Spatially-explicit water quality responses to freshwater discharge in a tropical estuary, Hilo Bay, Hawai‘i

doi:10.21203/rs.3.rs-2118273/v1

Download PDF

Research Article

Spatially-explicit water quality responses to freshwater discharge in a tropical estuary, Hilo Bay, Hawai‘i

https://doi.org/10.21203/rs.3.rs-2118273/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 27 Feb, 2023

Read the published version in Environmental Monitoring and Assessment →

You are reading this latest preprint version

Hilo Bay, Hawai’i is an estuary of great importance to its neighboring coastal community, but is threatened by impaired water quality indicated by excessive turbidity and chlorophyll a associated with river discharges of sediments and nutrients. The Wailuku River in the western half of the bay is the primary source of freshwater discharge, hypothesized here to form a surface-water dominant half of the bay with different water quality traits than the groundwater dominant, eastern half of the bay where the spring-fed Wailoa River discharges. The water quality of both halves of the bay over different flow conditions of the Wailuku River is examined in this study using spatially explicit water quality sampling which collects hundreds of samples in either half of the bay at about every 40 meters. The dense sample shows significant differences between the two halves of the bay, with greater salinity dilution and turbidity in the surface-water dominant area. Both salinity and turbidity have a predictable relation to discharge, with salinity decreasing and turbidity increasing in higher flow conditions. Chlorophyll a, however, has a more complex relation to discharge, as chlorophyll a concentrations are greatest in high-flow conditions, but this may be confounded by the different seasons of the water quality samples. Furthermore, significantly greater chlorophyll a concentrations in the groundwater dominant half of the bay in low-flow conditions shows that discharge may be spuriously correlated to chlorophyll a, and further studies of the effects of surface-water discharge on chlorophyll a concentrations are warranted.

Estuary

Hilo Bay

Water quality

Freshwater discharge

Spatial analysis

Tropical estuaries are ecologically diverse environments (Polanco et al. 2021). Because of their biodiversity and frequent proximity to coastal development, tropical estuaries are subject to a wide variety of human uses and central to many communities. It is estimated that in the Hawaiian Islands alone estuaries account for tens of millions of activity days of public use per year, which translates into hundreds of millions of dollars in economic revenue (Pendleton 2009, p. 159). Paradoxically, what makes estuaries widely used (i.e. their proximity to population centers) is what threatens their environmental quality and ecosystem functioning. Estuaries are located downstream of development which includes 40 percent of the population in the United States (Rouleau et al. 2021). The water that drains from these developed areas carries with it substances which can impair water quality and lead to events such as harmful algal blooms (Cloern et al. 2016). The nature of estuaries as receiving bodies of freshwater means that their water quality is subject to the transient effects of river discharge, the extent of which should be studied spatially.

What sets tropical estuaries apart is that they have been relatively understudied compared to their temperate counterparts (see Garcia and Barría de Cao 2018; La Colla et al. 2018; Coffin et al. 2021 for example), and they are subject to more pronounced changes due to river flows (Ringuet and Mackenzie 2005). This is because episodic rainfall events often account for a large percentage of total discharge into tropical estuaries. Combine this with the small, mountainous watersheds that occur in Hawaii, and tropical estuaries can show large water quality responses to specific discharge events (Ringuet and Mackenzie 2005; Wiegner and Mead 2009; Wiegner et al. 2013). The studies referenced and many others like them employ a spot sampling method to show a point location’s water quality response to discharge, often near the river outflow to the estuary. This approach leaves open questions of how the entire estuary responds to a discharge event, the spatial extent of that response, and how the spatial distribution of different water quality parameters relates to one another.

An estuary is organized by the location of its river outflow. This is why water quality spot sampling is often conducted near the river outflow and at a distance removed from the outflow (see Wiegner et al. 2013 for example). In this way, the effect of an outflow and how that effect dissipates at a distance is assessed. However, the change in a water quality parameter over that distance is not linear due to changes in water depth and current. For this reason, a spatially explicit water quality sampling method is required to map how a water quality parameter changes in an estuary (Wiegner et al. 2017). Such an approach shows how water quality is influenced by different physical features of an estuary, such as built structures and different sources of freshwater. Worldwide, breakwater structures are built to attenuate waves in estuaries, and in watersheds where the runoff to rainfall ratio is small, groundwater, as opposed to surface-water discharge, is the dominant source of freshwater outflow to the estuary (United States Army Corps of Engineers 2009). Both of these features are expected to have an influence on estuarine water quality.

Hilo Bay, Hawai’i presents a rare opportunity to study the water quality influence of a surface-water dominant watershed, a groundwater dominant watershed, and a breakwater in one estuary. A mountainous watershed drains the Wailuku River, whereas a comparatively flat watershed drains the Wailoa River (Mead and Wiegner 2010). The difference in terrain results in one river (i.e. the Wailuku) which is highly responsive to rainfall, and another river (i.e. the Wailoa) which is primarily fed by the groundwater of a spring (United States Army Corps of Engineers 2009). The two rivers outflow on opposite sides of Hilo Bay, and establish a surface-water dominant half and groundwater dominant half of the bay. The groundwater dominant half of the bay is encircled by a breakwater, which introduces stark water quality contrasts on either side of the breakwater (Badlowski et al. 2021). These various influences of water quality in Hilo Bay are assessed here using a spatially explicit sampling method of many hundreds of water quality samples in the surface-water dominant half, groundwater dominant half, and at various locations around the breakwater. The samples evaluate salinity as an indicator of freshwater discharge into the bay, turbidity as a measure of sediments transported in surface-water discharge, and chlorophyll a, an indicator of nutrient loading which can lead to harmful algal blooms. The objectives of the research are to show the influence of freshwater discharge on estuarine water quality and the spatial distribution of that influence in and around a breakwater and distinct areas of surface-water and groundwater discharge.

