Estuaries and Coasts

, Volume 37, Supplement 1, pp 164–179

Extreme Eutrophication in Shallow Estuaries and Lagoons of California Is Driven by a Unique Combination of Local Watershed Modifications That Trump Variability Associated with Wet and Dry Seasons

Authors

    • Department of Ecology and Evolutionary BiologyUniversity of California
  • Peggy Fong
    • Department of Ecology and Evolutionary BiologyUniversity of California
Article

DOI: 10.1007/s12237-013-9687-z

Cite this article as:
Kennison, R.L. & Fong, P. Estuaries and Coasts (2014) 37: 164. doi:10.1007/s12237-013-9687-z

Abstract

Rapidly growing human populations have caused heavy modifications to the watersheds of many Mediterranean climate estuaries, subjecting them to excessive nutrient enrichment and harmful macroalgal blooms. Despite these impacts, comprehensive studies in these systems are rare and comparisons between systems are lacking. We surveyed five southern California estuaries that ranged in size from 93 to 1,000 ha and incorporated differing land usages and watershed sizes. We sampled environmental variables (sediment redox potential, organic content, total nitrogen and total phosphorus, water column nitrate, ammonium, and salinity) and macroalgal cover and biomass quarterly at three locations within each estuary over 15 months to compare spatial and wet vs. dry season patterns. Maximum mean water column nitrate concentration across all estuaries ranged from 47 to 1,700 μM, showing that all estuaries were highly enriched with nitrogen, at least at some times. Mean macroalgal biomass ranged from 0 to 1,500 g wet wt m−2. However, neither nutrient concentrations nor algal biomass showed consistent seasonal patterns as maximum values occurred in different seasons in different estuaries. Three-dimensional principal components analysis followed by regression analyses confirmed that macroalgal abundance was not directly related to water or sediment N concentrations. Rather each of these southern California estuaries showed individual patterns in all measured variables, which were most likely induced by a suite of physical modifications unique to each system and its watershed.

Keywords

EutrophicationMacroalgal bloomsNutrient enrichmentWatershed modificationShallow Mediterranean climate estuaries

Introduction

Shallow lagoonal estuaries typical of Mediterranean climates are arguably both the most vulnerable to human impacts and the least studied estuarine ecosystems worldwide (Fong and Kennison 2010). The repercussions of human impacts may be especially dramatic in Mediterranean climate estuaries due to their unique physical setting, seasonality, and generally large human populations. Many coastal watersheds in Mediterranean climates are undergoing rapid population growth that contributes to modification of hydrology, urbanization, habitat fragmentation, and impaired water quality (Castro et al. 2007; Carlier et al. 2008). In Mediterranean climates, distinct wet–dry seasonal cycles produce ephemeral river inflows that may carry excessive nutrients from developed watershed into these systems. Although there are fewer studies in Mediterranean climate estuaries, it has been widely documented in temperate systems that highly modified watersheds produce higher nutrient loads, often resulting in elevated levels of water column nutrients and eutrophic conditions (e.g., Valiela et al. 1997; Howarth and R. Marino 2006; Paerl 2006; Tett et al. 2007). The few Mediterranean climate systems that have been well-studied, such as Venice Lagoon, Italy (see Sfriso et al. 2005 for review), have shown that nutrient enrichment and eutrophication are severe (Viaroli et al. 1996; Hernandez et al. 1997; Menendez and Comin 2000; Astill and Lavery 2001), creating conditions that threaten ecosystem health in other more well-studied systems (Kennish 2002; Bricker et al. 2008). Thus, there is a compelling need for more information on the current nutrient and eutrophic status of Mediterranean climate estuaries.

In many shallow estuaries, nutrient enrichment results in eutrophication in the form of excessive macroalgal blooms often composed of opportunistic green algae in the genus Ulva (Morand and Merceron 2005). Generally, in shallow estuarine systems, where the photic zone encompasses the benthos, macroalgae tend to limit the success of phytoplankton, particularly in highly enriched systems (Fong et al. 1993a; Marcomini et al. 1995). Many field surveys have sought to establish correlations between biomass of bloom-forming macroalgae and spatial or seasonal patterns of water column (Sfriso et al. 1987; Raffaelli et al. 1989; Valiela et al. 1992; Pihl et al. 1996; Hernandez et al. 1997; Flindt et al. 1997; Menendez and Comin 2000; Nelson et al. 2003) and sediment (Sfriso et al. 1992) nutrient concentrations, but found that describing these relationships is far from straightforward. However, these initial investigations led to more mechanistic studies including assessing the physiological response of the algae, as well as secondary feedback mechanisms such as internal nutrient cycling, grazing, hydrodynamics, and processes affecting early life stages (see Schramm 1999 for review and Cloern 2001). Ultimately, researchers found that estuaries have multiple nutrient supplies, including inputs from groundwater (Valiela et al. 1992; Conley et al. 2000), aerial deposition (Paerl 1997; Paerl et al. 2002), and storage in the sediments (Trimmer et al. 2000), and differences between systems are made more complex by a variety of additional factors such as river flow, wastewater discharge, seasonal changes in inputs, and a diverse mix of land uses. Therefore, estimates of total nutrient loading may be needed in order to accurately assess the relationship between nutrients and biomass. Quantifying loading rates is a complex and comprehensive process that has been accomplished in several well-studied estuaries, including some in Denmark (Conley et al. 2000), Portugal (Lillebø et al. 2005), and the East Coast of the USA (Lee and Olsen 1985; Hunchak-Kariouk and Nichols 2001; Latimer and Rego 2010; Kinney and Valiela 2011; Giordano et al. 2011), including extensive studies in Waquoit Bay (Valiela et al. 1992, 2000; Lyons et al. 1995; Hauxwell et al. 1998; Fox et al. 2008). However, quantifying nutrient loading is costly and often not fiscally realistic while measuring nutrient concentrations can provide an important first step, particularly when management decisions need to be made in a time-sensitive manner. Thus, in understudied Mediterranean climate estuaries, a logical first step to understanding eutrophication status is to quantify spatial and wet vs. dry season patterns of water column nutrient concentrations and macroalgal abundance. This will provide an initial evaluation of eutrophic conditions and help to provide focus for more mechanistic studies.

