, Volume 827, Issue 1, pp 391–404 | Cite as

Biomass across space and tide: architecture of a kelp bed with implications for the abiotic environment

  • Tiffany A. Stephens
  • Matthew J. Desmond
  • Christopher D. Hepburn
Primary Research Paper


The complex, stratified seaweeds within kelp forests provide habitat to a multitude of organisms and can alter the physical and chemical parameters of their surrounding environment. It is unclear, however, how patterns in the architecture of these beds change as the tide ebbs and floods. We investigate biomass distribution of floating and stipitate canopies within a kelp bed during low and high slack tides to determine how biomass interacts with common environmental parameters (nutrients, light, and mass-transfer). Floating canopy biomass remained consistent despite differences in depth, likely driven by an interaction between stipe density and individual biomass. Biomass was distributed inconsistently throughout the water column, in which biomass at the surface roughly doubled at low tide relative to high. Despite an increase in kelp biomass at the surface of the water column during low tide, more light reached the benthos than at high tide, suggesting that seawater optical properties independent of algal canopy better explain light attenuation. Seawater nutrients were consistent throughout the bed. Rates of mass-transfer decreased from the exterior to the interior of the bed and also attenuated with depth. This study highlights the structural complexity of kelp beds and the localized effects on important environmental variables.


Macrocystis Temperate reef Canopy Sub-canopy Light Nutrients Mass-transfer Water motion Tidal cycle Tidal height 


Temperate reefs are typically dominated by large brown algae, primarily of the orders Fucales and Laminariales (Barrales & Lobban, 1975; Kain, 1989). Members of these groups form diverse and vertically stratified kelp forest communities, some of which include large conspicuous canopies (e.g., Macrocystis pyrifera (Linnaeus) C. Agardh; Nereocystis luetkeana (K. Mertens) Postels & Ruprecht) that float in the water column above other macroalgae (Dayton et al., 1984; Graham et al., 2007). By direct and indirect interference of the local environment (Wernberg et al., 2005; Irving & Connell, 2006), multi-layered canopies influence the recruitment, productivity, and physiological performance of other algal species by modifying light (Wing & Patterson, 1993; Arkema et al., 2009), water motion (Goldberg & Kendrick, 2004; Gaylord et al., 2007), and, sometimes, nutrient concentrations (Jackson & Winant, 1983; Fram et al., 2008). Interspecific competition for light is an important mechanism that can structure communities (Dean & Jacobsen, 1984) and is a likely driver behind the varying structural morphologies of canopy-forming species in algal assemblages, as is observed in terrestrial forests (Connell, 1989; Spies & Franklin, 1989).

Macrocystis pyrifera, for example, typically out-competes other algal species, because its canopy can reduce light to understory algae by more than 95% (Pearse & Hines, 1979; Reed & Foster, 1984). Although such shading is not constant over time due to natural variation in biomass (Dayton et al., 1999), the effect that Macrocystis imposes on benthic light availability has been well documented via field experimentation and correlative studies (reviewed by Dayton, 1985). Typically, high canopy biomass results in a reduction of understory algal density and/or biomass (Schiel & Foster, 1986). Water motion (e.g., waves and currents), however, can ameliorate negative impacts from canopy-induced shading by mechanically disrupting canopy formation, allowing for flecks of light to reach understory algae (Wing et al., 1993).

Water motion also influences algal physiology via mechanisms involving molecular exchange between algae and the water column, known as “mass-transfer”. Increased water motion reduces the thickness of the diffusion boundary layer enveloping algal tissues and allows for increased rates of the exchange of chemical resources and wastes (Hurd, 2000). Such biophysical processes are particularly important when ambient seawater nutrients are low. In a nitrogen-limited environment, for example, kelp at the perimeter of the bed can display higher growth rates compared nearby kelp that grow just a few meters within the kelp bed due to increased rates of mass-transfer despite similar availability in nutrients, light, and temperature (Stephens & Hepburn, 2014). Such fine-scale processes within kelp beds warrant further investigation to understand the drivers of primary productivity.