Study Area

Two rivers largely contribute to the freshwater input into Hilo Bay: the Wailuku River to the west, and the Wailoa River to the east. Both rivers are sources of surface-water for the bay, however the Wailuku is the largest river in Hawai’i and thus the largest surface-water source for Hilo Bay (Wiegner and Mead 2009). While also a source of surface-water, the Wailoa is the bay’s main source of groundwater, delivering an estimated 1.8 million cubic meters of groundwater into the bay daily (M & E Pacific 1980). With the Wailuku and the Wailoa Rivers, Hilo Bay has a main surface-water source to the west, and a main groundwater source to the east, as well as a breakwater enclosing half of the bay’s ocean outlet. For this study, the bay was separated into two zones; a surface-water dominant zone and a groundwater dominant zone. The discharge area of the Wailuku River and the outlet to the Pacific Ocean make up the surface-water dominant zone of the study area, while the discharge area of the Wailoa River, enclosed by the eastern breakwater encompasses the groundwater dominant zone (Fig. 1). By designating these two zones, the effects of freshwater discharge on water quality patterns can be differentiated between the eastern and western sides of the bay.

Water quality statistics and Pearson correlations

High-density water quality measurements of salinity, turbidity, and chlorophyll a distributed at about every 40 m were collected using a Dataflow system (Madden and Day 1992) as described in Badlowski et al. (2021) on 19 sample dates between October 31, 2008 and January 21, 2011. Despite the age of the sample, the data are still relevant in distinguishing the effects of freshwater discharge on tropical estuary water quality in Hilo Bay. The units of each water quality measurement are salinity in parts per thousand (ppt), turbidity in nephelometric turbidity units (NTU), and chlorophyll a in micrograms per liter based on a calibration curve of fluorescence (abbreviated here as ugL cal). The means of the collected data were calculated for each zone (see Fig. 1), separated by the measured water quality parameters. Each parameter had a different mean corresponding to the surface-water dominant zone and the groundwater dominant zone. Pearson correlations were calculated on these means, using the following formula:

$${r}_{pb}=\left(\sum \left(x-{m}_{x}\right)\left(y-{m}_{y}\right)\right)/\left(\sqrt{\left(\sum {\left(x-{m}_{x}\right)}^{2}\sum {\left(y-{m}_{y}\right)}^{2}\right)}\right)$$

The variable m_y is the mean of the water quality vector y, and m_x is the mean of the freshwater discharge vector x. The Pearson correlations were calculated for each measured parameter, and compared against the measured 3-day average discharge of the Wailuku River relative to the time of sampling in cubic meters per second (cms). The 3-day average was used as it approximates the generally accepted residence time of water in Hilo Bay (US Army Corps of Engineers 2009). The correlation coefficients and the p-values were generated in Python, utilizing the Scipy Stats package.

Flow rate thresholds

Discharge flow thresholds were calculated to classify the sample dates as “low-”, “average-”, or “high-flow” days, depending on the rate of flow measured from the Wailuku River at the time of sampling, in cubic meters per second. Daily-mean discharge data between the first and last sample dates were downloaded from the United States Geological Survey and converted to 3-day averages as an approximation of the residence time of water in the bay (United States Army Corps of Engineers 2009). The 3-day average discharge thresholds were equal to the 10th percentile (low flow), mean (average flow), and 90th percentile (high flow). Water quality sampling dates having the least difference to the flow thresholds were then used as representative low- (January 21, 2011), average- (June 24, 2009), and high-flow (April 14, 2009) days (see Fig. 2). The measurements for these three dates were used to compare the effects of the Wailuku River discharge on salinity, turbidity, and chlorophyll a, compared across the surface-water dominant (west) and groundwater dominant (east) zones of Hilo Bay.

Examination of low-, average-, and high-flow dates

Box and whisker plots of water quality data by zone were generated for the low-, average-, and high-flow sampling dates. The statistics of the plots followed the definition of Tukey (1977, p. 39–43) in which the horizontal line in the box is the median, the horizontal lines at either end of the box are the 25th and 75th percentiles, the whiskers (vertical lines outside the box) extend to 1.5 times the interquartile range (75th percentile minus 25th percentile), and water quality measurements above or below the extent of the whiskers appear as points. These statistics were compared by zone for differences between the surface-water dominant and groundwater dominant sides of Hilo Bay.

Cumulative distribution functions of each of the three water quality parameters were calculated for each of the flow days, between the surface-water dominant and groundwater dominant zones. A statistical comparison of cumulative distribution functions from surface-water and groundwater dominant zones was conducted using a two-sample Kolmogorov–Smirnov (KS) test. This was done as follows:

$${D}_{n,m}={sup}_{x} |{F}_{1,n}\left(x\right)-{F}_{2,m}\left(x\right)|$$

where the variables ${F}_{1,n}$ and ${F}_{2,m}$ are the empirical distribution functions of the surface-water dominant and groundwater dominant samples, respectively, and ${sup}_{x}$ is the supremum function. The results were added to the cumulative distribution figures for further analysis in which statistically significant differences between the functions, such as a p-value < 0.05, could be examined. The KS statistic and the p-values were generated in Python using the Scipy Stats package.