In shallow estuaries (<3 m), benthic–pelagic coupling may be extremely strong (Grenz et al. 2000; Sundbäck et al. 2003; Engelsen et al. 2008), making knowledge of the nutrient status of sediments key to understanding their eutrophic condition (Kemp et al. 2005; Conley et al. 2007). Many field surveys have compared patterns of sediment nutrients, water column nutrients, and macroalgal biomass (Thybo-Christensen et al. 1993; Sundbäck et al. 2003; Boyle et al. 2004; Ferguson et al. 2004). Results of this correlative work suggest that sediments of shallow estuaries may stimulate and maintain algal blooms by cycling and retaining nutrients and organic matter. Experimental studies found that macroalgal biomass was stimulated by nutrients released from the sediments (McGlathery et al. 1997; Trimmer et al. 2000; Tyler et al. 2001; Rozen et al. 2002; Sundbäck et al. 2003; Kamer et al. 2004), and in situ measurements showed that nutrients flux out of sediments when water column concentrations are low (Fong and Zedler 2000; Caffrey et al. 2002), which may explain how high algal biomass is sustained. Sediment sources may be a result of microbial nutrient regeneration that is related to organic content (Hopkinson et al. 1999; Dong et al. 2000) and redox potential, a quantitative measure of the ability of the sediments to oxidize or reduce substances (Koch et al. 1992; Mitsch and Gosselink 1993). Sediments can also take up ammonium from decomposing macroalgal mats and function as a nutrient sink (Lavery and McComb 1991). Finally, levels of sediment nutrient enrichment can have negative upward cascading effects on the food chain (Armitage and Fong 2004; Armitage et al. 2006). Therefore, quantifying spatial and wet vs. dry season patterns of sediment characteristics is essential to assessing the eutrophic status of shallow Mediterranean climate estuaries.

Mediterranean climates in the Northern Hemisphere are typically characterized by warm dry summers (May–October) and cooler wet winters (November–April), with moderate temperatures falling within a relatively narrow range. The associated marine layer and high pressure cells reduce storm frequency, and rainfall can be absent for 4–6 months during the dry season. Estuaries with Mediterranean type climates and modified watersheds receive terrestrial nutrients in pulses from episodic storm events during the short wet season, often resulting in transient nutrient availability (Martins et al. 2001; Boyle et al. 2004). Although some studies in Mediterranean climate systems have documented peak macroalgal biomass in spring and summer (e.g., Sfriso et al. 1987; Marcomini et al. 1995; Martins et al. 2001; Naldi and Viaroli 2002; Marques et al. 2003), others found that macroalgae were able to proliferate in any season (Hernandez et al. 1997; Astill and Lavery 2001; Martins et al. 2001; Villares and Carballeira 2003; Viaroli et al. 2005). There is experimental evidence that warmer water enhanced biomass accumulation (Fong and Zedler 1993; Bintz et al. 2003). However, rapid nutrient uptake rates (Fujita 1985; Pedersen 1994; Naldi and Viaroli 2002; Lartigue and Sherman 2005; Kennison et al. 2011) and a high tolerance for a wide range of temperature (Fong and Zedler 1993) and salinity (Edwards et al. 1987; Young et al. 1987; Kamer and Fong 2000) may allow opportunistic green algae to dominate under a variety of environmental conditions (Kennison et al. 2011). For example, two studies in southern California showed opposite patterns, with maximum peaks found in summer (Kamer et al. 2001) and winter (Rudnicki 1986). However, a synoptic comparison of wet vs. dry season patterns of blooms in Mediterranean climate estuaries has yet to be undertaken. We hypothesized that blooms may occur both in wet and dry seasons and are driven by variance in local nutrient supply rather than relatively moderate seasonal changes in temperature and salinity characteristic of these systems.

For the vast majority of southern California estuaries, we lack even basic field surveys relating water column and sediment nutrients to macroalgal abundance. However, all existing studies concluded that these systems are extremely enriched (Rudnicki 1986; Page et al. 1995; Fong and Zedler 2000; Kamer et al. 2001; Fry et al. 2003; Boyle 2004). A few documented severe alterations in watershed nutrient inputs (Page et al. 1995; Boyle et al. 2004; Kamer et al. 2004), although there was some additional evidence identifying reduced sediments as a supplemental source of nutrients (Fong and Zedler 2000; Boyle et al. 2004). The objectives of this study were to quantify spatial and wet vs. dry patterns of water column and sediment nutrients, sediment organic content and redox potential, and macroalgal cover and biomass to evaluate eutrophic conditions within and between five southern California estuaries. Ascertaining these patterns will be valuable in evaluating and comparing eutrophic conditions of Mediterranean climate estuaries in southern California that will help justify larger scale efforts that may include nutrient loading rates.

Methods

Study Locations

Southern California, delineated to the north by Point Conception and to the south by a political boundary with Mexico, is characterized by coastal mountains and limited area of coastal plain. Both estuaries and their watersheds typically are small, with estuaries comprised of large intertidal mudflats and shallow subtidal areas (Fong and Kennison 2010). California's climate is Mediterranean, characterized by warm dry summers (May–October) and cooler wet winters (November–April). Macroalgal blooms are typically dominated by two opportunistic green algae (Chlorophyta), Ulva intestinalis and Ulva expansa, which are both one to two cell layers thick (Kamer et al. 2001). In addition, they are able to take up nutrients quickly and utilize large nutrient pulses over a sustained period (Kennison et al. 2011). Most rainfall occurs during discreet storm events within the wet season, and the frequency and magnitude of these events are tremendously variable both within and among years (Zedler 1982). With small watersheds and limited rainfall, the hydrology of many lagoons along the southern California coast is dominated by tidal action for much of the year. Longshore movement of sand coupled with high waves results in migration of the lagoon mouth (connection to the ocean) and may have caused natural seasonal lagoon closure in some systems (Ambrose and Orme 2000), although at present most are maintained open for water quality reasons. We conducted a 15-month survey to assess the relationship between water and sediment nutrients, sediment organic content and redox potential, and macroalgal abundance in five southern California estuaries: Carpinteria Salt Marsh Reserve (CSMR), two arms of Mugu Lagoon—Mugu-West (Mugu W) and Mugu-Calleguas Creek (Mugu CC), Upper Newport Bay (UNB), and Tijuana River Estuary (TJ) (Fig. 1a). These systems were chosen because they constituted a latitudinal gradient across the southern California region and included a range of watershed sizes and land use practices (Table 1). Precipitation data from October 2001 through January 2003 for each estuary were taken from weather stations closest to each estuary (Table 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs12237-013-9687-z/MediaObjects/12237_2013_9687_Fig1_HTML.gif
Fig. 1

