2.1. The ‘One Water’ Concept
The One Water concept characterizes water in all its forms [9
], and allows a demonstration of the impact of urbanization as illustrated in Figure 1
, where Figure 1
a schematically depicts the directions of water flow in a predevelopment, natural area, and Figure 1
b shows the results of urbanization. Comparing Figure 1
a,b, infiltration decreases due to more paved surfaces, and more water being needed to meet water demands and, hence, frequent reliance on groundwater as water source, as well as the importing of additional surface waters. With little water reuse occurring, substantial wastewater needs to be treated. Further, a subtle but very important dimension is related to evaporation/evapotranspiration (E/E): in pre-development, the E/E is a relatively constant return to the atmosphere via the water cycle, but E/E following urbanization results in the rate being discrete and substantial; namely, high only for limited periods of time immediately following precipitation events. For this reason, in Figure 1
b, the lines demarking the E/E rate have been depicted using dashed lines in Figure 1
b, to reflect the large variations over time in the post-E/E timeframe. More explanations will be described in the following sections.
An important element depicted in Figure 1
b is that with urbanization, increased impervious surfaces result in less natural water infiltration to the groundwater aquifer. Meanwhile, urban areas attract more citizens, which positively reinforces the likelihood of increases in impervious surfaces, diminishing groundwater recharge, and ultimately, lowering groundwater levels. Urbanization processes intensify the challenges: more flooding, increased demand for drinking water, and water quality impacts on waterbodies.
Note that in Figure 1
a,b, the approximate magnitudes of percentages of water and importation of water (later in this paper), runoff, infiltration, wastewater production, etc., pertinent to the region being assessed, are presented proportionally in terms of the “arrow” sizes. Figure 1
a shows typical water percentages under natural ground cover conditions (pre-development), and Figure 1
b shows the modified percentages under conditions of 75–100% impervious cover or full urbanization (post-development). The intent in the schematic is to demonstrate that the surface runoff volumes immediately following rainfall are dramatically higher with urbanization, and evaporation from impervious surfaces change greatly. Evaporation is high immediately after rainfall for the urban area but minimal otherwise so, while percentagewise there is a tendency to consider evaporation in post-development situations as if it is continuous and constant, evaporation is not continuous; relative to the magnitudes during heavy rainstorms, evaporation is relatively small. As an indication of context, Toronto’s Wet Weather Flow Master Plan (2003) indicates that 50% of the total average volume of rainfall events in Toronto occur during just a few events within the year (Toronto, 2020), indicating the time-sequence of the two types of pre-development E/E, relative to post-development E/E, are very different.
Various colors used in Figure 1
indicate conditions that imply good water quality (blue) to poor (dark grey) indicating the potential for the quality of the water to be deteriorated. Particularly noteworthy are the significantly lower percentages of water migrating to both deep and shallow groundwater infiltration, as well as the color change, comparing Figure 1
a,b. It follows that there is much more water to be managed at the surface during a sizable rainfall event as a result of urbanization, and less going to groundwater [11
Historically (i.e., prior to the 1970s), typical responses following urbanization were to manage storm water by moving the water as quickly as possible to the nearest watercourse, ultimately discharging to the ocean. Since this approach puts enormous pressures on downstream cities and rivers/streams, alternative options were essential. The growth of cities with large populations and associated demands for water supply creates the need to meet the demand. Meanwhile, however, climate change is complicating the issues; hence, the future challenges are ominous, as urbanization continues and climate change intensifies.
2.2. Implications of Sponge City
One Water is the provision of a focused strategy to improve the understanding of water movement rates. The One Water concept demonstrates that there continue to be major issues of infiltration to both deep and shallow groundwater needing to be addressed to avoid city subsidence. This issue worsens as population flux continues into cities (e.g., 65% of the world’s population is expected to be in urban areas by 2050 [12
]), creating spikes in water demand as a result.