Hydrodynamic forces beyond a threshold, however, can be deleterious because taller canopies are sometimes disadvantaged due to exposure to higher wave stress (Dayton, 1985). In California, nearly 600 ha of Macrocystis were removed during a sequence of large storms, which imposed long-term reductions in critical food and habitat for numerous associated reef species (Dayton & Tegner, 1984). The effect on understory macroalgae was moderate, because floating canopies of Macrocystis buffered incoming wave energy. Similar storm events have been observed in other beds dominated by smaller canopy-forming macroalgae [i.e., Ecklonia radiata (C. Agardh) J. Agardh], but dislodgement was far less severe (Wernberg et al., 2005) and can primarily be attributed to the capacity of physical buffering by biogenic structure across reefs and to the attenuation of water motion as depth increases (Jackson & Winant, 1983; Rosman et al., 2010).

This study aims to better understand the how multiple environmental factors shape the vegetative architecture within Macrocystis-dominated kelp beds, and vice-a-versa, and the potential trade-offs to such distribution of biomass. We investigate this by measuring rates of mass-transfer, concentration of seawater nutrients, ambient light, and the distribution of biomass of Macrocystis and dominant sub-canopy algal species across three-dimensional space. We also assess how the distribution of floating biomass changes during high and low slack tidal events and how this may influence the light environment in the sub-canopy. We anticipated that Macrocystis biomass and density would increase as one moves from the exterior to the interior of the kelp bed (see Dayton et al., 1992, 1999) and expected that the rates of mass-transfer would be inversely related to kelp density and biomass. We also anticipated that the rates of mass-transfer would reduce as depth increases, because wave energy decreases as depth increases (see Gaylord et al., 2012). That Macrocystis beds in SE New Zealand are relatively narrow (i.e., water masses moving through the bed should have low residency time) and that coastal waters in SE New Zealand are well mixed with weak upwelling (Heath, 1975), we expected that seawater nitrogen concentrations would be similar throughout the bed. Finally, it is suggested that upright fronds of Macrocystis have relatively little influence on light penetration to the bottom compared with the floating canopy fronds (Wing et al., 1993). Therefore, we hypothesized that less light would reach the benthos at low tide because vertically sitting fronds at high tide will be compressed into a horizontal plane at low tide, reducing the penetration of light.


Study site

This study was conducted throughout February 2014 (austral summer) at Horseshoe Bay, Stewart Island (S46°52′40.299″, E168°8′50.461″), which is situated off the southeast corner of New Zealand. The kelp bed used resembles other semi-sheltered sites in the region: Macrocystis fringes the shoreline (extending about 13–20 m away from shore) and the subtidal is characterized by a gently sloping, rocky reef (to approximately 12–15-m depth). The reef is colonized by numerous algal species (see Desmond et al., 2015), forming both floating and stipitate canopies.

Density and biomass

The density, biomass, and height of floating and stipitate canopy species were quantified throughout the reef. The floating canopy was composed of only Macrocystis pyrifera. The stipitate canopy included Ecklonia radiata, Carpophyllum flexuosum (Esper) Greville, Cystophora platylobium (Mertens) J. Agardh, Cystophora scalaris J. Agardh, Landsburgia quercifolia Harvey, Marginariella urvilliana (A. Richard) Tandy, Sargassum sinclairii J.D. Hooker & Harvey, and Xiphophora gladiate (Labillardiére) Montagne ex Kjellman. The densities (individuals m−2) of all seaweeds were quantified across the reef at 2-, 4-, 6-, 8-, and 10-m-depth bins (mean lower low water, MLLW; see Fig. 1) within 30-m transects; this was achieved by quadrat counts (1 m2) using SCUBA. The placement of quadrats within each depth band was assigned using five random numbers, ranging from 1 to 30. Individuals are defined as all biomass associated with a single holdfast; there was no coalescence of multiple holdfasts into one large holdfast observed for any of the species identified in this study.
Fig. 1

Representative canopy-forming macroalgae in this study site and their distribution across the reef (2-, 4-, 6-, 8-, and 10-m horizontal band; MLLW) and across depth (2-, 6-, and 10-m vertical band; MLLW), as determined using benthic surveys. Macrocystis pyrifera is the only species that contributes to the floating canopy, while the stipitate canopy is comprised of many brown macroalgae (see key in bottom left corner). The white circles indicate the approximate location of where dissolution blocks were deployed across the reef and across depth (2, 6, and 10 m)

