RESEARCH PROJECTS

Background: High-dimensional states can often leverage a latent low-dimensional representation. This inherent compressibility enables those high-dimensional states to be reconstructed or predicted from sparse measurements through sparse sensing. As a promising technique in data compression, reconstruction, and prediction, sparse sensing has not been widely used in environmental engineering and geosciences. In this project, some efforts have been made to reconstruct and/or predict streamflow and water quality (e.g., nitrate and phosphorus concentrations) time series across watersheds using sparse sensing. These works focused on exploring the applicability of sparse sensing on environmental signals and pursuing effective strategies to reduce the required sampling efforts. Other potential applications of sparse sensing include sensor location optimization, gap filling, and making predictions, especially through integration with data fusion.

Research questions: 1) How well can we predict streamflow and water-quality time series and reduce sampling frequency using sparse sensing? 2) In which types of watersheds does sparse sensing-based regionalization work the best? 3) Is there any spatial pattern of streamflow predictability using sparse sensing and how does it relate to watershed characteristics?

Approaches: Compressed sensing; Data-driven sparse sensing

Related publication: Zhang, K., Bin Mammoon, W., Schwartz, E, & Parolari, A.J. (2023). Reconstruction of sparse stream flow and concentration time-series through compressed sensing. Geophysical Research Letters, 50, e2022GL101177.

Zhang, K., Luhar, M., Brunner, M.I., Parolari, A.J. (2023). Streamflow prediction in ungauged watersheds in the United States through data-driven sparse sensing, Water Resources Research, 59, e2022WR034092.

Background: Adaptive management for resilient water quality is fundamentally a sensing and control problem that plays out over years to decades. Water managers and utilities are challenged to collect a wide range of quantitative and qualitative data and combine it to evaluate, compare, and choose from a suite of infrastructure investments intended to maintain or improve water quality. The quality of these decisions and management success depends strongly on the uncertainty inherent to the data and the ability to forecast how decisions will translate into water quality protection. In this project, low-dimensional models will be integrated with legal analysis to define the safe operating spaces of water quality within our quantitative forecasting framework under different uncertainties, policies, and decision-making strategies.

Research questions: 1) How does water quality trading program impact the trajectory and resilience of the watershed water quality? and 2) How does the safe operating space change with different levels of uncertainties and decision-making strategies?

Approaches: Legal analysis; Low-dimensional water quality modeling

Background: Green infrastructure (GI) has gained traction as a preferred stormwater management practice for its benefits in mimicking pre-development hydrology, removing non-point source pollution, and improving water quality. GI has been adopted as a key component of the Metropolitan Sewerage District’s (MMSD’s) strategic plan. There is a strong and urgent need to quantify the system-scale impact of existing and future GI implementations and prioritize GI investments. However, this is not a simple task because the GI impact can be diluted by uncertainties from the climate and those associated with the complex environment such as the changing land cover. To address these challenges, we proposed a data-model integrated approach in the first phase of this project, during which we proposed recommendations for data collection, modeling and analysis, and developed and applied a GI modeling workflow to study GI performance in the combined sewer area. In this project, we will integrate data-based analysis with hydrologic & hydraulic models to determine 1) the system response and priority sewersheds for GI considering recent decadal and future forecasted precipitation, 2) the system response and priority sewersheds for GI in the whole MMSD service area, and 3) the priority sewersheds for GI considering various climatic, hydrogeologic, hydraulic, and socio-economic factors.

Background: Urbanization increases impervious cover and involves urban drainage infrastructure which dissects the subsurface and induces artificial controls on streamflow by draining water from the subsurface soils, e.g., inflow and infiltration (I&I). With this effect, sometimes referred to as “urban karst”, the slow subsurface flows can be redistributed to fast sewer flow, and the baseflow recession of streamflow can be altered. Therefore, understanding the volume and dynamics of I&I and its impact on urban hydrology is a key in urban hydrology characterization. 

Research questions: 1) What fraction of the urban water balance is discharged through I&I? 2) How does subsurface drainage such as I&I affect baseflow recession in urban streams? and 3) Considering I&I, how does enhanced infiltration by green infrastructure affect the stormwater distribution?