Map of California with locations of estuaries used in the survey along the southern California coast (a). Carpinteria Salt Marsh Reserve is the most northwestern system (b), followed by Mugu Lagoon (c), Upper Newport Bay (d), and Tijuana River Estuary (e). Key depicts land usage, waterways, development, and roads. Symbols depict sampling locations within each estuary; head (triangle), middle (circle), mouth (inverted triangle). Land use abbreviations: Dev. residential and urban development, Agr. agricultural land use

Table 1

General comparative information on estuary size, watershed size, and land use. Mugu Lagoon (West and Calleguas Creek) share a watershed

Estuary

Size (ha)

Watershed size (km2)

Land use (%)

Reference

Carpinteria Salt Marsh Reserve (34°24′ N, 119°31′ W)

93

283

72 % agriculture

Page et al. (1995)

   

28 % urban

 

Mugu Lagoon (34°06′ N, 119°05′ W)

 

888

 

LWA (2008)

 Mugu Lagoon—West

ND

 

Agriculture and freshwater ponds

LWA (2008)

 Mugu Lagoon—Calleguas Creek

607

 

50 % open space

LWA (2008)

   

26 % agriculture

 
   

24 % urban

 

Upper Newport Bay (33°37′ N, 117°53′ W)

304

271

71 % urban

Rose (2006)

   

24 % open space

 
   

5 % agriculture

 

Tijuana River Estuary (34°33′ N, 117°05′ W)

1,000

4,403

Unknown

Zedler et al. (1992)

ND no data

Table 2

Precipitation from months sampled, from weather stations closest to estuaries (CSMR = Santa Barbara, Mugu Lagoon = Ventura, UNB = Laguna Beach, TJ = San Diego airport) as well as annual precipitation for 2001 and 2002. Data generated from http://www.countyofsb.org/pwd/pwwater.aspx?id=31668, Santa Barbara County Flood Control District, www.countyofsb.org, and http://www.cnrfc.noaa.gov/rainfall_data.php. Salinity and water temperature for each estuary with one measurement at each site (mouth, middle, head)

Estuary

Season

Annual precipitation (cm)

Monthly precipitation (cm)

Salinity (ppt)

Water temperature (°C)

    

Mouth

Mid

Head

Mouth

Mid

Head

CSMR

 December 2001

W

53.34 (2001)

5.66

27.0

15.0

2.0

13.9

17.0

17.0

 February 2002

W

17.78 (2002)

1.17

22.0

12.0

2.0

13.0

16.0

16.6

 June 2002

D

 

0.01

36.0

25.0

10.0

24.3

32.0

29.0

 September 2002

D

 

0.20

NS

30.00

21.0

19.6

23.8

NS

 December 2002

W

 

4.59

40.0

35.0

NS

17.4

17.8

14.9

 February 2003

W

 

2.91

26.0

10.0

3.0

9.2

11.5

14.6

Mugu W

 December 2001

W

35.72 (2001)

9.65

34.0

NS

28.0

14.3

NS

13.1

 February 2002

W

15.4 (2002)

13.82

34.0

34.0

35.0

14.9

15.1

20.0

 June 2002

D

 

0

40.0

NS

32.0

NS

NS

NS

 September 2002

D

 

0

38.0

39.0

36.0

NS

NS

NS

 December 2002

W

 

1.32

32.0

40.0

35.0

16.1

15.7

13.6

 February 2003

W

 

0.36

35.0

28.0

24.0

14.8

21.5

20.0

Mugu CC

 December 2001

W

35.72 (2001)

9.65

22.0

15.0

2.0

15.2

11.0

12.5

 February 2002

W

15.4 (2002)

13.82

25.0

16.0

10.0

18.0

17.4

16.8

 June 2002

D

 

0

21.0

14.0

5.0

24.3

20.6

NS

 September 2002

D

 

0

26.0

16.0

9.0

NS

NS

NS

 December 2002

W

 

1.32

16.0

8.0

5.0

15.7

14.2

14.5

 February 2003

W

 

0.36

20.0

16.0

10.0

17.1

15.3

15.7

UNB

 December 2001

W

42.04 (2001)

3.43

28.0

23.0

15.0

14.2

15.4

15.0

 February 2002

W

16.25 (2002)

0.76

26.0

13.0

10.0

19.2

17.8

17.4

 June 2002

D

 

0.13

40.0

31.0

30.0

NS

25.6

NS

 September 2002

D

 

0

37.0

34.0

25.0

20.5

NS

NS

 December 2002

W

 

7.21

37.0

35.0

19.0

16.0

16.6

16.0

 February 2003

W

 

15.24

32.0

27.0

27.0

13.8

20.5

16.6

TJ

 December 2001

W

25.12 (2001)

1.78

34.0

10.0

4.0

14.3

15.3

14.4

 February 2002

W

13.39 (2002)

0.43

35.0

28.0

NS

22.7

21.0

NS

 June 2002

D

 

0

39.0

35.0

37.0

21.4

27.3

24.9

 September 2002

D

 

0.79

40.0

40.0

40.0

21.3

18.9

17.5

 December 2002

W

 

5.87

41.0

40.0

36.0

16.4

20.1

17.8

 February 2003

W

 

12.4

30.0

0.0

0.0

17.0

19.9

20.0

NS not sampled

Carpinteria Salt Marsh Reserve

CSMR, the most northwestern estuary, is located approximately 11 km east of Santa Barbara, California (Fig. 1b). It is the smallest estuary and the watershed is mainly developed by agriculture (Table 1). There are two major streams that enter the marsh, Santa Monica Creek and Franklin Creek (Page et al. 1995). Two of our sites were located on a dredged channel west of Santa Monica Creek and its tributaries; portions of the tributaries drain into the channel. This channel is fairly straight and narrow (3–4 m) and relatively deeply incised (<2 m) with steep banks, making tidal flushing rates high for most of its length. The third site was further inland, where the channel shallows and widens out to broader mudflats (4–6 m), and water flow slows down considerably (Kennison 2008).

Mugu Lagoon (West and Calleguas Creek)

Mugu Lagoon is located within a Naval Base at Pt. Mugu, approximately 70 km northwest of Los Angeles and is the second largest estuary in our study (Table 1). For this study, we divided the estuary into two arms, Mugu Lagoon-Calleguas Creek (Mugu CC) and Mugu Lagoon-West (Mugu W) because these two areas have different watersheds, bathymetry, hydrological regimes, and physical habitat types (Fig. 1c). The two arms of Mugu Lagoon drain a total of 888 km2 (Larry Walker Associates 2008).

