the Query Processing Subsystem, which endeavors to query multiple data sources in order to satisfy the user's query. In the following section, the operations of the Query Processing Subsystem are described in greater ...

Context 32

... this context it should be mentioned that eco-balancing is already since years a classical tool of ecological planning, but the term was successfully (re)introduced by business management because of the origin of balances in economy. Hence Life Cycle Assessments (LCA), business managers much more often carry out process-or business-eco-balance than ecologist. If there are classical ecological planning units and procedures ( e.g. regions, landscapes, watersheds, and related tools like land consolidation) then we should consider the already developed approaches as weil as the high importance of having "state of the art" ecological indicators. These indicators (listed in Fig. 1) and the further quantification of their status are the basis of the quality of every ecological balancing, besides the tools provided by environmental ...

Context 33

... technical system is rather elegant and powerftll, and especially in combination with scenarios, it started to get accepted within the administration. Nevertheless, only a few people in the district administration are able to run the software -although it became a standard one in all districts in several states of Germany -, and hence this aspect is still the most relevant restriction of the use ofthe system. In addition, many other feed backs on the system from our target groups where like: "technical and scientifically good, but too complicated/oversized for policy and administration .... " What all of them appreciated very much, was a graphical illustration of the outcome ofthe environmental balancing in a very aggregated way (cf. Ten Brink 1991), as shown in Fig. 1 (cf. WG ...

Context 34

... adequ ate farmin g Fig. 1: Amoeba with Environmental Indicators and action fields "Regional Eco-balances in the district of Pfaffenhofen". Tue darker the gray color, the less good is the environmental status of an indicator in the action filed, ...

Context 35

... study area consists of a number of catchments in the mid-Murrum bidgee catchment near Wagga Wagga in southeastem Australia (Fig. 1). The application ofFLAG is the first step to direct more intensive field sampling. The maps produced will be evaluated by state agencies to assess the usefulness ofFLAG as an additional tool in research and extension ...

Context 36

... coordination of agricultural and wetland releases with reservoir releases of good quality snow-melt water on the east-side of the San Joaquin Basin has been suggested as a means of improving San Joaquin River water quality for all beneficial uses Karkoski, Quinn and Grober. 1995;Quinri et al., 1997;Quinn and Karkoski, 1998). In a previous technical paper Quinn (1999) described the results of a demonstration project of real-time monitoring and management of agricultural drainage and east-side reservoir releases which produces weekly forecasts of San Joaquin River assimilative capacity for salt ( Figure 1 ). After four years of operation, agencies such as the US Bureau of Reclamation are starting to use the web-posted forecasts of river water quality and local water districts such as West Stanislaus Irri gation District are utilizing monitoring data and forecasts of the water quality of their riparian diversions from the main-stem ofthe ...