Modelling urban stormwater and irrigation management with coupled blue-green infrastructure in the context of climate change

Today, the world is facing rapid global urbanization, climate change and increasing water scarcity, and cities are confronted with the consequences of global warming such as heavy rainfall, drought and heat, which defines the management of the urban water cycle as one of the most important future challenges for urban planners (Larsen et al. 2016). We can therefore say that we are in the midst of a paradigm shift in the way we manage water. As a result, the relationship between urban development and water resources is becoming an important part of international and national discussions to develop integrated adaptation and mitigation strategies (UNEP-DHI 2014).

To be prepared, urban planners need to consider urban water management alongside the built environment, pollution control policies and solid waste and stormwater management (Furlong et al. 2017). An integrated urban water management (IUWM) approach can help to improve the way resources are managed throughout the urban water cycle. It aims to include all parts of the water cycle, recognizing it as an integrated system that takes into account water needs for residential, industrial, agricultural and ecological uses, and provides a framework for planning, designing and managing urban water systems (Belmeziti et al. 2015; Larsen et al. 2016). IUWM aligns with the UN Sustainable Development Goal on water (SDG 6), which is key to creating sustainable communities (SDG 11 on cities), and directs UN member states to implement integrated water resource management at all levels by 2030 (United Nations 2015). IUWM also brings together water supply, sanitation, stormwater and wastewater management and integrates them with land use planning (Brudler et al. 2016; Fang et al. 2016; Sørup et al. 2020). Successful implementation requires cooperation between multiple jurisdictions over which the urban area is spread, and can help cities and water utilities develop more robust water systems to meet their needs now and in the ever-expanding urban future (Furlong et al. 2017). In essence, IUWM reimagines a city's relationship with water, and rethinks how water resources and associated infrastructure can be designed and managed.

Within the IUWM, stormwater management has great potential for improvement. To date, the main objective of management has been to control flooding, i.e., to collect rainwater and discharge it into the sewerage system so that it is no longer available in the city. However, changes in the distribution of rainfall, the length of dry periods and the increased frequency of heat periods are leading cities to consider countermeasures such as the integration and implementation of urban blue-green infrastructure (BGI).

In general, however, the main focus of engineered BGI is on rainwater retention, stormwater management, infiltration or climate regulation in urban green spaces (Oberndorfer et al. 2007; De Vleeschauwer et al. 2014; Voskamp & Van de Ven 2015; O'Donnell et al. 2020). Increased urban densification and expansion, combined with extreme rainfall events, has led to an increase in stormwater runoff, putting tremendous pressure on existing infrastructure such as networks and centralized wastewater treatment plants (Hoffmann et al. 2015; Khurelbaatar et al. 2021). Concepts have been developed to manage and control stormwater in a decentralized and integrated manner, including low impact development (Pati & Sahoo 2022; Szeląg et al. 2022), water sensitive urban design (Kuller et al. 2017; Nguyen et al. 2021), sponge city concepts (Li et al. 2017; Chen et al. 2022; Siehr et al. 2022) and BGI (Jayasooriya & Ng 2014; Busker et al. 2022; Kvamsås 2022; Knappe et al. 2023).

The BGI concept is currently being integrated into a new development project called Leipzig416 (www.leipzig416.de) in Germany (see also www.ufz.de/leipzigerblaugruen). A major challenge in the integration of BGI in the context of urban water management is the competition for space between underground (underground car park, infiltration trench and storage infrastructure) and above-ground structures (roads, pavements, playgrounds, cycle paths and car parks), as well as the need for natural land use (e.g., tree planting and green spaces). However, urban districts can be designed in a way to combine both, with engineered BGI for ecosystem service benefits and a more compact city (McDonald et al. 2023). The overall objective of the project is to prevent flooding, increase the local water availability, to reduce the load on the sewerage system, improve the microclimate at the urban block level and to achieve a climate resilient urban water management through the use and integration of sustainable and resilient urban BGI.

Many studies have examined irrigation demand and water use as an important response to climate change, mainly focusing on the needs of large-scale agricultural systems (Paschold & Beltz 2010; Lupia et al. 2017; Schwarz-v.Raumer et al. 2023). For example, for outdoor vegetable crops, there is an irrigation control method, the ‘Geisenheim method’, which considers a crop coefficient kc for many vegetable crops at different stages of development (Hochschule Geisenheim 2021). Schwarz-v.Raumer et al. (2023) presented a web-based tool designed to harmonize rainwater and greywater drainage while meeting the water needs of vegetation in urban green spaces. This tool integrates geographical information service (GIS) data to assess the rainwater and greywater harvesting potential within a catchment. It also calculates daily water requirements for different vegetation types, taking into account local weather patterns, shading, soil moisture and short-term soil conditions. However, the existing literature does not comprehensively address this issue in an urban context to investigate how BGI can reduce runoff and how much of this runoff is available to support BGI with irrigation.

