Comprehensive scenario management of sustainable spatial planning and urban water services

In the INIS joint research project ‘SinOptiKom – Cross-sectoral optimisation of transformation processes in municipal infrastructures in rural areas', funded by the German Federal Ministry of Education and Research, a software-based decision support and optimisation model for long-term transformations of urban water infrastructures is developed. The overall structure of the optimisation system consists of a pre-processing tool with a database and a scenario-manager, a mathematical optimisation model and an interpretation tool for the visualisation of results. The optimisation model determines adaptation strategies for existing water supply and wastewater systems, based on the implemented objective functions and shows their implementation over a specific period of time (Baron et al. 2015). The model is tested in two rural municipalities in the southwest of Germany.

Special attention has to be paid to the increasing uncertainty of future constraints and impact factors (Pearson et al. 2010; Stojanović et al. 2014). A realistic prognosis for different dynamic drivers including their interdependencies has to be examined to produce representative solutions within the model. Scenarios are required to generate quantitative dynamic input and evaluation data for the period of investigation. A characteristic of future water systems are the long capital commitments of central water infrastructures for time periods of 50 to 80 years and challenges like demographic, climate and socioeconomic developments. The development of consistent scenarios requires the cooperation of water management and town planning.

Different scenario management techniques are effectively applied in the business world and the scenario approach is increasingly used in the context of spatial planning (Andrienko et al. 2003; Krawczyk & Ratcliffe 2006; Myers & Kitsuse 2009; Stojanović et al. 2014) and the development of future urban water infrastructures (Lienert et al. 2006; Nowack & Günther 2009). One approach is the generation of future scenarios with the engagement of stakeholders (e.g. Truffer et al. 2010; Lienert et al. 2014). A multi-stage procedure with stakeholder interviews results in the definition of mostly three to four scenarios. This approach supports infrastructural planning processes but the selection of stakeholders is very crucial even with a systematic approach. Stakeholders have a subjective point of view about current conditions and a likely future. In general the development of three or four scenarios or storylines is highly dependent on the quality of assumptions and projections (Urich & Rauch 2014). In other projects only a limited number of drivers are included in scenarios, for example a combination of population growth and spatial development scenarios (e.g. Mikovits et al. 2014). For policy examples complex approaches based on participatory, computer-assisted approaches using algorithms are explored (e.g. Bryant & Lempert 2010).

This paper presents the development of a scenario management approach for the generation of scenarios by combining different drivers. It is designed for the integration of a large range of important drivers of future water infrastructures with the aim of generating an optimised transformation strategy of urban water systems. Since several different factors influence transformation strategies simulations with only three to four scenarios cannot take all important developments of each driver into account. Simulations with more scenarios are important for generating robust transformation strategies and exploring the underlying uncertainties. Stakeholders are involved in the identification of drivers of water infrastructures. The user of the model, an expert, then arranges the drivers to scenarios. In the project several rural municipalities are considered and therefore the transferability of the method is important.

In this paper, methods for the prediction of each identified driver are described and for some drivers their development in the context of a model municipality is illustrated. The terms scenario, driver and influencing factor are subject matter. In this paper a scenario is defined as a consistent combination of prognoses of different drivers. The development of each driver is affected by divers influencing factors.

The scenarios are developed on three decision and modelling levels: macrolevel, mesolevel and microlevel. On the macrolevel, the development of the settlement in general, the institutional and legal framework, energy prices and effects of climate change are considered for an association of municipalities. In Germany, an association of municipalities is a local authority which consists of neighbouring municipalities. The fee system is also designed on the macrolevel. For every municipality (mesolevel) the development of population, settlement and water demand is explored. The microlevel considers units of municipalities, e.g. street sections or neighbourhoods. On this level the population and settlement development is presented in detailed scenarios.

The different methods and techniques of predicting each identified driver are presented below. The future development of the drivers ‘demographic and spatial development’ as well as ‘water demand’ is illustrated for a rural model village within the project's case study region (mesolevel, microlevel). The model village has about 800 inhabitants with a specific water demand of 115 L/(C·d) at present. A central combined sewer system is installed and the wastewater is treated in a wastewater treatment plant together with the wastewater of three neighbouring villages. The future developments of the other drivers are illustrated for the rural case study region in the project (macrolevel).