Cumulative distribution functions were reorganized to show differences across the three flow days for a given water quality parameter and study area zone (e.g. salinity of the surface-water dominant zone on the low-, average-, and high-flow days). The three functions shown in these figures required two different two-sample KS tests. The first test compared the low-flow sample to the average-flow sample, and the second test compared the average-flow sample to the high-flow sample. Thus, differences in low- and high-flow days were examined relative to average-flow conditions (e.g. does the high-flow day differ from average?).

A Welch’s t-test of the two study area zones was calculated using the following formulas:

$t=\frac{\varDelta \stackrel{-}{X}}{{S}_{\varDelta \stackrel{-}{X}}}=\frac{{\stackrel{-}{X}}_{1-}{\stackrel{-}{X}}_{2}}{\sqrt{{s}_{{\stackrel{-}{X}}_{1}}^{2}+{s}_{{\stackrel{-}{X}}_{2}}^{2}}}$ and ${s}_{{\stackrel{-}{X}}_{i}}=\frac{{s}_{i}}{\sqrt{{N}_{i}}}$ .

The variables ${\stackrel{-}{X}}_{i}$ and ${s}_{{\stackrel{-}{X}}_{i}}$ are the ith sample mean and its standard error, respectively, for a given sample standard deviation ${s}_{i}$ and sample size ${N}_{i}$. The Welch’s t-test statistic was calculated between the surface-water dominant zone and the groundwater dominant zone for each water quality parameter, separated by the flow days. Both the t-test statistic and the p-value were generated for these tests in Python, utilizing the Scipy Stats package.

Water quality relative to discharge

When comparing the three water quality parameters to the discharge 3-day average, salinity had a statistically significant negative correlation in both the surface-water dominant zone and the groundwater dominant zone, with salinity concentration decreasing as discharge flow increased and little change in this pattern between the two zones (Fig. 3). For turbidity, there was a slightly positive trend relative to discharge flow on the surface-water dominant side, and little indication of any trend on the groundwater dominant side. Neither correlation was significant, but the plots do indicate that in both zones turbidity increased with discharge up to about 2 cms, above which turbidity shows little response to increases in discharge. For chlorophyll a, there was a positive correlation to the discharge flow in both of the zones, though only the correlation of the groundwater side was significant and chlorophyll a of the surface-water side decreased at discharges greater than about 2 cms, suggesting chlorophyll a and discharge follow a quadratic form in the surface-water dominant zone.

Water quality of low-, average-, and high-flow dates

Water quality parameters were generally higher in the surface-water dominant zone compared to the groundwater zone, across the low-, average-, and high-flow dates (Fig. 4). Salinity was greater in the surface-water side than the groundwater side, except for the date of high flow, when salinity in the surface-water side was reduced to a level relatively similar to salinity in the groundwater side. Statistical tests showed significant differences in salinity between the two zones on low- and average-flow dates, but that difference was not present on the high-flow date, when salinity was relatively similar throughout the bay (Table 1).

Turbidity was greater in the surface-water side of the bay compared to the groundwater side, and this was particularly evident on the high-flow date (Fig. 4). The median and interquartile range of turbidity on the average-flow date are visibly greater in the surface-water side than the groundwater side, but that difference is not visible on the low-flow date, when the difference in turbidity between the two sides of the bay was the least. The statistical tests of the two sides of the bay returned significant differences in turbidity, regardless of the date, with a greater level of turbidity in the surface-water side of the bay on low-, average-, and high-flow dates (Table 1).

Chlorophyll a, like the other water quality parameters, was greater in the surface-water side of the bay compared to the groundwater side, with the exception of the low-flow date, when chlorophyll a was largely not present in the surface-water side of the bay (Fig. 4). On the average-flow date, the surface-water side of the bay had an average chlorophyll a concentration nearly double that of the groundwater side (Table 1). Chlorophyll a concentrations increased throughout the bay on the high-flow date, but the surface-water side had a smaller range of greater concentrations than the groundwater side of the bay (Fig. 4), resulting in a statistically significant difference (Table 1).

Table 1

The means for each zone, the Welch’s t-test statistic, and the t-test p-value for each water quality parameter are listed by flow day. Statistically significant p-values < 0.05 are denoted by an asterisk (*), and p-values below 0.01 are written as “p-value < 0.01”
Parameter	Statistic	Low-flow (2011-01-21)	Average-flow (2009-06-24)	High-flow (2009-04-14)
Salinity
	Surface-water Zone Mean	31.9	23.5	16.1
	Groundwater Zone Mean	30.2	21.6	18.7
	T-test Statistic	15.06	16.53	-1.04
	P-value	p-value < 0.01*	p-value < 0.01*	0.30
Turbidity
	Surface-water Zone Mean	0.4	1.3	3.2
	Groundwater Zone Mean	0.3	0.9	1.0
	T-test Statistic	-2.27	13.97	36.62
	P-value	0.02*	p-value < 0.01*	p-value < 0.01*
Chlorophyll a
	Surface-water Zone Mean	0.3	2.5	10.5
	Groundwater Zone Mean	1.3	1.4	9.1
	T-test Statistic	-8.78	31.17	12.45
	P-value	p-value < 0.01*	p-value < 0.01*	p-value < 0.01*
*Significant difference between zones (p-value < 0.05)

