A construction method of visual conceptual scenario for hydrological conceptual modeling

https://doi.org/10.1016/j.envsoft.2021.105190Get rights and content

Highlights

  • Visual conceptual scenario is a medium through which modeling cognition is expressed.

  • Visual conceptual scenario can facilitate modeler's understanding of hydrological system.

  • Modelers can exchange modeling ideas during the construction of conceptual scenario.

Abstract

With the increasing complexity of hydrological systems, hydrological modeling by modelers from different research areas has been regarded as an effective way to solve complex hydrological issues. As the first step of hydrological modeling, conceptual modeling plays an important role in supporting modeling idea communication among interdisciplinary modelers. Currently, conceptual modeling methods usually show the modeling ideas by using the block-based diagrams. However, further research need to be explored to express the spatial and temporal distribution of these modeling elements and their dynamic interaction relationships, thus promoting the communication of modeling ideas and reaching a common understanding of the modeling system among modelers. Visual conceptual scenario is the production of hydrological conceptual modeling and can be used to express modelers’ cognition of modeling system. This article proposes a construction method of visual conceptual scenario. The conceptual component that represents the modeling elements, the rules that constrain the scenario construction, and a dynamic interaction method that supports the visualization of dynamic hydrological process are designed. Finally, a study case of identifying the impact of energy base water project on the groundwater is designed to illustrate the feasibility and practicability of the proposed construction method of visual conceptual scenario.

Introduction

Hydrological modeling plays an important role in explaining complex hydrological phenomena and exploring the relationships among hydrological processes, and it can support water resource management and decision-making (Demeritt and Wainwright, 2005; Granell et al., 2010; Granell et al., 2013). To solve hydrological problems, researchers from all over the world have conducted a series of hydrological modeling studies and produced numerous hydrological models, such as the Soil and Water Assessment Tool (SWAT) (Douglas-Mankin et al., 2010; Koo et al., 2020a, Koo et al., 2020b), the Storm Water Management Model (SWMM) (Gironás et al., 2010), the Variable Infiltration Capacity (VIC) Macroscale Hydrologic Model (Liang et al., 1994), and the Soil and Water Integrated Model (SWIM) (Krysanova et al., 2000). However, when considering a complex hydrological system, there is a growing understanding that a single model from a specific research area cannot sufficiently represent all the details of the modeling system (Argent, 2004; Voinov and Cerco, 2010). Coupling multidomain models and conducting hydrological integrated modeling to solve complex hydrological problems have become a trending topic (Granell et al., 2013; Belete et al., 2017).

The long-term practice of hydrological (integrated) modeling has shown that the mutual understanding among modelers may be the first step in hydrological modeling (Refsgaard and Henriksen, 2004). For example, in order to explore the impacts of groundwater exploitation on watershed ecological environment, a groundwater quantity model, a groundwater water quality model, a vegetation growth model and a surface runoff module (from a basin hydrological model) may be involved in the modeling system. However, each type of these models has numerous individual computational models (e.g., there are many basin hydrological models, such as SWAT, TOPMODEL, HBV, etc.). Each individual model was built on specific modeling principles and each has its own application scope. In this way, how to choose the right individual models or modules to conduct integrated modeling has become an important modeling task for hydrological problem solving. Moreover, modelers with a background in hydrology may not understand the modeling principles of vegetation growth models, which makes it difficult to communicate modeling ideas. Therefore, it is necessary to make clear the modeling elements and their interaction relationships before conducting hydrological integrated modeling, so that modelers with different backgrounds can reach a common cognition of the modeling system and then guide the construction of hydrological computational model. In this article, the process by which modelers with different backgrounds express and communicate their understanding of the modeling system is called hydrological conceptual modeling.

The Open Geographic Modeling and Simulation (OpenGMS, http://geomodeling.njnu.edu.cn/) provides a modeling resource-sharing environment (such as model, data, and computational resources), which aims to support collaborative modeling and complex environmental problem solving (Wen et al., 2006; Chen et al., 2011; Yue et al., 2015; Chen et al., 2016; Yue et al., 2016; Wen et al., 2017; Wang et al., 2018; Chen et al., 2019; Zhang et al., 2019; Chen et al., 2020, Chen et al., 2021; Wang et al., 2020). The OpenGMS divides the integrated modeling process into three steps: conceptual modeling, logical modeling and computational modeling (Chen et al., 2020). In this article, the hydrological conceptual modeling is conducted within the context of integrated modeling in the OpenGMS framework, which focuses on the expression of the modelers’ cognition of the modeling systems.