To determine biomass, the same section of reef was immediately revisited on subsequent dives for algal collection. At each depth, ten individuals of each species identified in the quadrat counts were collected (excluding Macrocystis), unless that species was absent at a given depth. These were transported to the lab, where their biomass (wet weight) was quantified. Due to constraints in boat capacity and time, Macrocystis biomass was quantified at only 2-, 6-, and 10-m depths (the bed’s interior, middle, and exterior; see Fig. 1). Using SCUBA, fronds of Macrocystis individuals were bundled together with flagging tape, the holdfasts were removed from the benthos, and the entire individual was allowed to float to the surface for collection via snorkelling and boat. Three individuals were collected at each bed location. The kelp was immediately transported to a nearby wharf, where each frond from each individual was removed (cut immediately above sporophyll), labelled using flagging tape, and the length measured (from cut above the sporophyll to the apical meristem at the end of the frond). Data on the number of stipes per individual and the number of fronds that reached the surface were borne from this processing. Afterwards, the sporophylls and fronds were transported to a nearby lab to record the biomass (wet weight). The biomass of each frond from each individual was quantified in 1-m segments; the first measured segment for each frond was at the bottom of the stipe, nearest to the holdfast. Holdfast biomass was not measured because of the abundance of difficult-to-remove organisms (e.g., sponges, woven algae, and colonial ascidians) living within the haptera. We report biomass for two different surface bins: 2 m and 1.41 m. Biomass distribution in the top 2 m of the water column was included and analysed for direct comparison with data presented by elsewhere (i.e., North, 1971). Although reporting on the biomass in the top 2 m of the water column is more easily standardized across studies, we also reported on all biomass growing in the top 1.41 m of the water column in this study, because the magnitude of tidal exchange at Horseshoe Bay is 1.41 m (annual mean; calculated using tidal data from Mr. Tides version 4.08, based on NOAA predictions) and it is possible that biomass–tide interactions are strongest when considering only the biomass that is directly influenced by the tidal swing. To investigate how biomass at the surface of the water column fluxes throughout tidal cycles (i.e., high versus low tide), it was necessary to also consider the mean tidal exchange of 1.41 m, which is ecologically and functionally relevant in this system.

Mass-transfer and seawater nutrients

Gypsum dissolution rates were used to quantify mass-transfer rates throughout the kelp bed. As water flows around these blocks, they dissolve to release mass over time and integrate both turbulent and unidirectional flow into a single number related to mass-transfer (Jokiel & Morrissey, 1993). This should not be considered as a measure of water motion but as an index of diffusion (Porter et al., 2000), a key-limiting factor for many biological processes (Jokiel & Morrissey, 1993) and of interest here. Due to the diffusivities of the major ions in gypsum (calcium and sulphate) being similar to dissolved inorganic nutrient species such as phosphate, nitrate, and ammonium, these methods provide most tractable index nutrient mass-transfer around complex biogenic structures and have been widely used (e.g., Stewart & Carpenter, 2003; Falter et al., 2005; Lowe et al., 2005; Hepburn et al., 2007).

Blocks were prepared using gypsum (calcium sulphate hemihydrate 100%) mixed with Milli-Q water in a ratio of 3:2 (w:w). This mixture was then transferred into 3-cm−3 molds and each block received a looped cable tie for attachment purposes before the mixture set. The blocks were dried at 65°C and the initial weights recorded. Three zones spanning the width of the bed (cross-shore; interior, middle, and exterior bed), as well as three depths (2, 6, and 10 m) were established for block deployment (see Fig. 1). In each zone and at each depth, ten blocks were attached to ten separate Macrocystis fronds using SCUBA. After approximately 48 h, blocks were collected, dried at 65°C, and final weights recorded. A full set of dissolution blocks were deployed at three separate sampling events in Feb 2014, with 3–4 days between each sampling event. Dissolution rates are expressed as weight lost per hour (g h−1).

Seawater nitrate (NO3) and ammonium (NH4+) concentrations, in replicates of three, were collected at the same kelp bed locations as dissolution block deployment. Nutrients were sampled at five intervals: Dec 2013, Jan 2014, Jun 2014, Jul 2014, and Aug 2014. HCl-washed equipment was used throughout sampling, and seawater was immediately filtered (Whatman® GF/C) and frozen. Samples were analysed using a Lachat QuikChem® 8500 automated ion analyzer.