The proportion of observations at a given water quality concentration are shown in cumulative distribution functions comparing the two zones of Hilo Bay (Fig. 5). Water quality concentrations are visibly greater in the surface-water dominant zone in most cases, with cumulative distribution functions set to the right of their companions from the groundwater dominant zone. This is the case in (1) salinity concentrations of the low- and average-flow dates, (2) turbidity concentrations of the average- and high-flow dates, and (3) chlorophyll a concentrations of the average- and high-flow dates, all of which have significantly different distributions. Deviations from this general pattern include (1) salinity concentrations in the groundwater zone that exceed those in the surface-water zone on the high-flow date, (2) turbidity concentrations in both zones that are largely less than 1 NTU on the low-flow date, and (3) chlorophyll a concentrations in the groundwater zone greater than those of the surface-water zone on the low-flow date.

The cumulative distribution functions are reorganized for a visual comparison of each water quality parameter across the different flow dates (Fig. 6). Low- and high-flow dates have cumulative distribution functions that are significantly different from the average-flow date according to KS statistical tests (not shown). The visual comparison of cumulative distribution functions shows the trend of salinity increasing throughout Hilo Bay on days of lower flow from the Wailuku River. The distributions of salinity, however, follow notably different patterns in the surface-water dominant zone versus the groundwater dominant zone. In the former, salinity is largely centered around the median, whereas the variation in salinity around the median is greater in the groundwater dominant zone. Turbidity, like salinity, also follows a trend relative to Wailuku River flow in which turbidity increases throughout the bay on days of higher flow. However, there is a greater increase in turbidity on days of higher flow in the surface-water dominant zone compared to the groundwater dominant zone. A similar pattern of increasing chlorophyll a concentrations on days of higher flow exists throughout the bay, and both zones of the bay follow similar distributions on each respective day, with (1) a large proportion of observations near zero on the low-flow date, (2) little variation around the median on the average-flow date, and (3) large variation around the median on the high-flow date.

The cumulative distribution functions have the advantage that they can be used to assess the proportion of sample points that exceed a water quality standard, and these points can then be mapped (Fig. 7). Water quality standards of Hilo Bay are that turbidity and chlorophyll a should not exceed 3 NTU and 4.5 ug/L, respectively, more than ten percent of the time (Clean Water Branch 2014). Although the temporal component of this standard cannot be evaluated here, the proportion of points exceeding the standard were mapped to show where water quality standards were not met on low-, average-, and high-flow days. Elevated levels of turbidity that exceeded the water quality standard mainly occurred in the surface-water dominant zone, and accounted for 82%, 71%, and 94% of sample points that exceeded the standard on low-, average-, and high-flow days, respectively. The remaining percentage of sample points that exceeded the turbidity water quality standard occurred in the groundwater dominant zone directly at the mouth of the Wailoa River and beyond the breakwater in the Pacific Ocean. The water quality standard of chlorophyll a was more equally exceeded in the surface-water and groundwater dominant zones, including the average-flow day when the standard was essentially not exceeded (except for two of 1,658 sample points), and the high-flow day when the standard was exceeded almost entirely inside the breakwater. The low-flow day had select locations in both zones that exceeded the chlorophyll a standard, including the mouth of the Wailuku River in the surface-water zone and parts of Reeds Bay west of the breakwater in the groundwater zone.

This study spatially organizes an estuary and illustrates how estuarine water quality varies in surface-water and groundwater dominant discharge areas using a spatially explicit sampling technique (Madden and Day 1992). The concept of spatially organizing an estuary is what sets this study apart from other studies, most of which employ point-specific spot sampling (see Wiegner et al. 2013 for example), and therefore, cannot examine how estuarine water quality is spatially organized. Other studies that have employed the high-density sampling methodology of Madden and Day (1992) have observed spatial patterns in estuarine water quality (Badlowski et al. 2021), which led to the design of the current study. In Hilo Bay, a plume of shallow surface-water discharge in high-flow conditions is observed exiting the western half of the bay from the Wailuku River, particularly in the form of turbidity, whereas that same plume is largely absent from the eastern half of the bay (Fig. 7). Further evidence of surface-water discharge exiting the western half of the bay can be seen in a spatially interpolated rendering of water quality sampling from October 16, 2009 (Fig. 8), derived following the interpolation of Badlowski et al. (2021). The date is one of moderately low flow (Fig. 2) in which surface-water discharge from the Wailuku River can be seen hugging the western shore of the bay, as evident in diluted salinity concentrations and elevated chlorophyll a concentrations. Also of note are low salinity concentrations in Reeds Bay due to groundwater discharge (Wiegner and Mead 2009). The lower temperatures and dissolved oxygen, collected in the same water quality survey, are also evidence of groundwater discharge into Reeds Bay (Wiegner and Mead 2009; Wiegner et al. 2017). The prevalence of surface-water discharge in the western half of the bay and groundwater discharge in the eastern half of the bay ultimately leads to the spatial organization of water quality in Hilo Bay.