Recently, a series of related studies on hydrological conceptual modeling has been carried out. The European Community (EC) funded project, Harmonising Quality Assurance (HarmoniQuA), attempts to offer computer-based guidance for supporting the full modeling process, which covers all water management domains, different types of users, different types of modeling purposes and different levels of modeling complexity (Scholten et al., 2004; Kassahun et al., 2005; Refsgaard et al., 2005; Højberg et al., 2007; Scholten and Refsgaard, 2010; Black et al., 2014). Rushton (2004) points out that a conceptual model can illustrate the mechanisms of the hydrological system being modeled and can allow other modelers to assess current thinking and to provide further insights. Jakeman et al. (2006) summarize ten iterative steps in the development and evaluation of environmental models, which incorporate the idea that conceptualization can help in thinking and finding underlying information about the system.

The capabilities of hydrological conceptual modeling can be summarized as follows:

  • (1)

    At the beginning stage of hydrological modeling, hydrological conceptual modeling can be regarded as a process in which the modeling problems are analyzed, the modeling objectives are specified, and the dynamic mechanisms of the hydrological system is clarified (Rushton, 2004; Jakeman et al., 2006; Yazici and Akkaya, 2000). For example, in studying the influence of vegetation on hydrological processes in an area according to specific application scenarios, modelers may first consider the following issues, among others: (a) how many processes are involved from the beginning of precipitation to the formation of runoff; (b) how vegetation types influence these processes; (c) what the relationships among these processes are; and (d) how vegetation affects the spatial and temporal distributions of rainfall, runoff and evapotranspiration. Such points need to be clarified during the conceptual modeling phase, which can help modelers to better understand the modeling problem. Although the understanding of modeling problems may be simple and incomplete at the beginning of conceptual modeling, it provides the possibility for further communication with other researchers.

  • (2)

    In the hydrological modeling process, hydrological conceptual modeling can support the transfer of modeling ideas among modelers to achieve a common understanding of the modeling problems. Generally, when faced with comprehensive and interdisciplinary hydrological modeling problems, modelers often have different viewpoints due to their different research backgrounds. For example, modelers with experience studying surface hydrological processes may be accustomed to applying infiltration excess runoff, saturation excess runoff, and partial infiltration excess runoff to describe the mechanisms responsible for runoff generation. However, modelers with groundwater research backgrounds usually emphasize the contribution of the underground seepage system to surface runoff generation (Rubin, 2003). In this respect, conceptual modeling can provide the capacity to express different ideas and can help different modelers understand each other.

  • (3)

    Hydrological conceptual models embody the principles of hydrological models, which can facilitate the sharing and application of hydrological models in the later stage of hydrological modeling (Clark et al., 2015). With the increasing complexity of hydrological models, the hydrological modules involved and the relationships among these modules become more complex (Argent, 2004; Belete et al., 2017; Yue et al., 2019). It is difficult for users, especially interdisciplinary researchers, to understand these comprehensive hydrological models, which limits the application of the models. A conceptual model is the product of conceptual modeling, which is very helpful for users who aim to understand a hydrological model quickly at the conceptual level. By comparing conceptual models, users can quickly choose appropriate models for their applications from a large number of candidate models, rather than reading the explanatory documents for each model in turn.

In short, hydrological conceptual modeling runs through the whole process of hydrological modeling and plays an important role in promoting the transmission and exchange of modeling ideas. With the development of hydrological conceptual modeling research, three main types of hydrological conceptual modeling methods have been developed.

  • (1)

    Conceptual sketching. Sketching is a powerful way to explore modeling ideas (Rushton, 2004; Jakeman et al., 2006). Modelers usually drew a conceptual sketch directly on paper to express and communicate their modeling ideas. This approach is very effective within a specific research team due to its simplicity and convenience. However, the conceptual modeling results are difficult to share with others and to reuse because of the informal and individual conceptual modeling approach (Wand and Weber, 2002). This makes it difficult for modelers with different backgrounds to communicate with each other.

  • (2)

    Graph-based conceptual modeling. To formalize modeling ideas, the influence diagram, block diagram, or bond graph is introduced by modelers to express the modeling elements (e.g., geographic entities and hydrological processes) and their relationships, which can aid thinking and modeling (Gawthrop and Smith, 1996; Muetzelfeldt and Massheder, 2003; Jakeman et al., 2006; Leonard et al., 2017; Lukyanenko et al., 2017). Examples include GBMS/SM (Chari and Sen, 1998), GME (Generic Modeling Environment) (Davis, 2003) and GeoVISTA Studio (Gahegan et al., 2002). These methods usually use block primitives and lines to express specific elements and their relationships. Although this approach has advantages in expressing logical reasoning, it has limitations in expressing the details of the modeling thought process, such as the spatial and temporal relationships among different modeling elements. For example, canopy interception and surface infiltration are different steps in surface hydrological processes, and they occur at different times and places. However, in a graph-based modeling approach, the lines between block primitives cannot express such spatial and temporal relationships among different modeling elements.