Mugu CC comprises the central portion of the estuary and is subject to a constant flow from a river that drains agricultural fields. Although large portions of this watershed are mountainous open space (Table 1), agricultural development is immediately upstream on the coastal plain and drainage from agriculture and urban runoff results in year-round water flow (CMWD 2006). Calleguas Creek is the main river flowing into the estuary; it was historically dredged to 9 m but has filled through time to 5 m by sedimentation from the river and the ocean inlet (CMWD 2006) and is approximately 60 m wide. This system tends to be well flushed by tides, with velocities near the mouth of 5–6 knots (Larry Walker Associates 2008).

The western arm of the estuary (Mugu W) is more lagoonal, but is highly modified with roads and runways interrupting the normal flow of tidal water (Fig. 1c). This portion of the estuary receives some drainage from agricultural plains of Oxnard as well as industrial drains originating from the Oxnard drainage ditches located beyond the Naval Station (Table 1, Larry Walker Associates 2008). This arm of the lagoon is approximately 4 km long and the lower portion consists of a broad and gently sloping mudflat (~150 m wide) with a shallow channel (<1 m). Several bridge culverts restrict flow resulting in muted tidal flushing.

Upper Newport Bay

UNB is located 56 km south of Los Angeles and is intermediate in size with a relatively small and highly urbanized watershed (Fig. 1d, Table 1). San Diego Creek and its major tributary drain 85 % of the watershed and are the main freshwater inflow (Rose 2006). The main channel is wide, approximately 100 m, with extensive broad mudflats and shallow banks; however, the center of the channel is dredged to 5 m below sea level for sediment retention purposes. UNB is separated from the Pacific Ocean by the lower bay, which has been dredged and developed into a marina with no natural area remaining. As a result, our most oceanward site is adjacent to the lower bay. Water residence time in the Upper Bay can be up to 2 weeks during neap tide (Pednekar et al. 2005).

Tijuana River Estuary

TJ is the largest estuary we surveyed, with the largest watershed (Fig. 1e, Table 1). TJ is located at the US/Mexican border with three quarters of the watershed located in Mexico and is highly urbanized with major cities just upstream of the estuary. Due to a Mediterranean climate, natural freshwater flow is limited in the summer dry season although treated wastewater discharge occurs throughout the year (Zedler 1996). We sampled the main channel, which is wide (50–100 m), shallow (~1 m), and bordered by broad mudflats. This estuary required a boat to reach upestuary sites, but in February of 2002 water depth limited access to that site.

Sampling Protocols

We sampled at low tide six times in December 2001; February, June, September, and December 2002; and February 2003. In each estuary, three sampling sites were permanently established along a main channel. Sites include the head of the estuary where there is freshwater flow, the mouth of the estuary (nearest the ocean inlet), and mid-way between. We chose to sample at low tide because this maximizes measurement of nutrient-rich freshwaters and minimizes tidal mixing. This choice may have been most clearly reflected in the Mugu CC water column and sediment nutrient data as this system received year-round runoff from nearby agriculture.

During each sampling event, we first sampled the water column. Previous work has shown that UNB is rarely stratified (U.S. Army Corps of Engineers 1993). Generally, these estuaries are classified as well-mixed tidal lagoons dominated by shallow subtidal and intertidal habitat, tending not to stratify (Nezlin et al. 2009), so all samples were taken from the mid-water column. A temperature reading was taken and salinity was measured with a handheld refractometer and three water samples were collected, placed on ice in a dark cooler, and transported to the laboratory within 6 h of collection. Upon return to the lab, samples were filtered (Whatman GF/C), frozen, and sent to DANR Analytical Laboratory at UC Davis where they were analyzed for NO3 + NO2 (hereafter referred to as NO3) and NH4+. Both species of N were quantified because of their rapid uptake rates by opportunistic green algae (Pedersen 1994; McGlathery et al. 1997; Lotze and Schramm 2000; Kennison et al. 2011). NH4+ and NO3 were measured by the diffusion–conductivity method (Carlson 1978). Total P in the water was determined by atomic emission spectroscopy following microwave acid digestion. Total P in the water was only above the detection limit in Mugu CC (ranged from 8 μM at the mouth to 38 μM at the head) and in only two wet seasons at the middle and head in TJ (range 56–88 μM). Otherwise, total P in the water was close to or below detection limit at all other times and estuaries. Soluble reactive phosphorus (SRP) was determined spectrophotometrically following reaction with ammonium molybdate and antimony potassium under acidic conditions (APHA 1998). SRP in the water (which was only sampled for the last three sampling periods) ranged from 2 to 6 μM in Mugu W, CSMR, and UNB and reflected the same ranges in values as total P in Mugu CC and TJ (Table 3). These automated methods have detection limits of 3.57 μM for all forms of N, 3.226 μM for total P, and 1.613 μM for SRP. For N data that were below detection limit, half the detection limit was used in all analyses. Due to limited data, neither total P nor SRP in the water was statistically analyzed.
Table 3

Water column soluble reactive phosphorus (SRP) concentrations. Data presented are means with SE in ( )

 

SRP (μM)

 

Mouth

Middle

Head

CSMR

 September 2002

2.151 (0.11)

2.043 (0.11)

1.613 (0.00)

 December 2002

2.043 (0.11)

1.828 (0.11)

2.366 (0.11)

 February 2003

2.043 (0.11)

1.935 (0.00)

5.699 (0.11)

MW

 September 2002

3.011 (0.11)

3.118 (0.11)

2.903 (0.00)

 December 2002

2.043 (0.22)

2.742 (0.16)

1.935 (0.032)

 February 2003

3.118 (0.11)

2.903 (0.00)

2.581 (0.00)

MCC

 September 2002

7.312 (0.11)

15.914 (0.60)

25.699 (0.57)

 December 2002

8.065 (0.32)

13.226 (1.62)

16.237 (1.69)

 February 2003

10.538 (1.41)

9.892 (0.71)

31.075 (0.92)

UNB

 September 2002

2.258 (0.00)

2.581 (0.00)

2.258 (0.19)

 December 2002

2.366 (0.11)

3.118 (0.11)

1.613 (0.00)

 February 2003

2.796 (0.11)

3.763 (0.11)

2.796 (0.43)