The main objective of the City of Leipzig and the investor, in terms of stormwater management, is to ensure a zero-runoff urban district. It is, therefore, necessary to investigate the extent to which the integration of multifunctional BGIs can reduce the volume of rainwater runoff discharged into the main sewer networks, and the extent to which water is available for an optimal irrigation of green roofs, courtyards and parks, even in dry years, with rainwater collected in storage infrastructures and used for irrigation. The final results/findings will then be incorporated into the ongoing design/planning process.

The issue of incorporating or linking BGI practices for the dual purpose of stormwater management and meeting irrigation water needs in urban areas has not received sufficient attention. Therefore, in this study, we have developed and applied an irrigation model called blue-green infrastructure irrigation (B-GRIIN), which includes coupled BGIs and the possibility of including and using the collected water for irrigation.

The specific objectives of this study were (i) to perform simulations using historical rainfall data to achieve a balanced urban water system by integrating BGIs with the B-GRIIN model to eliminate network discharge; (ii) to identify the water/irrigation demand of BGIs and (iii) to determine the influence of the coefficient kc in the overall water balance of the urban district.

Stormwater management is one of the biggest challenges for the development of the urban district. The surrounding combined sewers of the municipal water utilities have already reached their maximum capacity and therefore, the city of Leipzig prohibits the discharge of rainwater into the existing network. The concept that is now being pursued is the development of a zero-runoff urban district to achieve a climate resilient urban water management through the use of sustainable and resilient urban BGI. It is planned that rainwater from private spaces will be collected/retained/evapotranspirated/stored/infiltrated in a decentralized manner on the respective properties. However, rainwater from public spaces (not covered in this study) will also be collected/retained/evapotranspirated and infiltrated semi-centrally.

In order to assess irrigation needs in space and time throughout the year, it is essential to consider meteorological patterns (Schwarz-v.Raumer et al. 2023). Daily variations in precipitation and evapotranspiration over the course of the year play a key role in influencing the availability of soil water to plants, thereby determining the need and amount of irrigation required. The irrigation module used was modified from Paschold & Beltz (2010). The aim of urban irrigation is to maintain the water content in the root zone at a level that allows plant growth and ensures the ecosystem services of the BGI throughout the year. The total substrate thickness of the extensive green roof and the underground car park (S_tot-green and S_tot-car) was defined as the root zone. In this study, we further defined that the available substrate field capacity (aFC) should not fall below 30% and should not exceed 80% according to Eppel et al. (2012) and Hochschule Geisenheim (2021). Irrigation pulses were set to 20 mm.

A plant substrate with plant-available field capacity (FC) of 19% by volume was assumed. This refers to commercially available green roof substrates with a maximum FC in the range of 20–50% (www.zinco.de/substrate, www.optigruen.de/produkte/substrate) and a corresponding plant-available FC in the range of 10–24%. The ability of green roof substrates to store water has commonly been expressed by the maximum water holding capacity (MWHC), as defined in the German Green Roof Guideline (Lösken et al. 2021), but studies have confirmed that the MWHC overestimates the substrates' water holding capacity and better represents the maximum structural load. Fassman & Simcock (2012) found that agronomic measurements, corresponding to water stored between FC and wilting point, better represent the water holding capacity of green roof substrates. Thus, B-GRIIN simply expresses the water content available to the plant in the system at any given time.

Potential evapotranspiration ET_p according to Penman Monteith (see Allen et al. 2005) was obtained from the German Weather Service (www.dwd.de; DWD climate station 2928 in Holzhausen, Germany) for 2018 and the years 2011–2022. For the calculation of the actual evapotranspiration Et_a, an annual average crop coefficient kc (scaling factor) for grass of 0.8 was used (Sun et al. 2012; Pittenger 2014; Hörnschemeyer et al. 2021; Schwarz-v.Raumer et al. 2023). However, it is recognized that there may be multiple kc values for a single plant species, depending on the developmental stage of the plant and the season (Nivala et al. 2022). For example, during germination and establishment, most of the evapotranspiration occurs as evaporation from the soil surface. Therefore, a simplified sensitivity analysis was performed for the dry year 2018 to assess the influence of the crop coefficient kc on the irrigation requirements.

The B-GRIIN model considers a constant water loss through interception I of 1 mm per day, which reduces the effective precipitation that infiltrates into the substrate and/or reduces the runoff from sealed spaces during each time step.

The overflow of the storage infrastructure is equal to the amount of water infiltrating into the groundwater from the infiltration ditch, hence GIF = Overflow_SI.

