Optimal selection of hydraulic indexes with classical test theory to compare hydraulic performance of constructed wetlands

doi:10.1016/j.ecoleng.2019.105687

Ecological Engineering

Volume 143, 15 January 2020, 105687

https://doi.org/10.1016/j.ecoleng.2019.105687 Get rights and content

Highlights

•
Inconsistencies within hydraulic indexes were revealed.
•
Hydraulic indexes were optimized to represent hydraulic performance.
•
Compatibility, discrimination, difficulty of hydraulic indexes was compared.
•
Mutual relationships of hydraulic indexes were also studied.
•
Three hydraulic indexes are recommended.

Abstract

Residence time distributions (RTDs), obtained from tracer experiments, are one of the main tools for investigating the hydraulic performance of constructed wetlands. However, the existence of various hydraulic indexes derived from RTDs has led to inconsistency in their application for comparing the hydraulic performance. In this work, eight hydraulic indexes were initially selected based on their popularity, and then divided into three categories: hydraulic efficiency indexes (HEIs) (λ_m, λ_e, λ_p, and MI), short-circuiting indexes (SIs) (t₅ and t₁₀), and mixing indexes (MIs) ( σ² and Morril index). Then, the hydraulic indexes were optimized for compatibility, discrimination, difficulty, and their mutual relationships between different categories. The results showed large inconsistencies among HEIs, and small inconsistencies among SIs and MIs. Among the four HEIs, λ_e performed best in terms of compatibility, discrimination, and difficulty. Among SIs, t₅ and t₁₀ differed little in the three aforementioned aspects. Among MIs, the Morril index performed better than σ² in terms of discrimination and difficulty. The significant correlation between short-circuiting flow and hydraulic efficiency highlights the necessity of reducing short-circuiting flow. Within each category, λ_e, t₁₀, and the Morril index are recommended to quantify hydraulic efficiency, short-circuiting flow, and mixing flow, respectively. This study brings clarity to the application of hydraulic indexes and provides uniform standards for the quantification of hydraulic performance.

Introduction

Flow pattern inside constructed wetlands (CWs) is identified to affect pollutants removal rates significantly (Schuetz et al., 2012; Wahl et al., 2010). Normally, flow in CWs is assumed to be ideal in the form of either plug flow or completely mixed flow (Persson et al., 1999). In plug flow, all fluid particles move from inlet to outlet with the same velocity. In completely mixed flow, fluid properties are spatially and temporally uniform within CWs. However, neither of these ideal conditions is achievable in the field; actual flow is represented best as a combination of plug flow and several continuously stirred tank reactors (Persson et al., 1999; Teixeira and Siqueira, 2008. Tracer experiments can reveal actual flow pattern within CWs and residence time distributions (RTDs), obtained from tracer experiments, are one of main tools investigating hydraulic performance of CWs (Khan et al., 2013). Numerous studies about the effect of CW design on hydraulic performance (Headley and Kadlec, 2007; Holland et al., 2004; Ioannidou and Pearson, 2019; Nuel et al., 2017; Sabokrouhiyeh et al., 2017; Schuetz et al., 2012) use various hydraulic index to compare hydraulic performance.

However, the existence of various hydraulic indexes derived from RTDs led to inconsistency in their application to compare hydraulic performance of CWs. The diversity of hydraulic indexes is shown in their different categories. Teixeira and Siqueira, 2008 collected 14 hydraulic indexes, and divided them into short-circuiting index (SI) and mixing index (MI) with 8 and 6 indexes included, respectively, according to their physical meaning. Wahl (2013) collected 16 indexes and categorized as single value indexes, composite indexes, global indexes, conditional probability functions according to their definition formula. Stamou and Noutsopoulos (1994) also divided hydraulic indexes into four categories, namely SI, MI, the degree of plug flow, and efficiency. Hence, in the study of effect of CW design (water depth, wetland shape, the aspect ratio, number and position of obstacles, vegetation and climate) on hydraulic performance of CWs (Ioannidou and Pearson, 2019), there is little consistency in the application of hydraulic indexes and researchers have different preference for hydraulic indexes (Bodin et al., 2012; Guzman et al., 2018; Holland et al., 2004; Persson et al., 1999). Moreover, different hydraulic indexes may render inconsistent conclusions. For example, Khan et al. (2013) obtained inconsistent ranks of CWs in hydraulic efficiency according two hydraulic efficiency indexes (HEIs) when studying how the layout of floating CW affected hydraulic performance.

Confronted with this situation, we urgently need to optimize the hydraulic indexes to appropriately represent the hydraulic performance. In this paper, eight hydraulic indexes were firstly selected, which were then divided into three categories based on their physical meaning. These hydraulic indexes were then evaluated using classical test theory (CTT) for compatibility, discrimination, and difficulty. Finally, relationships between hydraulic indexes in the different categories were explored. This paper can bring clarity to the application of hydraulic indexes derived from RTDs, and provide uniform standards for the quantification of the hydraulic performance of CWs.