The Chinese government has identified multiple strategies to address the urban flooding issue, particularly after many massive urban flooding events observed in the early 2010s. Approaches to reduce the hazard from flooding events in China include 30 designated cities that adopted Sponge City (SC) initiatives starting in 2015 [2
]. Starting in December 2013, there have been many policies and guidelines developed for building “Sponge City” scenarios. The main goal of those policies and guidelines is to “alleviate the adverse effects of urban construction and recycle 70–90% of stormwater in-situ, and with the hope this can be achieved through combining retention, storage, purification, and reuse before discharge by applying the green infrastructure concept” [3
As briefly mentioned previously, SC has very similar concepts in comparison with its North American counterpart: Low Impact Development (LID). In North America, LID plus stormwater management ponds have been widely implemented. The intent of all these various applications has been to decrease urban flooding, encourage infiltration, and minimize discharge (e.g., to maintain quantities to historical/pre-development levels) to nearby waterways. With the success of LID in North America, SC was seen as having the potential to increase infiltration, thus decreasing flow toward rivers, and instead, toward groundwater. However, since infiltration is a slow process, whether the SC initiatives will create sufficient infiltration to both shallow and deep groundwater recharge remains very challenging [13
Another important goal of SC is to reduce the magnitude of damage due to flooding, but the range of options are highly influenced by location. Constraints on SC installation arise due to land availability and economics: in China, with high land values, there are limitations due to spatial requirements. A list of some SC initiatives is provided in Table 1
, with the assigned numbers indicating the suitability of different parameters, clearly indicating SC are primarily functional for surface water balance, not for flood control.
SC applications can change where water travels (see Figure 2
) as well as changing impacts on the water quality, as indicated in Table 1
but since infiltration is a slow process, SC enables only modest influences to decrease flooding events in the absence of large volumes of retention storage (e.g., stormwater management ponds), which are not available in post-development China. A limitation of SC is also implied from Figure 2
, where the thickness of the arrow indicates that surface runoff is reduced, but still large. This indicates SC may be sufficient to treat frequent storms (relatively small surface runoff volumes) but not major storm events. Jiang et al. [13
] examined a case study area in London, Ontario, Canada for the feasibility of applying lot-level SC applications to reduce the damage from street and basement flooding [13
]. The results of Jiang et al. [13
] show that SC has the potential to eliminate flooding for a 2-year return period storm, by implementing a substantial array of various SC applications, using one each, for each single-detached house lot [13
]. If many lot-level SC applications are implementable, major storms such as the 25-year return event, flooding issues will still occur, but with the help of SC applications, the quantity of surface runoff can be reduced.
Thus, One Water indicates that possibilities may exist to restore pre-development conditions for the 2-year recurrence storm and/or less, with installation of a full array of SC options installed at lot level. One Water recognizes the limited opportunity to attenuate flooding caused by major storm events as being a key aspect; namely, to rethink how issues of urban flooding and city subsidence can be more fully addressed in China.
The utility of the color scheme throughout Figure 1
, presented earlier, and Figure 2
and Figure 3
, are to identify needed guidelines; the quality of the water needs to be maintained. For example, water departing from an industrial area or from heavy traffic roadways may have greases and oils, heavy metals, PAHs, etc., whereas water sourced after passing through vegetated swales will be much cleaner as a result of the slow passage time and filtration/sorption by the vegetation. Similarly, planting trees en masse can help to lower flooding by enhancing the soil’s ability to absorb rainwater and provide a multitude of additional benefits including greater carbon sequestration and surface soils which absorb water effectively (as trees grow, they create new cavities in soil, slowing the speed at which rainwater leaches out of the soil and enhancing infiltration) [15
In Figure 3
a, the addition of arrows (in comparison with Figure 1
b) reflects the quantities of water imported and sourced from groundwater to respond to the water demands from the megacity. Wastewater discharges implied by releases to the environment are also shown in Figure 3
a with minimal groundwater recharge. Figure 3
b is a schematic view characterizing desirable water management in an urbanized scenario with a more balanced approach. This approach is accomplished by:
intentional increases of acceptable quality (i.e., light grey) water being infiltrated to ensure that the integrity of the groundwater sustainability is preserved, while also increasing both shallow and deep groundwater infiltration [16
reduced water importation as the result of success at reducing water demands from the megacity;
enhanced evapotranspiration occurs due to sponge city components such as bioretention cells;
stormwater reuse arises from efforts to successfully reuse some of the stormwater by, for example, water capture to facilitate vegetative watering and/or for rainwater being used for toilet flushing.