TJ

 September 2002

2.151 (0.00)

2.581 (0.00)

2.688 (0.11)

 December 2002

2.258 (0.00)

2.043 (0.11)

2.796 (0.11)

 February 2003

2.419 (0.16)

93.871 (0.37)

77.634 (0.39)

Within each site, we sampled intertidal macroalgae along a 30-m transect parallel to the waterline and 1 m downslope from the vascular vegetation (Kamer et al. 2001). We estimated macroalgal abundance by measuring percent cover, which includes filaments that grow intercalated in the sediment and may be visible but not collectable, and algal biomass, which included both attached and detached mats. At five randomly chosen points along each transect, which were reassigned each sampling time, a 0.25-m2 quadrat with 36 evenly spaced intercepts (forming a 6 × 6 grid) was placed on the benthos. The presence or absence of each macroalgal species in the top layer under each intercept was recorded. When present, algae were collected from a 530.9-cm2 area circumscribed by a plastic cylinder placed on the benthos in the center of each quadrat. Each sample was placed in an individual ziploc bag in a cooler, transported to the laboratory, and refrigerated. Algal samples were transferred to low nutrient seawater where they were cleaned of macroscopic debris, mud, and animals; placed in a nylon mesh bag; spun in a salad spinner for 1 min; and wet-weighed. Macroalgal biomass was normalized to area (gram wet weight per square meter). In order to compare biomass measurements to other systems, we converted to dry weight as 10 % of wet weight (10:1 wet/dry, Kamer et al. 2004).

We measured sediment reduction–oxidation potential (redox), organic content, and nutrient content every sampling date within the same random quadrats along each transect as for macroalgae. In order to minimize disturbance to the sediment, redox was measured before algae were sampled. Redox potential was measured with six brightened platinum electrodes and a calomel reference electrode at a fixed depth of 5 cm into the sediment. The electrodes were allowed to stabilize for 10 min prior to measurements with a portable pH/millivolt meter. We collected three sediment cores (5 cm diameter) to 5 cm depth from each of the five quadrats making sure algae were eliminated. The three cores were composited, placed in a dark cooler on ice, and returned to the lab within 6 h. Sediments were dried at 60 °C to a constant weight, ground with mortar and pestle, and sieved to less than 2 mm. Organic content was determined by loss after ignition in a 400 °C muffle furnace. Total N in the sediment was analyzed at DANR using the combustion gas analyzer method, and total P in the sediment was analyzed by alkaline extraction and measurement by flow injection analysis (Olsen and Sommers 1982; Prokopy 1995). These automated methods have detection limits of 0.04 % for sediment total N and 0.01 % for sediment total P. For data that were below detection limit, half the detection limit was used.

Statistical Analysis

A 3D principal components analysis (PCA) was used to assess the relationship between environmental variables and macroalgae visually and statistically. Spatial and wet vs. dry season patterns were investigated between estuaries by comparing the average of each site (head, middle, mouth) within each estuary for each sampling date (e.g., CSMR, mouth, December 2001 and Mugu W, mouth, December 2001, etc., for n = 87; three sampling dates were removed because of missing data). All data were fourth root-transformed to satisfy the normal distribution requirement and were normalized for different scales (Clarke and Warwick 1994). We used linear regression to further investigate the relationships between environmental and biological variables identified as important in the PCA. We regressed NO3 and sediment P against algal biomass and cover and NH4+ against redox potential.

Significant differences in rank similarities between groups of samples were tested by ANOSIM (Clarke and Green 1988). ANOSIM is analogous to the univariate test, analysis of variance (ANOVA), but rather than testing differences in means, it utilizes the dissimilarity matrix generated by Euclidean distances and computes randomized permutation tests. The global R statistic indicates whether significant differences exist between groups (significance level of 0.05); R takes a value of 1 when replicates within groups are more similar than replicates between groups (Clarke and Gorley 2001). Lower number of replicates will affect p values, whereas the global R statistic is a function of rank similarities (Clarke and Warwick 1994). Pairwise comparisons are used to assess differences within groups and a Bonferroni correction accounts for the effect of differences in group size (significance level of 0.05/n, where n = number of comparisons, Clarke and Warwick 1994).

PRIMER calculates missing values for variables that meet certain assumptions by using the expectation-maximization (EM) algorithm (Clarke and Warwick 1994). Missing values for redox, organic content, and sediment N and P data met these assumptions and were replaced by EM. Water column data for TJ December 2001 at the mouth and February 2002 in the middle were missing so all data from these sampling dates were excluded because they did not meet assumptions of EM.

We compared wet and dry season data for water column NO3 and NH4+, macroalgal biomass and percent cover, and sediment organic content, redox potential, and total N and P by calculating means for each site (n = 5 for sediment and n = 3 for water) and graphing the averages of those means for the wet season (December 2001, February and December 2002, and February 2003) and the dry season (June and September 2002). For each estuary, data were generated from water column nutrient (n = 3), sediment (n = 5), and algal (n = 5) samples for each site (n = 3) and season (n = 6) for a total of n = 54 or n = 90, respectively. For all statistical analyses, data for the head site in TJ in February 2002 were not included due to missing values. To further explore correlations statistically, nonlinear regression (best fit) was used to quantify the relationship between sediment N and organic content and biomass and percent cover.

Results

Relating Macroalgal Blooms and Environmental Drivers

PCA was used to make comparisons between estuaries and sites within estuaries of environmental variables and macroalgal abundance. The dominant green macroalgae found were U. expansa and U. intestinalis, and PCA showed that macroalgae were not associated with any one estuary or site within an estuary and were not positively related to any specific set of environmental conditions (Fig. 2). High sediment N and organic content clustered towards the head and middle of TJ, the head of CSMR, and the middle of Mugu W. Mouth sites of TJ, UNB, and Mugu W are grouped together, toward macroalgae and away from water and sediment nutrients, suggesting that environmental variables are similar at these locations. Sites at the mouth of CSMR are grouped with high redox potential, suggesting that they are highly oxidized. Mugu CC was separated from other estuaries based on high water column NO3 and sediment P and low macroalgal abundance. The first three principal components explained 74 % of the total variance (Table 4). The variance in the first component (31 % of the total variance) was explained by the negative loading of algal abundance (as wet weight and percent cover), whereas the positive contributions were due to sediment P and water column NO3. Sediment organic content and total N dominated the second component and independently explained 27 % of the total variance. The variance in the third component was related to the positive contribution of redox potential in the surface sediment and negative loading of water column NH4+, explaining 17 % of the total variance. Where sediments were anoxic, there was high water column NH4+, possibly where regeneration was occurring. Regression analyses of NO3 and sediment P against algal biomass and cover and between NH4+ and redox potential showed no significant relationships (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs12237-013-9687-z/MediaObjects/12237_2013_9687_Fig2_HTML.gif
Fig. 2