Section snippets

Data sources

Many RTDs covering a wide range of hydraulic performance are needed for such research. Previously, Teixeira and Siqueira (2008) obtained these RTDs by tracer experiments conducted on five significantly different types of CWs. However, those CWs were unique and did not represent the diversity of CWs. The better option is to acquire these RTDs through tracer experiments on CWs under different factors, because CWs under different factors, such as layout of inlet and outlet, vegetation, aspect

Results

For revelation of inconsistency of hydraulic indexes, CWs were firstly ranked according to their value. Then, compatibility, discrimination, and difficulty of hydraulic indexes were evaluated by CTT. Finally, the relationships between HEIs and SIs, MIs were explored, as well as the relationships between SIs and MIs.

Analysis of hydraulic efficiency indexes

The characteristics of HEIs were explored to explain connection between them. Traditionally, the range of λ_m is between zero and one, meaning that the average residence time is not greater than the nominal residence time. However, of the CWs surveyed, there are four for which λ_m exceeds unity. Kjellin et al. (2007) reasoned that such a situation would be due to errors in measuring the flow rate and volume. Bodin et al. (2013) found that λ_m could exceed one when a tracer experiment was ended at 3

Conclusions

Eight hydraulic indexes were selected and classified as HEIs, SIs, MIs. Inconsistency of hydraulic indexes was revealed using ranks of wetlands, based on a large number of RTDs. Hydraulic indexes were optimized so as to represent the hydraulic performance. Compatibility, discrimination, and difficulty of hydraulic indexes were evaluated using CTT. The relationships between HEIs, SIs and MIs were analyzed. A significant correlation between short-circuiting flow and hydraulic efficiency was

Author contributions section

Junjie Liu: Conceptualization, Methodology, Writing-Original draft preparation,

Bin Dong: Writing-Review and Editing, Supervision, Funding acquisition.

Wangzi Zhou: Data curation, Investigation.

Zhongdong Qian: Supervision.

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

This material is based upon work supported by the National Natural Science Fundation of China under Grant No. 51779181. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The data and parameters used in this study are listed in the manuscript, supporting information, or available from the cited public sources. The supporting information additionally provides

References (42)

A. Angeloudis et al.
Flow, transport and disinfection performance in small- and full-scale contact tanks
J. Hydro-Environment Res.
(2015)
H. Bodin et al.
Tracer behaviour and analysis of hydraulics in experimental free water surface wetlands
Ecol. Eng.
(2012)
H. Bodin et al.
Influence of residence time analyses on estimates of wetland hydraulics and pollutant removal
J. Hydrol.
(2013)
T.-J. Chang et al.
Flow uniformity and hydraulic efficiency improvement of deep-water constructed wetlands
Ecol. Eng.
(2016)
F.E. Dierberg et al.
Relationship between hydraulic efficiency and phosphorus removal in a submerged aquatic vegetation-dominated treatment wetland
Ecol. Eng.
(2005)
C. Guo et al.
Tracer study of the hydraulic performance of constructed wetlands planted with three different aquatic plant species
Ecol. Eng.
(2017)
C. Guo et al.
Test study of the optimal design for hydraulic performance and treatment performance of free water surface flow constructed wetland
Bioresour. Technol.
(2017)
C.B. Guzman et al.
Island topographies to reduce short-circuiting in stormwater detention ponds and treatment wetlands
Ecol. Eng.
(2018)
T.R. Headley et al.
Conducting hydraulic tracer studies of constructed wetlands: a practical guide
Ecohydrol. Hydrobiol.
(2007)
J.F. Holland et al.
Effects of wetland depth and flow rate on residence time distribution characteristics
Ecol. Eng.
(2004)

V.G. Ioannidou et al.

The effects of flow rate variation and vegetation ageing on the longitudinal mixing and residence time distribution (RTD) in a full-scale constructed wetland

Ecol. Eng.

(2019)

J. Kjellin et al.

Controlling factors for water residence time and flow patterns in Ekeby treatment wetland, Sweden

Adv. Water Resour.

(2007)

L. Latrach et al.

Optimization of hydraulic efficiency and wastewater treatment performances using a new design of vertical flow Multi-Soil-Layering (MSL) technology

Ecol. Eng.

(2018)

J. Laurent et al.

Systemic models of full-scale surface flow treatment wetlands : determination by application of fluorescent tracers

Chem. Eng. J.

(2015)

O. Levenspiel et al.

Patterns of flow in chemical process vessels

Adv. Chem. Eng.

(1964)

A.Y.C. Lin et al.