Furthermore, spatial concepts regarding the association of municipalities' structural state and development on the macrolevel are concretised on the mesolevel in different variants, in order to synthesise the development of the single municipalities. The final step towards the integrated municipal spatial planning scenarios is their implementation into a geographic information system (GIS). Here, detailed data as the status quo of demographic structure, the inhabitants' milieu, building occupancy or condition is geo-referenced. Depending on the demographic prognoses and factors as the economic development or strategic decisions on infrastructures and public services, the municipal development is synthesised. The synthetic spatial development refers to the spatial development concept in order to assess changes in land use, building activity or spatial transformation, e.g. the revitalisation of municipal centres.

Different methods can be used for the prediction of the water demand, e.g. trend analysis or econometric models (cf. Schleich & Hillenbrand 2009). As the water demand in Germany has been declining since the mid-1990s, a trend analysis could result in unrealistic small water demands for a time horizon of 50 years. In the presented project the scenario technique (Dönitz 2009; Nowack & Günther 2009) is used to identify the future development of water demand. This approach allows an integration of important influencing factors and their uncertainties (Lienert et al. 2006). Previously presented developments of population and settlement are integrated in the approach.

The total water demand comprises the water demand of households, commerce and industry, public institutions, own needs of water supply companies as well as losses in the water supply network. For the determination of the total water demand, the above presented developments of inhabitants and settlement structure are integrated.

The future development of energy prices is very complex. Therefore, the project focuses on electricity prices. The development of electricity prices depends on national funding programs and legal developments, the development of the share of renewable energies, or formation of power grids. Since the presented project is not focusing primarily on energy, future development possibilities are derived from the literature (e.g. SRU 2011; UBA 2012; Öko-Institut 2015) and from results of the funding program ERWAS (Future-oriented Technologies and Concepts for an Energy-efficient and Resource-saving Water Management) of the German Federal Ministry of Education and Research (Table 1).

The investment and operational costs of water infrastructures, especially of new and innovative technologies, might change in the future. Therefore, two alternatives of cost development are considered. In the first version present prices of technologies of water supply, wastewater discharge and treatment as well as rainwater management are assigned with a medium rate of price increase of 3%.

The second version includes larger variations of current investment and operational costs. Investment costs can be divided into costs of well-established and comprehensively installed facilities like conduits and in new and innovative technologies like greywater treatment. Costs for well-established technologies will similarly remain in the future. However, expensive costs of new technologies might change significantly with wider dissemination and use. In addition, the implementation of new technologies benefits with a price reduction. Increasing personnel costs and maintenance efforts of facilities, e.g. flushing of conduits might affect the operational costs. For the quantification of cost developments of new technologies the concept of ‘production learning curves' is applied. The concept indicates that with every duplication of the accumulated production volume the total costs for a product decreases with a constant percentage. This reduction of costs is called learning rate. The concept is known from other sectors (e.g. chemical industry) and there are only few applications in the water sector (e.g. Zhou & Tol 2005; Hillenbrand 2009). Learning curves are designed for the implemented new technologies in the project and in Table 2 possible ranges of cost reduction for three examples are presented. The values were derived for Germany, but they might differ in other countries.

This driver allows the inclusion of future modifications of legal limits or amendments. These are increasing discharge limits of nutrients for wastewater, the introduction of thresholds for micropollutants or the introduction of rates for recycling of phosphorus. Another important change will be the withdrawal of the use of sewage sludge in agriculture in Germany. Currently, a large part of sewage sludge in rural areas is used locally as fertiliser in agriculture. Alternatives for the utilisation, like incineration, are integrated and costs for storage and transportation to larger incineration plants are considered. Moreover other methods for the recirculation of nutrients into the soil are examined.