The spatial organization of water quality evident in Hilo Bay may be observed in other estuaries around the world. For instance, a spatially explicit sampling methodology may distinguish between a maximum turbidity zone and oxygen minimum zone important in the carbon cycling of the Elbe estuary in Germany and other similar estuaries in Europe (Amann et al. 2012). Rivers are often disaggregated into tidal and non-tidal reaches in water quality studies (see Hoang et al. 2018; Post et al. 2018; Tiwari et al. 2022 among others). Estuarine water quality studies could benefit from a similar design. In the largest estuary of the United States, the different salinity zones of the Chesapeake Bay relate to various water quality and biological properties of the bay (Harding et al. 2019) and illustrate the importance of spatially organizing an estuary for the monitoring and management of water quality, as was the historic practice some decades ago in the Delaware estuary (Albert 1988).

The large water quality response of Hilo Bay to high-flow conditions (see Figs. 4, 6, and 7) is typical of a tropical estuary. High-flow events, although often short in duration, can account for up to 80% of total annual sediment and nutrient loads (Milliman and Syvitski 1992; Eyre 1995). The high-flow conditions studied here brought with it a large load of sediment and nutrients as indicated by elevated levels of turbidity and chlorophyll a, respectively. In a similarly focused study of Kāne‘ohe Bay, Hawaii, high-flow events carried sediment and nutrients in the form of a large plume far exceeding the extent of that in baseflow conditions concentrated only at the mouth of river outflows (Ringuet and Mackenzie 2005). This study exposes a similar dynamic of a large plume in high-flow conditions and small pockets of sediment and nutrients in lower flow conditions (Fig. 7), except instead of strategically locating spot sampling locations, a spatially explicit sampling approach is employed to reinforce the findings of previous studies. Spot sampling has shown the water quality response of Hilo Bay at distances near and far from the mouth of the Wailuku River (Wiegner and Mead 2009; Mead and Wiegner 2010; Wiegner et al. 2013), and now evidence exists of different responses between the surface-water dominant, western half of the bay and the groundwater dominant, eastern half of the bay.

There are clear and significant differences in the water quality between the surface-water and groundwater dominant halves of Hilo Bay (Table 1 and Fig. 5). Salinity, turbidity, and chlorophyll a samples varied depending on their location in the western half of the bay that receives surface-water discharge from the Wailuku River or the eastern half of the bay that receives groundwater discharge from the spring-fed Wailoa River (United States Army Corps of Engineers 2009). The difference was such that no matter the flow conditions (i.e. low, average, and high), the three water quality parameters were larger in the surface-water dominant half of the bay. The lone exceptions to this were the salinity of high-flow conditions and the chlorophyll a of low-flow conditions. In high-flow conditions, the salinity of the surface-water dominant half of the bay is diluted and less than that of the groundwater dominant half of the bay. This pattern of dilution and lower salinity following high-flow events in Hilo Bay has been observed before (Wiegner et al. 2013). What is a new finding here is that the dilution of surface-water discharge is concentrated in the western half of the bay (Fig. 8). This shows that surface-water discharge from the Wailuku is funneled out of the western half of the bay due to the force of the tide and strong winds (United States Army Corps of Engineers 2009), whereas the water trapped behind the breakwater in the eastern half of the bay has relatively greater salinity (Silvius et al. 2005). In terms of chlorophyll a, freshwater discharge carrying nutrients may collect behind the breakwater in low-flow conditions (United States Army Corps of Engineers 2009), allowing phytoplankton to grow in the groundwater dominant half of the bay independent of surface-water discharge (Venkataramana et al. 2017).

Salinity decreased with Wailuku River discharge (Table 1 and Figs. 3, 4, and 6), as has been observed in previous studies of Hilo Bay (Wiegner and Mead 2009; Mead and Wiegner 2010; Wiegner et al. 2013). The freshwater delivered by the Wailuku River can vary by more than an order of magnitude between low- and high-flow conditions (Fig. 2), and in the process, dilutes salinity in the bay by about one half. The response of salinity to discharge, namely its dilution, was strongest in the surface-water dominant half of the bay compared to its groundwater dominant counterpart. This is because surface-water discharge from the Wailuku River travels north and exits the western half of the bay (Fig. 8), as has been shown in previous water quality modeling (United States Army Corps of Engineers 2009). Previous spot sampling studies have speculated that the variation of salinity increases in high-flow conditions and that this retards phytoplankton growth and lowers chlorophyll a concentrations (see Wiegner and Mead 2009 for example). However, due to their limited sample size, spot sampling studies have not been able to confirm this. The present study confirms that salinity varies more in high-flow conditions, but that does not correspond to lower chlorophyll a concentrations (Fig. 6).

Turbidity is a measure of water clarity and suspended sediments in the water column, which increased with discharge in this study (Table 1 and Figs. 4, 6, and 7). The effect of discharge on turbidity was greatest in the western half of Hilo Bay where the Wailuku River outflows. The Wailuku River drains a mountainous watershed and carries with it up to ten times more sediment than its neighbor to the east, the Wailoa River (Wiegner and Mead 2009). This is thought to be the case because of differences in watershed size, relief, and land cover (Mead and Wiegner 2010). The Wailuku River watershed is the largest in the state of Hawaii, characterized by extreme relief (3,500-m headwater elevation) compared to the relatively gentle relief of the Wailoa River watershed (750-m headwater elevation). Combine this with a nearly six times greater percentage of barren land coverage in the Wailuku River watershed (https://hawaiiwatershedatlas.com/), and it makes sense that the Wailuku River carries far more sediment than the Wailoa River. This also explains why the western half of Hilo Bay had significantly greater turbidity in this study (Table 1), as well as the vast majority of sample measurements exceeding the turbidity water quality standard (Fig. 7).