Approaches: Hydrograph analysis; Surface-subsurface hydrologic modeling

Collaborators: Metroplitan Milwaukee Sewer District (MMSD)

Relevant publication: Zhang, K., Sebo, S., McDonald, W., Bhaskar, A., Shuster, W., Stewart, R., Parolari, A.J. (2023). The role of inflow and infiltration (I/I) in urban water balances and streamflow regimes: A hydrograph analysis along the sewershed-watershed continuum, Water Resources Research, 59, e2022WR032529.

Zhang, K., Parolari, A. (2022). Impact of stormwater infiltration on rainfall-derived inflow and infiltration at watershed scale: A physically based surface-subsurface urban hydrologic model. Journal of Hydrology, 610, 127938.

Background: Well-designed and implemented green stormwater infrastructure (GSI) can help to recover the natural hydrologic regime of urban areas. A large-scale GI planning requires a good understanding of the impact of GI spatial allocation on surface-subsurface hydrologic dynamics. This study developed and utilized a coupled surface-subsurface hydrological model (SWMM-MODFLOW) and other variably saturated hydrologic models to simulate two-way interactions between GI and subsurface hydrology at lot and catchment scales. The goals are to explore the optimal designs and spatial allocation patterns of GI that provide the best hydrologic benefits.

Research questions: 1) How does GSI perform in shallow groundwater environments? 2) What are the appropriate designs for GSI in such environments? 3) How does the spatial allocation of GSI affect the local and regional groundwater table fluctuation?

Approaches: SWMM-MODFLOW; Variably-saturated hydrologic models

Related publication: Zhang, K., Chui, T.F.M. (2020). Assessing the impact of spatial allocation of bioretention cells on shallow groundwater - an integrated surface-subsurface catchment-scale analysis with SWMM-MODFLOW. Journal of Hydrology, 586, 124910.

Zhang, K., Chui, T.F.M. (2020). Design measures to mitigate the impact of shallow groundwater on hydrologic performance of permeable pavements. Hydrological Processes, 34, 5146-5166. (IF: 3.784). 

Zhang, K., Chui, T.F.M. (2019). A review on implementing infiltration-based green infrastructure in shallow groundwater environments: Challenges, approaches, and progress. Journal of Hydrology, 579, 124089. (IF: 6.708)

Zhang, K., Chui, T.F.M, Yang, Y. (2018). Simulating the hydrologic performance of low impact development practices in shallow groundwater via a modified SWMM. Journal of Hydrology, 566, 313-331.

Zhang, K., Chui, T.F.M. (2018). Interactions between shallow groundwater and LID underdrain flow at different temporal scales. Hydrological Processes, 32(23), 3495-3512. (IF: 3.784)

Zhang, K., Chui, T.F.M. (2017). Evaluating hydrologic performance of bioretention cells in shallow groundwater. Hydrological Processes, 31, 4122-4135. (IF: 3.784)

Abstract: As part of the further expansion of Shek Wu Hui Sewage Treatment Works (SWHSTW) undertaken by AECOM Asia Co Ltd., a porous pavement study (RD1108) was initiated. Pavement trial panels were designed and constructed at SWHSTW and Stonecutter Island Sewage Treatment Works (SCISTW). The porous pavement trials include three pedestrian trial panels (3m x 1m each) and four vehicular trial panels (30m x 3m each), three at SWHSTW one at SCISTW. The surface pavers of the pedestrian ones were porous blocks, grass cover and resin bound surfacing material, while those of the vehicular ones are open cell pavers and two types of porous blocks at SWHSTW, and permeable interlocking concrete paver at SCISTW. Construction of trial panels commenced in June 2016 and completed in January 2017. A series of hydrologic testing and monitoring was conducted, which included the in-situ permeability tests, the seven-month monitoring under natural rainfall during the rainy season, the 12-month serviceability trial (vehicular traffic count and topographic survey) and the two sets of artificial rainfall experiments before and after the 12-month serviceability trial (in March 2017 and February 2018 respectively). The hydrologic performance was quantified by runoff peak reduction, volume reduction, peak delay, and was compared among panels and between the artificial rainfall experiments in 2017 and 2018. 

Related publication: Zhang, K., Huang, P., Chui, T.F. (2022). Runoff mitigation by underdrained permeable pavement in shallow groundwater environments: A field investigation? Journal of Hydrologic Engineering, 27(7), 04022011.