Climate change will cause changes in rain patterns and temperature. In the project the impacts of climate change on heavy rainfall events are integrated. For rainwater simulations the local precipitation intensity according to the KOSTRA-DWD-2000 (DWD 2005) is used. In the simulations the precipitation intensity for the design rainfall for rural areas with a duration of 15 minutes and a recurrence interval of one year is applied. In the climate change version the precipitation intensity is adapted to the next recurrence interval. Thus, for a rural area the value for a recurrence interval of two years is used. In the model municipality the corresponding local precipitation intensity is 125 L/(s·ha) at present and 161.4 L/(s·ha) in the climate change version.

Adaptations of water infrastructures can have an impact on the existing fee system of water supply and wastewater discharge and treatment. Depending on the transformation strategies calculated by the model, which, e.g. can imply a change from central to more decentralised water infrastructures, institutional changes, e.g. in the fee system might be required (Bellefontaine et al. 2010; Duffy & Jefferies 2011). Fees are not considered as a driver in the presented scenario management approach, because thereby the selection of adaptation strategies could be influenced or limited. Rather, a compatible fee system for the optimised solution can be determined. Adaptation possibilities of the present fee system can be deduced for the calculated transformation strategy and generalised as recommendations for the adaptation of a future water fee system.

The modifications can be an implementation of base fees and per unit used fees, one time payments for modernisation, or subsidies. The adaptation of the fee system depends on the effects of demographic change and has to take social fairness into consideration. A flat rate for potable water can be an option for securing the functional capability of the water supply network. Wastewater fees are calculated based on the consumption of potable water. They may no longer be suitable when innovative sanitary systems are implemented and the consumption of potable water is reduced further. Then, other factors like nutrient and energy content of wastewater or the type of treatment will be of importance and have to be taken into consideration in the fee system. If the existing urban drainage system is transferred to a decentralised system the ownership structure is of importance for the fee system. The owner and the operator of decentralised facilities on private properties can be the municipality or the private person. General adaptation possibilities of the fee system and the application for a model municipality are presented in Breitenbach et al. (2015).

The above presented future developments of different drivers are combined in a software-based scenario-manager. In the scenario-manager probabilities of (co-)occurrence are assigned to the different prognoses for each driver and the interdependencies of the drivers are integrated.

The scenario-manager is integrated into the database management system of the decision support and optimisation model. The user interface of the scenario-manager allows a clear selection and arrangement of scenarios. The underlying network and site-specific data for the optimisation model are processed for different time steps. In the next step objective functions are selected and weighted for the optimisation process (Schmitt et al. 2014). The optimisation model calculates adaptation possibilities and transformation strategies for water infrastructures, showing their chronological and spatial implementation over the considered period of time (Baron et al. 2015). Besides, the visualisation of model results, the underlying scenario with the developments of the selected drivers, e.g. demographic developments, is visualised in the interpretation tool.

The arrangement of different drivers to different scenarios allows a comprehensive illustration of future states. Sensitivity analyses will be undertaken with the optimisation model for different scenarios in order to examine the sensitivity of results and thus the robustness of transformation strategies.

For the presented model village three scenarios were defined and will be tested (see Table 3).

Scenarios are an important input for the decision support and optimisation tool. They have a big influence on the calculated transformation strategy. For the determination of developments of each driver different adequate methods were presented and illustrated for a model municipality. The presented scenario management approach bundles the different drivers' developments and ensures a realistic modelling. The approach includes a wide range of drivers and therefore differing scenarios covering a wide spectrum of factors can be derived. The specification of scenarios by the user has the advantage that the expertise of the user is included. The specification of scenarios can reduce the number of possible adaptation measures. Sensitivity analyses can then be undertaken for different scenarios in order to examine the robustness of calculated transformation strategies. Thus, the uncertainties in the modelling process can be reduced.

The scenario management approach is applied on two model municipalities in the project, but general recommendations for rural areas facing the same problems can be derived. The approach of combining spatial developments with population prognoses and connecting them to water infrastructures, especially enables an integrated scenario management. Formulating new recommendations for the fee system can support or even enable the implementation of innovative wastewater technologies.

The INIS joint research project ‘SinOptiKom – Cross-sectoral optimization of transformation processes in municipal infrastructures in rural areas' is funded by the German Federal Ministry of Education and Research and is conducted by a research consortium of ten cooperation partners. Their contributions to the project's progress and diffused input to the paper are highly appreciated.