The 3D PCA correlates environmental and biomass data with estuaries depicted by color (TJ = orange, UNB = purple, CSMR = green, Mugu W = blue, Mugu CC = red), locations are depicted as symbols (head (triangle), middle (circle), mouth (inverted triangle) and wet vs. dry seasons (wet = W and dry = D). Circle around red symbols show separation of Mugu CC grouping along PC2 toward sediment total P and NO3, and circle around mouth sites show separation of mouths of TJ, Mugu W, and UNB

Table 4

Three-dimensional principal components analysis for correlations between environmental variables (redox = Mv, sediment nutrients = %N, P, organic content = % dry wt, water column nutrients = micromolar, and biomass = wet weight per square meter). Italics indicates the dominating variable on that PC

Eigenvalues

   

PC

Eigenvalues

% Variation

Cum.% variation

1

2.46

30.8

30.8

2

2.13

26.6

57.4

3

1.34

16.8

74.2

4

0.831

10.4

84.6

5

0.763

9.5

94.1

Eigenvectors (coefficients in the linear combinations of variables making up PCs)

Variable

PC1

PC2

PC3

Redox

0.012

−0.178

0.562

Organic

0.06

0.639

0.033

Sed N

0.031

0.641

0.05

Sed P

0.453

0.17

0.426

Wet wt

0.514

0.183

0.28

% Cover

0.545

0.144

0.279

NH4+

0.058

0.251

0.525

NO3

0.474

0.054

0.259

Environmental variables were significantly different among sites (two-way ANOSIM global test, factors site × estuary, for site R = 0.41, p = 0.001). A pairwise comparison (estuaries × site) using the Bonferroni correction (p = 0.005, n = 10) showed mouths different from head (ANOSIM, R = 0.524, p = 0.001) and middle (ANOSIM, R = 0.457, p = 0.001) sites. Environmental variables were also well separated among estuaries (two-way ANOSIM global test, for estuary R = 0.63, p = 0.001) with Mugu CC the most different from other estuaries (Table 5, R > 0.7 for all comparisons), and environmental variables for Mugu CC and TJ were the most different from each other (Table 5). Two-way ANOSIMs (factors = season × estuary and season × site) showed no significant differences between seasons and sites (global R = 0.024, p = 0.341, global R = −0.084, p = 0.998, respectively). Overall, during the duration of our sampling time, macroalgae might be prevalent any time of year, with no consistent spatial pattern and no consistent relationship to the measured environmental variables.
Table 5

Pairwise tests for two-way ANOSIM (site significant differences in italics, p < 0.001)

Estuary effect

R

Statistic

Level

Significance

CSMR, Mugu W

 

0.685

 

0.001

CSMR, Mugu CC

0.727

 

0.001

 

CSMR, TJ

0.48

 

0.001

 

CSMR, UNB

0.631

 

0.001

 

Mugu W, Mugu CC

0.79

 

0.001

 

Mugu W, TJ

0.539

 

0.001

 

Mugu W, UNB

 

0.377

 

0.003

Mugu CC, TJ

0.883

 

0.001

 

Mugu CC, UNB

 

0.798

 

0.001

TJ, UNB

0.565

 

0.001

 

CSMR Carpinteria, Mugu W Mugu West, Mugu CC Mugu Calleguas Creek, UNB Upper Newport Bay, TJ Tijuana River

Spatial and Wet Vs. Dry Season Patterns

All estuaries received more rain in 2001 than 2002, and February 2003 was a particularly wet month in the southern estuaries, UNB and TJ. Within years, December and February were the wettest months sampled (Table 2) and in all estuaries precipitation was <1 cm in the dry season months of June and September 2002. Annual precipitation in 2001 was highest in CSMR, and in 2002, precipitation was lower and comparable among estuaries.

There were strong salinity gradients, at least in the wet season, in four of the five estuaries (Table 2). The exception was Mugu W where the range of salinity was smaller (24–40 ppt) suggesting little freshwater supplied from the watershed at any time of year. During most sampling periods, there were salinity gradients in CSMR, Mugu CC, and UNB, with the lowest salinities at the head of the estuary indicating there was some freshwater flow throughout the year. In contrast, a salinity gradient occurred only in the wet season in TJ. In the dry season, water was hypersaline and did not vary across sites in the estuary, suggesting little freshwater flow and high evaporation at this time. Overall, water temperatures were colder in the wetter seasons and the mouth during all sampling periods except December 2002 and were within the wide range of tolerance for macroalgal growth.

Water column NO3 in Mugu CC was extremely high in both wet and dry seasons (Fig. 3a); NO3 was 1–2 orders of magnitude higher than in other estuaries in the dry season, probably due to agricultural runoff. In addition, salinities were low at the mouth, probably due to a combination of agricultural runoff in the dry season and river flow in the wet season. In contrast, the other four estuaries showed a pattern of higher NO3 values in the wet season. Mean nitrate in all other estuaries ranged from 47 ± 18 to 300 ± 81 μM in the wet season and 2 ± 1 to 100 ± 52 μM in the dry season. The highest values for each of these ranges occurred in CSMR, making it the second most nutrient-enriched of our study systems.
https://static-content.springer.com/image/art%3A10.1007%2Fs12237-013-9687-z/MediaObjects/12237_2013_9687_Fig3_HTML.gif
Fig. 3LR

Seasonal patterns in each estuary of water column and sediment nutrients and macroalgae. Wet season is December 2001, February and December 2002, and February 2003; dry season is June and September 2002. Means and standard errors of water column NO3 (a), and NH4+ (b), sediment N (c), organic content (d), sediment P (e), redox potential (f) and macroalgal percent cover (e) and biomass (f). Note scale break for a and b

As demonstrated by PCA, means of water column NH4+, sediment N, and organic content showed similar patterns among estuaries that were driven by maximum values in TJ (Fig. 3b–d). In TJ, mean water column NH4+ in the wet season (200 ± 43 μM) was an order of magnitude greater than all other estuaries, most likely associated with discharge of sewage upstream. For all other estuaries, NH4+ was within the range of 10–23 μM NH4+ in both seasons. Sediment N was highest in TJ, followed by CSMR, MW, and MCC with UNB the lowest. Sediment organic content was high in TJ in both the wet and dry seasons (max. 5 ± 0.7 %), while organic content was overall similar between wet and dry seasons in CSMR, MW, and MCC and lowest in UNB (1.6 ± 0.1 %). Sediment P, like water column nitrate, had maximum values in MCC in both the wet and dry seasons and overall similar values between wet and dry seasons in all other estuaries (Fig. 3e). MW had the second highest sediment P followed by CSMR, TJ, and UNB. Redox potential showed no consistent pattern and was highly variable in all five estuaries (Fig. 3f).