Three loggers (HOBO® Pendant 64 K—UA-002-64) were deployed for 3 months: one on land above the water’s surface and two underwater at − 2.0 m and − 10.0 m (MLLW). The logger deployed on land was necessary to quantify the percent of surface irradiance that reached the subtidal loggers. For analysis, only light data recorded during 1-h intervals surrounding high and low slack were compared to determine whether kelp influenced shading intensity at these two tidal elevations. In addition, hour-long intervals overlapping with darkness (i.e., early dawn or late dusk) were excluded from analysis. The HOBO® sensors recorded light intensity in units of Lux and it was necessary to convert these data to PAR values. Calibration was achieved through simultaneous recording (at different depths throughout the water column) using a HOBO® logger and an LI-COR® underwater quantum sensor (LI-192SA coupled with an LI-250A), after which a direct relationship between Lux and PAR (mol photon m2) was determined and used for calibration (Long et al., 2012). Full calibration methods are described by a seperate study that shared the same field site (see Desmond et al., 2015). Light data are expressed as both percent surface irradiance and as mean dose per slack tide. The latter is a modification of the “dose per day” reporting method, where “dose” is the cumulative amount of light, opposed to mean light, in a defined time period.

Statistical analysis

The differences in means of macroalgal density, total biomass, and bottom biomass (random factors) across the reef (bed position; fixed factor) were determined using one-way ANOVA for both Macrocystis and stipitate canopy species. Due to multiple statistical tests, i.e., eight testing the same independent variable (bed position), we adjusted individual P values by calculating their Benjamini–Hochberg critical value using the formula (i/m) × Q, where i is the individual P value’s rank, m is the total number of tests, and Q is the false discovery rate (Benjamini & Hochberg, 1995). We selected 0.05 for the false discovery rate. To test whether the differences in means of the biomass of Macrocystis in the surface 2 m and 1.41 m of the water column (random factors) varied across both depth and slack tides (fixed factors), two-way ANOVA was used. The differences in the means of block dissolution and seawater nutrients were determined using repeated-measures ANOVA; fixed factors were bed position (interior, middle, and exterior) and depth from the surface (2, 6, and 10 m) and measures were repeated over independent sampling events. The difference in the means of light values (percent surface irradiance) and dose per slack tide (random factors) across slack tides (low and high; fixed factors) and across depth (2 and 10 m; fixed factors) were determined using two-way ANOVA. Regression analysis was used to determine the relationship between canopy biomass (floating and stipitate) and block dissolution within those canopies. Differences in means were determined using Tukey’s honestly significantly different (HSD) post hoc test. Significance was set at the 5% level (α = 0.05). Dissolution and nutrient data fulfilled prerequisites of normality (Kolmogorov–Smirnov test with Lilliefors correction) and equal variance (Levene’s median test) for parametric tests. All statistical analyses were carried out using the software package RStudio® version 1.1.383.


Macrocystis: spatial and tidal characteristics

The density of individuals significantly varied across the reef; density in the interior bed was approximately 350% higher than observed at the middle and exterior bed (see Fig. 1 for reminder of bed locations). Density at the middle and exterior bed was similar (Table 1). The mean biomass of individuals (kg individual−1) increased as depth increased, but kelp biomass per area reef (kg m−2) was statistically similar across the reef despite increases in depth (Table 1). Although the number of stipes per individual increased as depth increased across the reef, stipe density on the substratum (stipes m−2) was the highest at the shallow interior of the reef, because individual counts were highest (Table 1). The percentage of fronds within individuals of Macrocystis that reached the surface of the water column significantly decreased from the interior of the bed to the exterior (Table 1), and the biomass of each frond (kg individual−1) significantly increased as depth increased (Table 1). Mean frond biomass at the interior and middle of the bed is approximately 68% and 8% lower, respectively, than the frond biomass at the exterior of the bed.
Table 1

(Left) Mean density and biomass measurements (± 1 SE) for different Macrocystis thallus parameters at different bed positions (and their respective depth)


Bed position

Bed position statistics

Interior (2 m)

Middle (6 m)

Exterior (10 m)

df, residuals


P 0


Density, individual (no. m−2)