Several studies have linked salinity and turbidity conditions of Hilo Bay and other Hawaiian coastal waters to microbial pollution. Wiegner et al. (2017) found the fecal indicator bacterium, Enterococcus, related to salinity (-), water temperature (+), and turbidity (+) in Hilo Bay, with similar relationships between salinity, water temperature, and Clostridium perfringens also observed. Economy et al. (2019) identified Staphylococcus aureus and fecal indicator bacteria in Hilo Bay and associated rivers and watershed soils, finding negative and positive correlations with salinity and turbidity, respectively. Both of these studies draw linkages between storm conditions and elevated risk of microbial pollution in Hilo Bay. Considering these relationships between microbial pollutants, salinity, and turbidity results from the present study suggest that the western (surface-water dominant) portion of Hilo Bay is most susceptible to stormwater-induced microbial pollution.

Turbidity is expected to be related to chlorophyll a in that suspended sediments limit light, phytoplankton growth, and chlorophyll a concentrations (Wiegner and Mead 2009). At first glance, this does not appear to be the case in this study since chlorophyll a increases overall with turbidity (Table 1). However, this demonstrates why a spatially explicit analysis is important. In a previous analysis of spatially explicit samples, it was shown that significantly high measurements of turbidity and chlorophyll a were spatially offset (Badlowski et al. 2021). A plume of elevated turbidity exiting the Wailuku River displaced elevated chlorophyll a concentrations and presumably pushed phytoplankton toward the breakwater. This spatial pattern is evident in this study in comparing the high turbidity of the surface-water dominant half of the bay at the mouth of the Wailuku River and similarly high chlorophyll a measurements of the groundwater dominant half of the bay near the breakwater (compare high-flow measurements in Figs. 4 and 6). Thus, it appears that once again the highest measurements of turbidity and chlorophyll a are at locations offset from one another. However, this does not preclude the comingling of sediment and phytoplankton in the plume of surface-water discharge exiting the Wailuku River (see the high-flow conditions in Fig. 7).

Chlorophyll a concentrations were significantly higher in the surface-water dominant half of the bay than in the groundwater dominant half in high-flow conditions (Table 1 and Fig. 5), which is evidence contrary to the hypothesis of surface-water discharge plumes limiting light and phytoplankton growth (Wiegner and Mead 2009). However, it could be that phytoplankton was growing in the bay prior to the high-flow conditions. In this hypothesis, the Wailuku River supplies a steady flow of nutrients, even in low-flow conditions (Wiegner et al. 2017). Phytoplankton then have the potential to grow in the outflows of the Wailuku River. At the onset of high-flow conditions, phytoplankton mix with the sediment of the outflowing plume, and some days later (e.g. three to eight days) the phytoplankton population and chlorophyll a concentrations recede (Ringuet and Mackenzie 2005), either due to light limitations (Wiegner and Mead 2009) or because the phytoplankton have been expelled from the bay (United States Army Corps of Engineers 2009).

Chlorophyll a concentrations are not wholly dependent on river flows (Figs. 3 and 7). Nutrients are delivered to the bay in both low- and high-flow conditions (Wiegner et al. 2017). Therefore, phytoplankton have the potential to grow and elevate chlorophyll a concentrations regardless of flow conditions. In fact, phytoplankton may grow the greatest in low-flow periods of summer (Wiegner and Mead 2009). This is because phytoplankton growth and chlorophyll a concentrations follow the seasons in which summer is a peak growing season (Hoang et al. 2018), and phytoplankton may be expelled from Hilo Bay in high flows (United States Army Corps of Engineers 2009). Low-flow sampling of this study was conducted in the winter, and it may very well be that chlorophyll a concentrations of the low-flow conditions would have been greater if the sampling had been conducted in the summer growing season. It could also be that the elevated chlorophyll a concentrations of the average- and high-flow conditions in this study (Fig. 4) are a function of the season, since both of those samples were collected during the phytoplankton growing season (Wiegner and Mead 2009).

The spatial organization of water quality in an estuary is illustrated in this study using a spatially explicit and dense sampling technique. In the tropical estuary of Hilo Bay, water quality is organized in a surface-water dominant area around the mouth of the Wailuku River, and a groundwater dominant area where the effects of river flows on water quality parameters are less pronounced. Water quality parameters follow patterns of surface-water discharge from the Wailuku River in which higher flows dilute salinity and increase turbidity. Turbidity is a function of discharge, and it is significantly greater in the surface-water dominant part of the bay than in the groundwater dominant area. Salinity is diluted the most in the surface-water discharge part of the bay, and it varies more with higher flows, but this does not translate to lower chlorophyll a concentrations as hypothesized in previous studies. Rather, in this study, chlorophyll a concentrations are greatest in high-flow conditions with the greatest variation in salinity. Chlorophyll a concentrations show a relation to discharge in this study, but this may be due to the different seasons of the water quality samples. There is evidence of elevated chlorophyll a concentrations during a high-flow event in this study, but the effect of the surface-water discharge plume, and its elevated turbidity, needs to be studied more with repeat sampling in the days following the outflow of the plume. This should be done using a spatially explicit sampling method as in this study to show the spatial distribution of the plume, its elevated turbidity, and chlorophyll a concentrations in and around the plume.