In essence, SC illustrates the need to be concerned with the transformation embodied to approach Figure 3
b. However, while helpful as a guiding principle, limits exist to which SC can manage the water, as per the One Water concept. At present, the key points are:
One Water is useful to portray the principles of decision-making in relation to water movement and organize the thinking and the potential effectiveness of specific initiatives;
One Water does not easily reflect the implications of different return periods of flooding, e.g., the 2-year impact versus the 100-year impact;
One Water can portray the need to be highly cautious about the water quality impacts of infiltration.
One of the bases for SC is to restore the surface water balance to pre-development conditions. End-of-pipe treatment works but SC initiatives do not have the storage magnitudes as needed, since storage is expensive. Knowing and understanding that infiltration is a slow process is essential, because without storage, “restore” cannot be accomplished.
Although little research has been conducted on urban evaporation, Ramamurthy and Bou-Zeid (2014) found that impervious surfaces can promote significant evaporation and are able to evaporate at higher rates than vegetated surfaces when wet, especially just after raining and when combined with intensive solar radiation, due to the heat storage capabilities prior to a rain event [17
]. They reported that evaporation from concrete pavements, building rooftops and asphalt surfaces is discontinuous and intermittent, accounting for ~18% of total latent heat fluxes during a relatively wet 10-day period, with a significant impact on the urban surface energy balance during the 48 h following a rain event when evaporation from impervious surfaces is the highest. Thus, understanding the temporal variability of water flux in urban models is critical. Alternatively, during dry periods, there is limited water availability to evaporate [18
]. Hence, when flooding is high, evaporation is high from impervious surfaces, hence demonstrating the rationale for the dashed lines in Figure 1
, Figure 2
and Figure 3
2.3. The Groundwater Perspective
Due to the profound reduction of groundwater recharge schematically shown in Figure 1
but continued withdrawals of water for water supply for urban areas, groundwater levels will continue to lower, and concomitant subsidence will occur. As a result of the growth of megacities, issues of flooding and city subsidence result from groundwater withdrawal to meet water supply demands, have been exacerbated. Numerous examples throughout the world are already evident, as demonstrated for select cities around the world, as presented in Table 2
. To put in context, subsidence rate in many coastal cities have now exceeded the absolute sea-level rise by a factor of 10 [19
In Beijing, the subsidence rate increased up to 52 mm per year from 2003 to 2010 [21
]. The root cause of such damage is that water demands substantially increased due to the rapid growth in Beijing’s population [22
City subsidence levels indicate massive groundwater withdrawals are evident and substantial. Singh et al. [23
] reported that in Beijing, the annual groundwater extraction reached 2.6 × 109
/year, which included roughly 1 × 106
/year for exploitation. Gao et al. [24
] also suggested that the South-to-North canals now provide nearly two-thirds of urban water supplies for Beijing, where these amounts were previously supplied by groundwater withdrawal. Due to such large amounts of historical withdrawal and long-term groundwater exploitation, both groundwater level declines and land subsidence were the result. Cao et al. [25
] reported that 66% of the Beijing plain has been affected by more than 50 mm land subsidence, covering an area of 4.2 × 103
. Ye et al. [26
] have also provided evidence showing the overall land-subsidence affected areas in China have exceeded 90,000 km2
by 2015. On all accounts, overexploitation of groundwater causing land subsidence is life-threatening to citizens [27
] and large groundwater withdrawals are leading to land subsidence [28