3.00 ± 0.45

1.00 ± 0.32

0.80 ± 0.20





Biomass, each (kg individual−1)

4.20 ± 0.75

12.22 ± 0.91

16.62 ± 3.55



< 0.001


Biomass, reef area (kg m−2)

12.99 ± 1.94

12.24 ± 3.87

13.30 ± 3.32





Biomass, water column (kg m−3)

3.77 ± 0.74

1.62 ± 0.15

1.15 ± 0.07





Stipe count (no. individual−1)

9.00 ± 1.00

15.67 ± 1.53

18.00 ± 1.53





Stipe density (no. m−2)

27.00 ± 3.00

15.67 ± 0.88

14.40 ± 1.22





Surface fronds (% total fronds)

60.65 ± 1.22

32.85 ± 5.95

17.55 ± 10.13





Surface frond length (% depth)

137.68 ± 5.76

114.25 ± 2.53

113.82 ± 8.53



< 0.001


The values calculated for “biomass, water column (kg m−3)” represent the averages between MLW and MHW. (Right) One-way ANOVA statistics testing in the means for each thallus parameter. Original P values (P0) and P values adjusted using the Benjamini–Hochberg procedure (PBH) are both presented, the latter being more conservative considering the use of multiple statistical tests and the only statistical value used when determining significance. Bold P values indicate significant values

n/s not significant

The percent of Macrocystis biomass (% individual weight) in the surface 2 m of the water column varied across the reef (df = 2,15; F = 23.60; P < 0.001), in which interior individuals exhibited a higher percentage of their total biomass near the surface than kelp individuals growing in the middle and exterior bed positions (Table 2). Across the reef, percent biomass in the surface 1.41 m averaged 47.3% (SE = 17.5) at low tide and 23.8% (SE = 11.5) when the tide was high; see Table 2 for specific values for each bed position and Fig. 2 for spatial distribution of biomass. Canopy biomass per individual (in the top 1.41 m) showed little cross-bed variation, regardless of tidal height. The mean biomass of the canopies of individual Macrocystis thalli (kg individual−1; Table 2) in the surface 1.41 m at low tide, pooled across the reef (4.15 kg individual−1, SE = 0.41), was higher (df = 1,15; F = 9.56; P = 0.007) than the biomass at high tide (1.96 kg individual−1, SE = 0.21). Canopy biomass per volume (in the surface 1.41 m) was approximately threefold higher at the bed’s interior compared to the middle and exterior (df = 2,15; F = 18.32; P < 0.001). The mean biomass per volume (Table 2) in the surface 1.41 m at low tide, pooled across the bed (4.34 kg m−3, SE = 1.50), was approximately double the biomass density high tide (2.23 kg m−3, SE = 1.02) (df = 1,15; F = 5.61; P = 0.032).
Table 2

Mean canopy biomass (± 1 SE) for Macrocystis at different depth stratifications during low and high tide

Bed position

Biomass (% individual)

Biomass (kg individual−1)

Biomass (kg m−3)

Low tide

High tide

Low tide

High tide

Low tide

High tide

Top 1.41 m


81.88 ± 5.20

46.74 ± 3.27

3.45 ± 0.29

2.01 ± 0.19

7.34 ± 0.38

4.27 ± 0.14


34.87 ± 3.89

13.42 ± 2.70

4.13 ± 0.34

1.58 ± 0.40

2.93 ± 0.11

1.12 ± 0.03


25.52 ± 2.28

11.26 ± 2.21

4.86 ± 0.36

2.29 ± 0.44

2.75 ± 0.06

1.30 ± 0.03

Top 2.00 m


92.04 ± 3.33

63.45 ± 2.42

3.86 ± 0.13

2.71 ± 0.07

5.79 ± 0.19

4.07 ± 0.10


42.84 ± 2.17

22.35 ± 1.58

5.20 ± 0.11

2.64 ± 0.04

2.60 ± 0.06

1.32 ± 0.02


31.80 ± 7.00

17.03 ± 4.29

5.92 ± 0.41

3.36 ± 0.14

2.37 ± 0.17

1.34 ± 0.06

The first bin is set at 1.41 m, because it is the mean tidal exchange between MLW and MHW at this study site, and, thus, biomass in the top 1.41 m may float at the water’s surface during slack low and high tides most days, excluding when tidal elevations reach higher than MHW. The biomass (kg m−3) indicates the amount of biomass in the water per the reef area below the canopy