Acknowledgments

The authors would like to acknowledge University of Hawai‘i Hilo student Judy K. Walker for assistance in collecting the data used in this study, as well as Dr. Tracy N. Wiegner and the University of Hawai‘i Hilo Marine Science department for equipment and boat support that made this project possible. Monmouth University allocated funds to support the work of Sydney N. Lucas on this research. Gina A. Badlowski, a graduate of Monmouth University, is also a valuable member of the team in contributing to a previous publication which informed the direction of this work.

Funding: University of Hawai‘i Hilo Marine Science department funds supported boat time and equipment for data collection. Monmouth University’s Geographic Information Systems program funded the graduate assistant position of Sydney N. Lucas to complete this work.

Conflicts of interest / Competing interests: The authors report no conflicts of interest and no competing interests relevant to this study.

Availability of data and materials: Data presented in this study are available from the authors upon request.

Code availability: Not applicable

Author contributions:

As defined in CRediT Contributor Roles Taxonomy (https://credit.niso.org/)

Lucas (Formal analysis; Investigation; Writing – original draft; Writing – review and editing)

Fouad (Conceptualization; Methodology; Validation; Writing – original draft; Writing – review and editing; Visualization; Supervision; Project administration; Funding acquisition)

Adolf (Methodology; Validation; Resources; Data curation; Writing – review and editing; Supervision; Funding acquisition)

Albert, R. C. (1988). The historical context of water quality management for the Delaware estuary. Estuaries, 11(2), 99–107. https://doi.org/10.2307/1351997.

Amann, T., Weiss, A., & Hartmann, J. (2012). Carbon dynamics in the freshwater part of the Elbe estuary, Germany: Implications of improving water quality. Estuarine, Coastal and Shelf Science, 107(10), 112–121. https://doi.org/10.1016/j.ecss.2012.05.012.

Badlowski, G. A., Adolf, J. E., & Fouad, G. (2021). Spatial analysis of water quality parameters in Hilo Bay, Hawai'i, using a combination of interpolated surfaces and hot spot analysis. Environmental Monitoring and Assessment, 193, 118. https://doi.org/10.1007/s10661-021-08894-6.

Clean Water Branch. (2014). Water quality standards. Hawai‘i: State of Hawai‘i, Department of Health. Available at https://health.hawaii.gov/cwb/files/2013/04/Clean_Water_Branch_HAR_11-54_20141115.pdf.

Cloern, J. E., Abreu, P. C., Carstensen, J., Chauvaud, L., Elmgren, R., Grall, J., Greening, H., Johansson, J. O. R., Kahru, M., Sherwood, E. T., Xu, J., & Yin, K. (2016). Human activities and climate variability drive fast-paced change across the world’s estuarine-coastal ecosystems. Global Change Biology, 22(2), 513–529. https://doi.org/10.1111/gcb.13059.

Coffin, M., Knysh, K., Roloson, S., Pater, C. C., Theriaul, E., Cormier, J. M., Courtenay, S. C., & van den Heuvel, M. R. (2021). Influence of nutrient enrichment on temporal and spatial dynamics of dissolved oxygen within northern temperate estuaries. Environmental Monitoring and Assessment, 193, 804. https://doi.org/10.1007/s10661-021-09589-8.

Economy, L. E., Wiegner, T. N., Strauch, A. M., Awaya, J. D., & Gerken, T. (2019). Rainfall and streamflow effects on estuarine Staphylococcus aureus and fecal indicator bacteria concentrations. Journal of Environmental Quality, 48(6), 1711–1721. https://doi.org/10.2134/jeq2019.05.0196.

Eyre, B. (1995). A first-order nutrient budget for the tropical Moresby estuary and catchment, North Queensland, Australia. Journal of Coastal Research, 11(3), 717–732.

Garcia, M. D., & Barría de Cao, M. S. (2018). Anthropogenic pollution along the coast of a temperate estuary: effects on tintinnid assemblages. Hydrobiologia, 809, 201–219. https://doi.org/10.1007/s10750-017-3465-z.

Harding, L. W., Mallonee, M. E., Perry, E. S., Miller, W. D., Adolf, J. E., Gallegos, C. L., & Paerl, H. W. (2019). Long-term trends, current status, and transitions of water quality in Chesapeake Bay. Scientific Reports, 9, 6709. https://doi.org/10.1038/s41598-019-43036-6.

Hoang, H. T. T., Duong, T. T., Nguyen, K. T. Le, Q. T. P., Luu, M. T. N., Trinh, D. A., Le, A. H., Ho, C. T., Dang, K. D., Némery, J., Orange, D., & Klein, J. (2018). Impact of anthropogenic activities on water quality and plankton communities in the Day River (Red River Delta, Vietnam). Environmental Monitoring and Assessment, 190, 67. https://doi.org/10.1007/s10661-017-6435-z.

La Colla, N. S., Botté, S. E., Negrin, V. L., Serra, A. V., & Marcovecchio, J. E. (2018). Influence of human-induced pressures on dissolved and particulate metal concentrations in a South American estuary. Environmental Monitoring and Assessment, 190, 532. https://doi.org/10.1007/s10661-018-6930-x.

Madden, C. J, & Day, J. W. (1992). An instrument system for high-speed mapping of chlorophyll a and physico-chemical variables in surface waters. Estuaries, 15, 421–427. https://doi.org/10.2307/1352789.