  • (3)

    Icon-based conceptual modeling. This approach builds a visual conceptual modeling scenario from the perspective of the modelers' ideas about the hydrological system (Chen et al., 2011). In this approach, different modeling elements are represented by different icons, and these icons are “organized” by the layout rules to build a constrained and cognitive conceptual scenario. Such a scenario provides a flexible way for the modelers to communicate modeling ideas intuitively. However, due to the limitation of two-dimensional (2D) and static icons, it is difficult to express the three-dimensional (3D) time-space relationships of the modeling elements (e.g., nested relationships). Moreover, the hydrological processes are dynamic, which is difficult to express in such a static and 2D scenario.

In conclusion, although these conceptual modeling approaches provide effective tools to express modeling elements and their relationships, they have limitations in expressing the spatial and temporal distribution of modeling elements and related dynamic hydrological processes. Recently, serious games, virtual environment and geographic scenario are increasingly used as tools to help stakeholders participate in the research of environmental issues (Chen et al., 2013; Lin et al., 2013a, Lin et al., 2013b; Lin and Chen, 2015; Craven et al., 2017; Chen and Lin, 2018; Khoury et al., 2018; Smith and Lohani, 2018; Lü et al., 2018, Lü et al., 2019; Den Haan et al., 2020; Li et al., 2021a, Li et al., 2021b; Yang et al., 2021). Inspired by these studies, visual conceptual scenario is proposed to express modelers’ cognition of modeling system. The visual conceptual scenario is the production of hydrological conceptual modeling, which provides a visual way for cross-domain modelers to express their modeling ideas at conceptual level, thus facilitating the communication among these modelers.

The conceptual scenario consists of numbers of conceptual components, which represent geographic entities (e.g., rivers, vegetation, and soil) involved in modeling system and related hydrological processes (e.g., surface runoff, precipitation, and evaporation). The cognitive differences among different modelers can be addressed by designing different conceptual components. Namely, modelers can express their knowledge of the geographic entities or hydrological processes by designing corresponding conceptual components. To construct the visual conceptual scenario, these conceptual components need to be “organized” according specific rules. Therefore, a series of rules (e.g., context rules, layout rules, and semantic rules) that reflect modelers’ cognition of modeling system are designed to constrain the construction of the scenario. Meanwhile, the spatial and temporal distribution of conceptual components can be constrained by these rules. For example, the layout rules can dictate that “cloud” must be above the “ground”, and the semantic rules can specify that “rain” comes before “surface runoff”. Since the dynamics of hydrological processes can be expressed by the interaction among conceptual components, a method that represents such interaction relationships at conceptual level is proposed. In this way, the temporal and spatial distribution of modeling elements is constrained by specific rules, and the dynamic hydrological process of the modeling system is driven by dynamic interaction method proposed in this article. During the hydrological conceptual modeling, modelers can exchange modeling ideas through the visual and dynamic conceptual scenario, rather than through abstract block primitives.

The remainder of this article is structured as follows: The construction framework of visual conceptual scenario is introduced in Section 2. Section 3 introduces the methodology of the construction of conceptual scenario. In Section 4, a case study involving two conceptual scenarios is developed. Finally, conclusions and directions for future work are presented in Section 5.

Section snippets

The construction framework of the visual conceptual scenario

Fig. 1 shows the construction framework of the visual conceptual scenario. When considering a hydrological system, modelers may work together to communicate their perceptions of how to model the system. The following aspects, among others, will be considered in understanding a hydrological system: the geographic entities involved in the system, the space-time distribution pattern of these entities, how many hydrological processes take place in the system, and the relationships among these

The design of the conceptual components

A conceptual component is the comprehensive representation of a geographic entity and is the basic building block of a conceptual scenario. The basic idea of designing the conceptual component is to use the visualization and scripting techniques provided by the Unity engine (Unity, 2021) to express the attribute and behavior characteristics of the geographic entity. Fig. 2 shows the detailed design method of a conceptual component.