Fig. 2

Kite diagrams depicting the mean biomass distribution for Macrocystis individuals across depth during low tide (top) and high tide (bottom). The width of the kites represents mean total biomass at the respective depth. The horizontal, dashed line in the high tide figure indicates MLW; the biomass above MLW is compounded with biomass below MLW as the tide shifts to low (effectively doubling surface biomass)

Macrocystis versus stipitate canopy species

The density (Table S1), biomass (Table S2), and height (Table S3) of each were quantified throughout the reef. Excluding Macrocystis, E. radiata, and M. urvilliana were the most dominant species. Although the collective density of the stipitate canopy species (individual m−2) was significantly higher across depth (df = 4,28; F = 9.63; P < 0.001), Macrocystis was the dominant species in biomass throughout the kelp bed surveyed (Fig. 3). At the bed’s interior, Macrocystis accounted for 81.9% (SE = 14.9) of the total biomass (kg m−2) in the water column, 61.9% (SE = 24.3) at the middle of the kelp bed, and 87.3% (SE = 27.4) at the exterior. Ignoring the floating Macrocystis canopy; however, the biomass (kg m−2) of the stipitate canopy dominated the bottom 1 m (stipitate canopy) of the water column (Fig. 3c), in which stipitate canopy species accounted for 65.2% (SE = 25.7), 89.1% (SE = 13.0), and 71.9% (SE = 10.7) of total biomass (for interior, middle, and exterior reef positions, respectively).
Fig. 3

Cross-bed density and biomass measurements for Macrocystis versus the collective stipitate canopy. A Density, B mean biomass of whole individuals for the entire water column, and C mean biomass in the bottom 1 m (approx. mean height of other browns) of the water column. Error bars: ± 1 SE; (n > 3); the letters above the error bars represent post hoc grouping

Mass-transfer and seawater nutrients

Block dissolution (a measurement of mass-transfer) varied across the reef and by depth (Table 3), in which dissolution rates decreased from the exterior to the interior of the reef and decreased as depth increased. At the centre of the reef (middle of the reef at 6 m depth), the rates of block dissolution were the lowest (Fig. 4), suggesting that a cross-bed and cross-depth interaction magnified bed-scale boundary layer dynamics within the heart of the bed.
Table 3

Two-way ANOVA statistics testing for the differences in the means across both bed position and depth for observed abiotic factors. Level of significance is set at α = 0.05


df, residuals



Block dissolution





 Bed position




Seawater NO3





 Bed position




Seawater NH4+





 Bed position




Light (% surface irradiance)

 Tidal height



< 0.001




< 0.001

 Tidal × depth




Light (mean dose at slack tide)

 Tidal height







< 0.001

Tidal × depth




Fig. 4

Relative mass-transfer rates throughout the kelp bed. The upper right-most (exterior position, 2-m depth) had the highest mass-transfer rate and is, therefore, set at 100%; the other values show the mass-transfer of their respective bed position but relative to 100%, i.e., mean mass-transfer rate at the exterior 6-m depth was 89.7% of the highest mass-transfer rate (100%)

Seawater NO3 and NH4+ concentrations did not significantly vary across the bed or across depth (Table 3). Concentrations of seawater NO3 at the interior, middle, and exterior bed positions (averaged across depth) were 1.68 µM (n = 3, SE = 0.23), 1.35 µM (n = 3, SE = 0.17), and 1.30 µM (n = 3, SE = 0.08), respectively. Concentrations of seawater NH4+ at the interior, middle, and exterior bed positions (averaged across depth) were 0.54 µM (n = 3, SE = 0.12), 0.43 µM (n = 3, SE = 0.09), and 0.63 µM (n = 3, SE = 0.16), respectively.