M & E Pacific. (1980). Hilo area comprehensive study—geological, biological, and water quality investigations of Hilo Bay. Hawai‘i: United States Army Corps of Engineers, Honolulu District.

Mead, L. H., & Wiegner, T. N. (2010). Surface water metabolism potential in a tropical estuary, Hilo Bay, Hawai‘i, USA, during storm and non-storm conditions. Estuaries and Coasts, 33, 1099–1112. https://doi.org/10.1007/s12237-010-9317-y.

Milliman, J. D., & Syvitski, J. P. M. (1992). Geomorphic/tectonic control of sediment discharge to the ocean: The importance of small mountainous rivers. The Journal of Geology, 100(5), 525–544.

Pendleton, L. H. (2009). The economic value of coastal and estuary recreation. In L. H. Pendleton (Ed.), The economic and market value of coasts and estuaries: What’s at stake?. Arlington, Virginia: Restore America’s Estuaries.

Polanco, A., Mutis Martinezguerra, M., Marques, V., Villa-Navarro, F., Borrero Pérez, G. H., Cheutin, M., Dejean, T., Hocdé, R., Juhel, J., Maire, E., Manel, S., Spescha, M., Valentini, A., Mouillot, D., Albouy, C., & Pellissier, L. (2021). Detecting aquatic and terrestrial biodiversity in a tropical estuary using environmental DNA. Biotropica, 53, 1606–1619. https://doi.org/10.1111/btp.13009.

Post, C. J., Cope, M. P., Gerard, P. D., Masto, N. M., Vine, J. R., Stiglitz, R. Y., Hallstrom, J. O., Newman, J. C., & Mikhailova, E. A. (2018). Monitoring spatial and temporal variation of dissolved oxygen and water temperature in the Savannah River using a sensor network. Environmental Monitoring and Assessment, 190, 272. https://doi.org/10.1007/s10661-018-6646-y.

Ringuet, S., & Mackenzie, F. T. (2005). Controls on nutrient and phytoplankton dynamics during normal flow and storm runoff conditions, southern Kaneohe Bay, Hawaii. Estuaries, 28, 327–337. https://doi.org/10.1007/BF02693916.

Rouleau, T., Colgan, C. S., Adkins, J., Castelletto, A., Dirlam, P., Lyons, S., & Stevens, H. (2021). Why estuaries matter. In The economic value of America’s estuaries: 2021 update. Washington, DC: Restore America’s Estuaries.

Silvius, K., Moravcik, P., & James, M. (2005). Hilo Bay watershed-based restoration plan. Hawai‘i: State of Hawai‘i, Department of Health. Available at https://eha-cloud.doh.hawaii.gov/cwb/api/documents/20/download.

Tiwari, N. K., Das Gupta, S., Swain, H. S., Jha, D. N., Samanta, S., Manna, R. K., Das, A. K., & Das, B. K. (2022). Water quality assessment in the ecologically stressed lower and estuarine stretches of river Ganga using multivariate statistical tool. Environmental Monitoring and Assessment, 194, 469. https://doi.org/10.1007/s10661-022-10007-w.

Tukey, J. W. (1977). Schematic summaries. In F. Mosteller (Ed.), Exploratory data analysis. Reading, Massachusetts: Addison-Wesley Publishing Company.

United States Army Corps of Engineers. (2009). Hilo Bay water circulation and water quality study. Hawai‘i: United States Army Corps of Engineers, Honolulu District.

Venkataramana, V., Sarma, V. V. S. S., & Matta Reddy, A. (2017). River discharge as a major driving force on spatial and temporal variations in zooplankton biomass and community structure in the Godavari estuary India. Environmental Monitoring and Assessment, 189, 474. https://doi.org/10.1007/s10661-017-6170-5.

Wiegner, T. N., Edens, C. J., Abaya, L. M., Carlson, K. M., Lyon-Colbert, A., & Molloy, S. L. (2017). Spatial and temporal microbial pollution patterns in a tropical estuary during high and low river flow conditions. Marine Pollution Bulletin, 114(2), 952–961. https://doi.org/10.1016/j.marpolbul.2016.11.015.

Wiegner, T. N., & Mead, L. H. (2009). Water quality in Hilo Bay, Hawaii, U.S.A., under baseflow and storm conditions. Hawai‘i: University of Hawai‘i at Hilo. Available at https://kohalacenter.org/archive/himoes/pdf/HiloBayFinalReport2009.pdf.

Wiegner, T. N., Mead, L. H., & Molloy, S. L. (2013). A comparison of water quality between low- and high-flow river conditions in a tropical estuary, Hilo Bay, Hawaii. Estuaries and Coasts, 36, 319–333. https://doi.org/10.1007/s12237-012-9576-x.

No competing interests reported.

Download PDF

Journal Publication

published 27 Feb, 2023

Read the published version in Environmental Monitoring and Assessment →

Editorial decision: Major revision
12 Jan, 2023
Reviews received at journal
28 Oct, 2022
Reviewers agreed at journal
16 Oct, 2022
Reviewers invited by journal
14 Oct, 2022
Editor assigned by journal
03 Oct, 2022
Submission checks completed at journal
03 Oct, 2022
First submitted to journal
29 Sep, 2022

You are reading this latest preprint version

Spatially-explicit water quality responses to freshwater discharge in a tropical estuary, Hilo Bay, Hawai‘i

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Conclusions

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1