Comparison of rhodamine WT and bromide in the determination of hydraulic characteristics of constructed wetlands

Ecol. Eng.

(2003)

J.J. Liu et al.

Variations of effective volume and removal rate under different water levels of constructed wetland

Ecol. Eng.

(2016)

C. Maucieri et al.

A review on the main affecting factors of greenhouse gases emission in constructed wetlands

Agric. For. Meteorol.

(2017)

M. Nuel et al.

Seasonal and ageing effects on SFTW hydrodynamics study by full-scale tracer experiments and dynamic time warping algorithms

Chem. Eng. J.

(2017)

F.R. Ouedraogo et al.

Impact of sludge layer geometry on the hydraulic performance of a waste stabilization pond

Water Res.

(2016)

J. Persson et al.

Hydraulics efficiency of constructed wetlands and ponds

Water Sci. Technol.

(1999)

Cited by (7)

Application of residence time distribution (RTD) to enhance the performance of perforated baffled constructed wetland under unsteady flow
2023, Journal of Water Process Engineering
The perforated baffle can significantly enhance the performance of the constructed wetland. In the present study, the effects of perforation shape, number of baffles, baffle position and perforation location on the flow pattern of wetlands were investigated using NaCl impulse-response tracer test, and perforated baffles were used to improve the nitrogen removal rate of the constructed wetland. The perforated baffle optimizes the flow pattern by reducing the peak height of the RTD curve and delaying the peak occurrence time. Triangle perforation exhibits better flow pattern optimization due to its smaller area. When the inlet flow rate is 100 mL/min, the mean residence time (T_m) increases by 0.25 h. The position of the perforated baffle has a more significant effect on the flow pattern, being able to better extend hydraulic residence time (HRT) when located at the height of H/2 substrate. In addition, the perforation locations of two adjacent baffles are not on the same vertical line, which can increase the horizontal water flow velocity and improve the space utilization of the wetland. Compared to the control check constructed wetland (CKCW) without baffles, the perforated baffled constructed wetland (PBCW) exhibited a higher NH₄⁺-N removal rate of 84.53 ± 7.48 % during stable operation. This is mainly because the perforated baffle improves the utilization of dissolved oxygen (DO) while improving the water flow distribution and extending the HRT. In conclusion, perforated baffles have considerable potential to alleviate shock loading and enhance the performance of constructed wetlands.
Correction of residence time distributions and hydraulic indexes affected by tracer release duration in surface flow constructed wetlands
2021, Journal of Hydrology
Citation Excerpt :
Brooks and Wise (2005) quantified the moment uncertainties of RTDs due to the random errors in tracer concentration measurements and found that the uncertainties were more sensitive to concentration error than they were to volume errors. Liu et al. (2020) screened the hydraulic indexes which were not easily disturbed by RTD errors from the perspective of numerical consistency of hydraulic indexes. Although these studies have contributed to the guidance of tracer experiments and RTD data processing, few of them involved the inaccuracy and correction of RTDs and hydraulic indexes caused by the unreasonable behavior of tracer release experiments.
In tracer experiments of surface flow constructed wetlands, residence time distributions (RTDs) are generally originated from the normalized concentration data of recovered tracers based on ideal instantaneous tracer release mode. However, this tracer release mode is commonly approximated by finite-continuous tracer release over a relatively short duration. As a result, the accuracy of RTDs and hydraulic indexes cannot be guaranteed if the RTD data and index values were still calculated based on the instantaneous release. To solve this problem, a method for correcting the original RTD data and the index values affected by tracer release duration was proposed. The correction of hydraulic indexes can be performed using analytical formulas, whereas the correction of the RTDs requires an RTD model (such as the gamma distribution and one-dimensional advection-dispersion model) or a smoothing algorithm (such as the simple moving average algorithm) to avoid the increasing effect of random measurement errors caused by the direct use of the RTD correction formulas. The correction method was successfully verified by a specially designed tracer test, and its practical application effect was exemplified. A Monte Carlo test also proved that the RTD correction owed certain robustness to tracer release instability, while the correction of some hydraulic indexes that reflect flow dispersion and comprehensive hydraulic efficiency had inferior robustness. Aside from this, the influence of tracer release duration on RTDs and hydraulic indexes without any correction was analyzed. It was found that the release duration caused the shape of the original RTDs to be more skewed in the positive direction of the x-axis and changed from “thin and high” to “fat and short”, while the duration of the release caused the positive relative errors of most indexes, which became more serious with an increase in the duration or worsening hydraulic performance. Based on these influence rules, 0.01 was defined as the threshold of the normalized release duration which needs to correct RTDs and hydraulic indexes, considering that the relative errors between the uncorrected and corrected indexes were less than 5% when the normalized duration was less than this threshold.