During the 3-month logger deployment, there were 67 high and 70 low tide events that were appropriate to include in analysis (i.e., occurred only during daylight hours). At both 2- and 10-m depths, the percent surface irradiance reaching the benthos during high tide was significantly lower than during low tide (Table 3); the mean irradiance reaching 2-m depth was 14.38% (SE = 1.27) of that at the surface during high tide, and the mean reaching 2 m during low tide was 27.07% (SE = 3.58). At 10-m depth, irradiance averaged 2.42% (SE = 0.21) at high tide but averaged 3.19% (SE = 0.31) when the tide was low. At both 2- and 10-m depths, the mean dose of light for each tidal slack event (mol photon m2 slack−1) was also significantly lower at high tide compared to low tide (Table 3). The mean dose reaching 2-m depth was 0.379-mol photon m2 slack−1 (SE = 0.040) at high tide and 0.505-mol photon m2 slack−1 (SE = 0.043) at low tide. At 10-m depth, the mean dose was 0.092-mol photon m2 slack−1 (SE = 0.008) at high tide and 0.097-mol photon m2 slack−1 (SE = 0.011) at low tide.


Cross-bed biomass and density

The mean biomass of Macrocystis individuals increased as the depth in which they grew increased; this is expected given that Macrocystis’ floating canopy will span a given water column provided that there is adequate light reaching the benthos for recruitment (see Deysher & Dean, 1986) and that individuals persist through hydrodynamic stressors (see Dayton et al., 1992). Despite the positive relationship between the biomass of individuals and depth, total biomass averaged across reef area (kg m−2 reef), unexpectedly, did not scale with water column depth and instead remained constant. The potential for increased total biomass (kg m−2 reef) with increased depth was likely countered by the pattern of Macrocystis density, which reduced gradually from the bed interior to the bed exterior (Table 1). That total standing biomass (kg m−2 reef) was relatively equivalent across the bed but not density informs on patterns in the architecture of a kelp bed. The biomass of Macrocystis located at the interior portion of the kelp bed was spread across more individuals, resulting in a more even distribution of biomass, and biomass at the exterior was associated with fewer individuals, resulting in patchier biomass distribution (see Fig. 5).
Fig. 5

Visual interpretation of how Macrocystis biomass is partitioned across kelp density for each cross-bed location. The number of circles and cumulative area within circles for each bed position reflect the density and biomass data, respectively, from this work

Of the factors that could explain lower kelp density at the edge of the bed, dislodgment and recruitment are both plausible drivers (see Dayton et al., 1992). Individuals at the edge of the bed experience higher drag and, in this study, dislodgment was observed only at the edge of the bed when opportunistically tagged Macrocystis disappeared between maintenance days. In addition to putative physical stressors, the deeper benthos associated with the bed’s edge receives less light. At this study site, irradiance at 10 m throughout much of the year (0.1–0.4-mol photons m−2 day−1; Desmond et al., 2015) is too low to support recruitment and survivorship (0.4–0.7-photons m−2 day−1 required; Deysher & Dean, 1986). Alternatively, the even distribution of individuals in the shallow, interior could be attributed to low-wave stress and ample light for recruitment (0.50–0.28-mol photons m−2 day−1; Desmond et al., 2015). Why, however, do these individuals grow fewer fronds, potentially lowering their biomass threshold? Although light is sufficient for recruitment (supporting high density), it is possible that once the individual grows into a functional adult, low light from self- and intraspecific shading due to high density retards the development of new fronds within each individual (a self-thinning response, see Lonsdale & Watkinson, 1982). This effect has not been previously reported with frond number, but it has been suggested that self-shading procures less bushy fronds that have lower biomass at the interior of Macrocystis beds (Stewart et al., 2009).