To identify a geographic entity, the following aspects will be

Study case

Groundwater is an important source of energy production in arid regions, so it is necessary to study the impact of enhanced human pressure on groundwater (Awawdeh et al., 2020). In this section, the conceptual modeling is conducted under the background of identifying the impact of energy base water project on the groundwater in the Subei Lake basin, Ordos, northwestern China (Hou et al., 2008; Liu et al., 2016). Two conceptual scenarios that reflect the key hydrological processes in the Subei

Conclusions and future work

This article proposes to construct visual conceptual scenario to facilitate the exchange of modeling ideas among modelers during hydrological conceptual modeling. The primary originality of this study is its development of a construction scheme that implements the construction of conceptual scenario, which can support modelers with different backgrounds in working together. First, modelers can express their conception of geographic entities by designing the conceptual components, with which the

Data availability statement

The constructed dynamic conceptual scenarios in the study case section can be accessed at http://geomodeling.njnu.edu.cn/modelItem/57a498ec-42e2-4f43-8a08-ce92ef752e48.

Editorial Conflict of Interest Statement

Given their roles as Environmental Modeling & Software Editor, Min Chen were not involved in the peer-review of this article and have no access to information regarding its peer-review. Full responsibility for the editorial process for this article was delegated to journal Chief Editor, Daniel P. Ames.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We appreciate the detailed suggestions from the anonymous reviewers. We also express heartfelt thanks to the other members of the OpenGMS team. This work was supported by the Natural Science Foundation of China (Grant Nos. 41930648, 41631175, 41871285 and U1811464).

References (67)

  • C. Granell et al.

    Service-oriented applications for environmental models: reusable geospatial services

    Environ. Model. Software

    (2010)
  • C. Granell et al.

    Seeing the forest through the trees: a review of integrated environmental modelling tools

    Comput. Environ. Urban Syst.

    (2013)
  • A.J. Jakeman et al.

    Ten iterative steps in development and evaluation of environmental models

    Environ. Model. Software

    (2006)
  • H. Koo et al.

    A global sensitivity analysis approach for identifying critical sources of uncertainty in non-identifiable spatially distributed environmental models: a holistic analysis applied to SWAT for input datasets and model parameters

    Environ. Model. Software

    (2020)
  • H. Koo et al.

    Position paper: sensitivity analysis of spatially distributed environmental models-a pragmatic framework for the exploration of uncertainty sources

    Environ. Model. Software

    (2020)
  • W. Li et al.

    A rapid 3D reproduction system of dam-break floods constrained by post-disaster information

    Environ. Model. Software

    (2021)
  • H. Lin et al.

    Virtual geographic environments (VGEs): a new generation of geographic analysis tool

    Earth Sci. Rev.

    (2013)
  • R. Muetzelfeldt et al.

    The Simile visual modelling environment

    Eur. J. Agron.

    (2003)
  • J.C. Refsgaard et al.

    Modelling guidelines––terminology and guiding principles

    Adv. Water Resour.

    (2004)
  • J.C. Refsgaard et al.

    Quality assurance in model based water management–review of existing practice and outline of new approaches

    Environ. Model. Software

    (2005)
  • A. Voinov et al.

    Model integration and the role of data

    Environ. Model. Software

    (2010)
  • S. Yue et al.

    Service-oriented model-encapsulation strategy for sharing and integrating heterogeneous geo-analysis models in an open web environment

    ISPRS-J. Photogramm. Remote Sens.

    (2016)
  • F. Zhang et al.

    Design and development of a service-oriented wrapper system for sharing and reusing distributed geoanalysis models on the web

    Environ. Model. Software

    (2019)
  • Unity
  • M. Awawdeh et al.

    Groundwater vulnerability assessment using modified SINTACS model in Wadi Shueib, Jordan

    Spatial Sci.

    (2020)
  • Blender
  • M. Chen et al.

    Virtual geographic environments (VGEs): originating from or beyond virtual reality (VR)?

    INT J DIGIT EARTH

    (2018)
  • M. Chen et al.

    A visualization method for geographic conceptual modelling

    Spatial Sci.

    (2011)
  • M. Chen et al.

    Real geographic scenario based virtual social environment: integrate geography with social research

    Environ. Plann. B

    (2013)
  • M. Chen et al.

    Position paper: open web-distributed integrated geographic modelling and simulation to enable broader participation and applications

    Earth Sci. Rev.

    (2020)
  • M. Chen et al.

    Geographic modeling and simulation systems for geographic research in the new era: some thoughts on their development and construction

    China Earth Sci.

    (2021)
  • M.P. Clark et al.

    A unified approach for process‐based hydrologic modeling: 1. Modeling concept

    Water Resour. Res.

    (2015)
  • Cited by (6)

    View full text