The highest macroalgal percent cover and biomass were found in Mugu W in the dry season, followed by UNB (Fig. 3g, h), despite low water column N concentrations in this season in both estuaries. The highest algal abundance also occurred in the dry season for MCC, though there was a little proliferation of biomass in any season despite extremely high water and sediment supplies year-round. In contrast, both TJ and CSMR had the highest algal abundance in the wet season. While high biomass corresponded to higher water column and sediment nutrient supplies in the wet season in TJ and CSMR, in UNB, it did not. Means of wet and dry seasons for sediment N were negatively related to organic content (y = −0.026x + 0.18, r2 = 0.52, p < 0.0189, DF = 8) and means of wet and dry season for biomass and percent cover were positively related (y = 16.68x − 251.18, r2 = 0.76, p = 0.0011, DF = 8).

Discussion

Our results indicated that Mediterranean climate estuaries of southern California rank as some of the most nutrient-enriched estuarine systems in the world. We found that anthropogenic inputs from the watershed resulted in extremely high nutrient concentrations in the water column and sediments of all five of our estuaries, at least at some times, most likely a consequence of the long history of extensive physical modifications (Zedler et al. 2001). Although variable, many of the estuaries in our study had water column NO3 concentrations orders of magnitude higher than estuaries worldwide; even the lowest NO3 concentrations were higher compared to East Coast systems (Valiela et al. 1992; Nixon 1995; Taylor et al. 1995) and comparable to the eutrophic Scheldt estuary in the Netherlands (Chen et al. 2007). Our water column nitrate values were often even higher than those in one of the most historically nutrified Mediterranean estuaries, Venice Lagoon, Italy, where water column concentrations ranged from 4 to 132 μM NO3 (Svensson et al. 2000), and NH4+ concentrations were 5–40 μM NH4+. We found that the sediments as well as the water column may be an important reservoir of nutrients as sediment total N in southern California estuaries was comparable to other shallow eutrophic systems (Eyre and Ferguson 2002; Sundback et al. 2003). We reason that Southern California systems may have overall higher nitrogen concentrations in water and sediments because each has a unique combination of agricultural/urban development in the watershed coupled with physical modifications within each estuary (filling, dyking, channelization) that may alter natural hydrology and water residence times. Like systems on the East Coast of the USA such as Waquoit Bay and New Jersey coastal bays, watershed development and increasing populations may also be driving eutrophic conditions (Hunchak-Kariouk and Nichols 2001; Fox et al. 2008; Kennish et al. 2007). However, it has been established that although southern California is characterized by one of the densest human populations in the USA, southern California estuaries are among the least studied (Bricker et al. 2008). While our study has just begun to identify and compare sources of nutrients in these highly modified systems, it has clearly identified the need for further mechanistic and process-oriented studies.

While it appeared that nitrogen was not consistently limiting in these systems, it is less clear what role phosphorous plays in limiting macroalgal blooms. With few exceptions, we found that total P and SRP in the water were low or below the detection limit, suggesting that these may be limiting, at least some times. However, there is some evidence suggesting that P may not play an important role in limiting macroalgal blooms in these estuaries. In microcosm experiments investigating nutrient limitation in southern California coastal lagoons, P was not found to limit algal growth across a wide range of N/P supply ratios and concentrations (Fong et al. 1993b). In Tijuana estuary, P was found in excess of algal demand in most seasons, suggesting N limitation; the exception was spring, when nitrogen-rich runoff entered the estuary and P or other factors may have been limiting (Fong 1986). Boyle et al. (2004) found that maximum mean total P concentrations in the water of UNB (~16 μM) were well above the means documented in the eutrophic Peel-Harvey estuary in western Australia (McComb et al. 1981) and the mesotidal Guadiana estuary, Spain (Domingues et al. 2011). Our study suggested that the sediments rather than the water column maybe an important reservoir of P, supplying additional nutrients for algal growth, as sediment p values at CSMR, UNB, and TJ were comparable to the highly eutrophic Venice lagoon, Italy (0.03–0.07 % dry wt, Marcomini et al. 1995) and to a previous UNB study (~0.05–0.07 % mean dry wt, Boyle 2004). Nutrient flux experiments in Formosa Slough and UNB showed sediment P fluxed into the water column when water column P was low (Fong and Zedler 2000; Kamer et al. 2004). To establish if P is a limiting factor in these systems, and whether the sediments play a key role, the next step is to investigate nutrient flux rates between sediments, algae, and the water column in these systems.

Eutrophication in these shallow Mediterranean climate estuaries took the form of dense blooms of macroalgae that could occur any time of the year. Four of five estuaries had macroalgal blooms of relatively high magnitude that were not restricted to either the wet or dry season or a specific site. Others have also found that, although green opportunistic macroalgae often bloom in the summer, blooms can occur in any season in Mediterranean climates (Hernandez et al. 1997; Patricio et al. 2007). The magnitude of blooms in southern California estuaries was similar to those quantified in other eutrophic systems. We found peak macroalgal biomass at a single site ranged from >200 g dry wt m2 in UNB to >300 g dry wt m2 in Mugu W, comparable to other Mediterranean systems that typically have blooms ranging from 300 to 500 g dry wt m2 (Flindt et al. 1997; Hernandez et al. 1997; Martins et al. 2001; Naldi and Viaroli 2002; Viaroli et al. 2005). While Mugu W in the dry season reached maximum peak biomass of 3 kg wet wt m−2 compared to 25 kg wet wt m−2 found in one site of Venice Lagoon, Italy in the 1980s (Sfriso et al. 2003), biomass from some sites in Venice Lagoon in the 1990s were comparable to those in our study (Sfriso et al. 2005). In Mugu W, and likely in Venice Lagoon, these mats were made up of drift algae. The magnitude of blooms and presence of biomass in all but one of our estuaries in both wet and dry seasons suggest that opportunistic algae can take advantage of elevated nutrients in any season (Kennison et al. 2011). These results suggest that future research should be focused on local modifications of watersheds that may enhance local nutrient supplies as well as changes between wet and dry seasons.