Spatiotemporal characteristics of hydraulic performance and contaminant transport in treatment wetlands
2021, Journal of Contaminant Hydrology
Wetlands have been proven to be efficient and cost-effective ecological treatment systems for municipal and domestic wastewater. It is essential to understand the hydrodynamic characteristics and the related contaminant transport process to optimize the treatment efficiency in free water surface wetlands. Thirty-six tracer experiments were conducted under different water depths and wetland configurations, such as installing static obstacles and dynamic disturbances. The particle image velocimetry and a novel color-concentration transform method were developed to reveal spatiotemporal flow velocity and residence time distribution. The flow fields are categorized into short-circuiting, circulation flow, dead zones, and the corresponding contaminant transport phenomena are flow advection, shear dispersion, and eddy diffusion. The flow circulation dominates the formation of the dead zone and decreases contaminant dissipation. The flow path could have effectively meandered, and the dead zone and short-circuiting could be reduced by increasing the length of the obstacles. The improved flow field is close to the plug flow, indicating enhanced hydraulic performance and treatment efficiency. The dynamic disturbance reflects the movement of fish in wetlands and provides momentum flux to promote the dissipation of pollutants in the circulation field and dead zone, alleviating the deterioration of water quality caused by pollutant accumulation. The findings of this study may provide a critical reference for the optimal design of wetlands.
Flow uniformity metrics for quantifying the hydraulic and treatment performance of constructed wetlands
2020, Ecological Engineering
Citation Excerpt :
In our cases, the correlation coefficient is 0.69, and the short-circuiting flow area is relatively small (0.01 < SC < 0.05), as shown in Fig. 3d. The significant correlation between short-circuiting flow and mixing highlights the profound effect on hydraulic performance (Liu et al., 2020), indicating the necessity of reducing short-circuiting flow (Guzman et al., 2018). We used the moment index (MI), Morril index (Mo), and the 5% normalized time (t5), which were presented by Wahl et al. (2010) and Farjood et al. (2015), to evaluate the hydraulic performance and the related short-circuiting.
The treatment efficiency of constructed wetlands is strongly related to hydraulic conditions. The occurrences of a dead zone and short-circuiting flow decrease flow uniformity and might decrease hydraulic performance. This study aims to investigate the relationship between spatial flow characteristics and wetland treatment performance. A hydrodynamic and contaminant transport model conducted ten sets of simulations with different water depths and five sets of simulations with varying numbers of emergent obstructions. Each experiment had a variant number, dimension, and allocation of obstructions. A metric of flow uniformity was proposed and discussed. We demonstrated how the dead zone and short-circuiting alter flow conditions and affect the hydraulic performance and pollutant treatment efficiency by examining the flow characteristics. The order-of-magnitude of time scales between the advective transport rate and diffusive transport rate was analyzed by calculating the Peclet number. The first-order model was used to calculate the pollutant removal ratio. The results revealed that the values of advection were approximately tens to hundreds of times those of diffusion, indicating that the flow velocity dominated the dispersion behavior. The significant improvement in flow uniformity was found when emergent obstructions were installed. The index value of the dead zone was 2 to 69 times that of the short-circuiting flow, which shows that the dead zone significantly affected the uniformity of the flow. In addition, we discovered that when the coverage ratio of the dead zone was less than 50%, the hydraulic efficiency increased from 0.55 to 1.00. Nevertheless, the short-circuiting flow was found to have minor effects on the hydraulic efficiency. The new methodology for distinguishing dead zones and short-circuiting flow areas is useful for evaluating hydraulic performance directly instead of executing water quality modeling or tracer experiments.
Clarifying the correlations between hydraulic indicators evaluating the hydraulic performance of free water surface constructed wetlands
2024, Environmental Science and Pollution Research
The development of an intelligent multilevel item bank model for the national evaluation of undergraduates
2020, Universal Journal of Educational Research

View all citing articles on Scopus

View full text

Optimal selection of hydraulic indexes with classical test theory to compare hydraulic performance of constructed wetlands

Highlights

Abstract

Introduction

Section snippets

Data sources

Results

Analysis of hydraulic efficiency indexes

Conclusions

Author contributions section

Declaration of Competing Interest

Acknowledgments

J. Hydro-Environment Res.

Ecol. Eng.

J. Hydrol.

Ecol. Eng.

Ecol. Eng.

Ecol. Eng.

Bioresour. Technol.

Ecol. Eng.

Ecohydrol. Hydrobiol.

Ecol. Eng.

Ecol. Eng.

Adv. Water Resour.

Ecol. Eng.

Chem. Eng. J.

Adv. Chem. Eng.

Ecol. Eng.

Ecol. Eng.

Agric. For. Meteorol.

Chem. Eng. J.

Water Res.

Water Sci. Technol.