In the model, SI_tot is set to 200 m³ as the total volume of the storage infrastructure. Excess water from SI flows into the infiltration trench (Overflow_SI). Overflow_SI is equal to the amount of water that infiltrates from the infiltration ditch to the groundwater, therefore, Overflow_SI = GIF.

For modelling purposes, we assume that the substrate and retention layers of the extensive green roof as well as from the underground car park roof are saturated on day t₀. Therefore, the total available substrate storage in the extensive green roof (S_tot-green) is 37 mm (19 mm available water stored in the substrate layer and 18 mm in the retention layer) and 208 mm (190 mm stored in the substrate layer and 18 mm in the retention layer) in the underground car park roof system (S_tot-car; Figure 3). Expressed as a sequential model, S_tot-green and S_tot-car depend on the previous day's storage.

As described above, the initial volume of SI_tot is set to 200 m³ as the total volume of the storage infrastructure, assuming that SI was full at the start of the simulation. The change in water volume at t₁ is defined by the inflow to SI as the sum of Outflow_green, Outflow_car and Runoff_sealed minus the water that is used for irrigation (Extraction_SI = IR) and the Overflow_SI.

In order to develop comparable stormwater management scenarios for private areas, a ‘model block’ was used for the calculation (see Section 2.3).

In order to simulate the water balance of the private spaces, the following scenarios were considered for the private and public areas:

Scenario 0 (S0): Baseline scenario. No consideration of the extensive green roof (GR_roof) and the underground car park roof system (GR_car).

Scenario 1 (S1): Consideration of both types of green roof (GR_roof and GR_car) but no irrigation.

Scenario 2 (S2): Consideration of both types of green roof (GR_roof and GR_car) and irrigation of the underground car park roof only (GR_car).

Scenario 3 (S3): Consideration of both types of green roof and irrigation of both (GR_roof and GR_car).

Based on the objectives of the City of Leipzig and the investor for the development of a zero-runoff urban district, the irrigation model B-GRIIN was developed to model the total water budget of residential, private areas. The basic principle for determining individual irrigation inputs is based on the usable FC in the root zone of the vegetation and its evapotranspiration. By integrating and coupling irrigated and non-irrigated green areas with sealed areas, infiltration and storage structures, the water balance with and without irrigation of sub-areas can be estimated for the model block.

To date, the literature on this topic is still scarce. However, Schwarz-v.Raumer et al. (2023) follow the same basic idea to better understand the relationship between urban water collection and irrigation needs, and address three key questions: (i) how much water can an area supply, (ii) what is the amount of water needed to sustain vegetation in that area? and (iii) what storage capacity is needed to bridge the gap between rainfall and irrigation periods?

The implementation of BGI in an urban context has become increasingly interesting and can contribute to climate change adaptation and mitigation. In this study, we developed and applied an irrigation model called B-GRIIN and identified the water/ irrigation demand of BGI using the collected/available water for irrigation. In order to achieve a positive and zero-water balance without network discharge, BGI was coupled and integrated into the B-GRIIN model. The modelling results over a 12-year period show that it is possible to achieve a zero-water balance by implementing BGI at the block level, and to provide additional water for the irrigation of the courtyard (underground car park roof) and possibly the extensive green roof. Irrigation is particularly important to establish stable urban vegetation throughout the year, in order to fully utilize the engineered water management and ecosystem services of the BGI. The B-GRIIN model uses a general crop coefficient factor (scaling factor) kc of 0.8 for grass, but we have also shown that this has a large influence on the predicted irrigation volume and thus on the overall water balance. This result highlights the importance of the scaling factor in process design and the need to specify the factor as a function of climate, vegetation, technology and site. Further research is needed to calibrate the model to real infrastructure at city scale. There is no doubt that B-GRIIN is a simplified model based on an idealized system, but it can serve as a benchmark to further investigate the influence of BGI on the overall urban water balance and to support the integrated planning process for climate resilient cities.

A practical implementation of the integrated concepts of a ‘Water Sensitive City’ requires (a) the coupling of heavy rainfall management and irrigation of BGI + urban green; (b) placing the availability of land and water as essential design variables at the centre of blue-green planning; (c) a precise (quantitative) definition of the targeted water management functions (flood control, irrigation and cooling) while maintaining the natural local water balance and (d) the interdisciplinary engineering design of coupled BGI.

This work was funded by the Federal Ministry of Education and Research (BMBF-FKZ: 033W110A) as part of the funding initative: RESZ-Verbundvorhaben LeipzigerBlauGrün – Blau-grüne Quartiersentwicklung in Leipzig – Teilvorhaben 1: BlauGrüne Systemarchitektur – Wasser, rechtlich-administrative Realisierungsbedingungen und Verbundkoordination and Blau-grüne Quartiersentwicklung in Leipzig „Leipziger BlauGrün II’.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.