Although there was no consistent relationship between biomass and nutrients, in two of our five estuaries in southern California, CSMR and TJ, we found higher nutrients and blooms in the wet season, suggesting that the watershed was a source of nutrients from storm events that may have stimulated blooms. Page et al. (1995) and Robinson et al. (2005) also found that maximum nutrient inputs were associated with runoff in the wet season from multiple watershed sources in CSMR. In TJ, algal biomass, though corresponding with seasonal peaks in nutrients, was moderate compared with other systems in our study, probably due to toxicity from large concentrations of NH4+. Treated wastewater is released directly upstream, and in the wet season, spills of untreated sewage can occur providing a constant source of organic matter and NH4+. Peckol and Rivers (1995) found that enrichment of ambient water by 100 μM NH4+ reduced NH4+ uptake by macroalgae and growth was inhibited, suggesting that toxic levels had been reached in that study. Although our NH4+ concentrations were lower than this threshold, they may still have reduced algal growth. Eutrophic conditions in the wet season in CSMR and TJ may have been stimulated by nutrient availability from anthropogenic sources delivered by runoff from winter storms.

However, our study revealed an overall uncoupling of spatial and wet/dry patterns of water column nutrient concentrations from macroalgal blooms in southern Californian estuaries. PCA results revealed either negative or no relationship between macroalgae and water column nutrients; as one example, Mugu W had both the highest algal biomass and lowest water column nutrients of the study estuaries. Other studies have also found that nutrient concentrations do not always directly relate to patterns of macroalgal biomass and, therefore, cannot be the only predictor of blooms, which has led to investigations of internal nutrient flux, hydrological processes, and nutrient loading rates (see Schramm 1999 for review and Cloern 2001). One possible explanation for the uncoupling is that internal cycling processes affected the relationship between nutrient supply, availability, uptake, and storage. Both laboratory and field studies in UNB found that sediments were a source to algae during the summer and fall, when water column nutrient concentrations were low, suggesting that simply measuring water column availability did not accurately represent the total nutrient supply to algae (Boyle et al. 2004; Kamer et al. 2004). Other studies suggested that macroalgal mats intercept sediment nutrient flux, uncoupling sediment regeneration from water column concentrations (McGlathery et al. 1997; Valiela et al. 1997; Trimmer et al. 2000; Tyler et al. 2001). Additionally, Sundback and McGlathery (2005) found that internal cycling of nutrients within the algal mats themselves sustained biomass. Our study demonstrated that water column nutrients are not an adequate predictor of eutrophic conditions and identified the need for studies that quantify the contribution of internal nutrient processing, variable hyrdrological regimes, and multiple external nutrient sources.

The most extreme case of uncoupling between nutrients and macroalgal abundance occurred in Mugu CC, where excessively high nutrients were found in both water and sediment and yet macroalgal abundance was either low or completely absent. One possible explanation is that in this deeply dredged system, high water velocities scour sediments and/or rip off juvenile algal filaments (Kennison 2008). In some lower energy systems, juvenile filaments can aid rapid biomass proliferation of floating mats by providing a source to adult mats; however, in this system, high water velocities may export filaments, preventing proliferation. Alternatively, although green algae are capable of taking up nutrients in excess of their physiological needs and reallocating when externals sources are low (Naldi and Viaroli 2002; Kennison et al. 2011), nitrate or pollutant concentrations may have been high enough to be toxic. Although there is little documentation of nitrate toxicity to algae, in a 4-week-long nutrient enrichment experiment, Fong et al. (2004) found no growth of macroalgae occurred in the 1,000-μM NO3 enrichment treatment. In addition, herbicides and pesticides from agricultural runoff could inhibit algal growth (Cohen et al. 2001). Regardless of the mechanism, while too much algae is known to have negative effects, having too little algal biomass disrupts the natural role of algae as primary producers facilitating nutrient retention and cycling (Tyler et al. 2001). Therefore, one possible serious consequence of a lack of algae in this excessively nutrient enriched system is that this system may be functioning as a conduit of nutrients and toxins from the agricultural fields upstream to the ocean, putting nearshore water quality at risk.

In Mugu West and UNB, a very different set of anthropogenic modifications may have enhanced algal biomass accumulation year-round despite lower nutrient supplies than in other systems. Both estuaries have muted tidal flushing and therefore longer water residence times and contain wide and shallow mudflats that comprise suitable habitat for algal growth (Larry Walker Associates 2008; Pednekar et al. 2005). Previous studies documenting high levels of macroalgal abundance and eutrophication in UNB suggested that excessive nutrient loading from the watershed and retention of nutrients by the sediments stimulated macroalgal biomass in the summer and fall when eternal supplies were low (Kamer et al. 2001; Boyle et al. 2004). In Mugu W, nutrients are released in the dry season from fertilizing agricultural fields upstream. In addition, the interruption of natural water flow from physical barriers such as culverts, roads, and a runway mutes the tidal flow, which increases algal depositional areas and creates stores of nutrients in the dry season. Similarly in Chesapeake Bay, Kemp et al. (2005) found that deposition of decomposing phytoplankton biomass stimulated a biogeochemical feedback, further intensifying the eutrophication process. For both UNB and Mugu W, the relatively long water retention times, chronic nutrient inputs with enriched sediments, and year-round blooms suggest that their unique patterns of nutrients and macroalgal abundance are due to complex secondary processes that create eutrophic conditions.

Eutrophic conditions in southern California estuaries have been affected by excessive anthropogenic source of nutrients, the effects of which may be exasperated by modifications to water flow and complex secondary feedback mechanisms. Some highly developed watersheds have intensive agricultural practices in close proximity to estuaries where agricultural and urban discharge provided chronic nutrient inputs throughout the year, while others are loaded with nutrients from human waste; both resulted in elevated nutrient inputs and storage in sediments, increasing macroalgal abundance. Even estuaries directly adjacent to each other can have very different timing and symptoms of eutrophication, so care must be taken to understand the processes that generate these unique patterns, if we hope to conserve or even restore them in the future.

Copyright information

© Coastal and Estuarine Research Federation 2013