Cross-depth biomass

A disproportionate percent of biomass often resides near the surface. For Macrocystis growing in 9-m depth, for example, North (1971) described that approximately 75% of individual biomass was in the top 2 m of the water column. For individuals growing in 10-m depth in this study, only about 25% (averaged across low and high tidal events) of total thallus biomass was positioned in the surface 2 m of the water column. It is difficult to interpret why the gap between the two proportions is so large, but it may be attributed to either (1) differences in the light environment and the subsequent response of deeper vegetative tissue, (2) fluctuations in biomass relative to the differences in tidal height during tidal exchange, or (3) phenotypic plasticity across independent populations. The site in this study is characterized by clearer waters that are rarely turbid (see Desmond et al., 2015) and it is possible that more vegetative, photosynthetic biomass is maintained at deeper frond sections relative to individuals growing in sites that are frequently turbid (see Scheffer et al., 1992). Kelp growing in turbid sites, therefore, would have a higher percentage of biomass in the top 2 m of the water column relative to kelp in clearer waters. Alternatively, the magnitude of tidal swing could contribute to differences in the proportion of biomass floating at the surface. It is unknown how the lengths of Macrocystis fronds scale with the mean tidal swing associated with the habitat in which they grow. For example, it is possible that with larger tidal swings Macrocystis fronds grow longer to reach the water’s surface during higher tides, thus increasing the total length and biomass that reaches the surface on average. This putative increased growth would then result in more biomass floating at the surface during low tides relative to beds that experience smaller tidal swings. The approximated tidal swing for North’s (1971) study area is 1.63 m, compared to 1.41 m in this study. It is possible that this larger swing could contribute to the proportion of canopy biomass reported that study compared to this study. The interaction between tidal height and the distribution of biomass within the water column is an intriguing concept that has been largely ignored. Rosman et al. (2010), perhaps the only other study that addresses tidal influence on the canopy biomass of Macrocystis, reported a 10–20% change in biomass at the surface. In this study, the canopy biomass (kg m−3) subject to the mean tidal swing during high tide was approximately 52% lower (averaged across the bed) than the biomass measured at low tide, or about 41% in the top 2 m.

Canopy interactions with abiotic factors

The 1-m tall stipitate canopy and Macrocystis’ floating canopy provide three-dimensional structure that directly modify water motion and thus rates of mass-transfer. As expected, mass-transfer rates decreased across the bed (exterior > middle > interior) and across depth (2 m > 6 m > 10 m). These results are unsurprising because water motion is often stronger at the top (typically wave-induced) and at the edges of kelp beds (typically current-induced) where there are fewer or no biogenic structures inhibiting water flow (Jackson & Winant, 1983). One would expect increased productivity at regions where mass-transfer is higher (see Hepburn et al., 2007). In this study, for example, the highest biomasses of individual Macrocystis was associated with the highest rates of mass-transfer (i.e., Macrocystis at the interior bed). It is important to consider that while higher standing stocks can be an indication of increased productivity, it can also be misleading. It is possible to underestimate total bed production using biomass as a sole metric because quantifying partial loss of kelp biomass (e.g., erosion and dislodgement) is difficult (see Reed et al., 2008). Kelp at the edge of the bed, for example, could have dynamic turnover rates via frond loss and regrowth, but still have small standing biomass. Quantifying biomass at one point in time may misrepresent recent biomass and total carbon fixation (sensu Mann, 1972; Brady-Campbell et al., 1984).

The concentrations of seawater nutrients (NO3 and NH4+) did not vary across the bed or across depth, which supports the hypothesis that the residency time of water masses moving through kelp beds is low, corroborating previous work (Stephens & Hepburn, 2014). Finally, the data do not support our hypothesis that light reaching the benthos at high tide is reduced by the increased Macrocystis biomass that accumulates at the water’s surface during low tide. This was surprising for the light environment at 2 m, particularly, because the biomass of Macrocystis in this reef location was spread more evenly across the canopy, and such macroalgal structure is more likely to have greater light capturing ability than populations composed primarily of larger kelp at lower densities (see Gerard, 1984). At the 2-m-depth band, the path length at low tide is much shorter than that at high tide (approx. 0.6 m versus 2.0 m) and attenuation reduces as pathlength shortens, so it is logical for more light to reach the benthos during low tide. Since there was generally more light at low tide with the presence of accumulated kelp canopy, it is implied that the shading effect from kelp biomass was weaker than the effect of reduced light attenuation associated with a shorter pathlength at low tide. It is also possible that the mechanical movement of the fronds via waves in this semi-exposed site facilitated the reduced effect of shading (e.g., Wing et al., 1993).

Kelp beds and their biomass are often considered across two-dimensional reef space, ignoring potential interactions between depth-dependent responses or patterns, and how these relationships change throughout a tidal cycle. This study highlights the structural complexity of kelp beds, informing upon the distribution of kelp biomass and the effect that such architecture has on the surrounding environment.



We thank the staff and students of the University of Otago Marine Science Department, the Portobello Marine Laboratory, and Liina Pajusalu for aiding with fieldwork. We also thank those that provided critical reviews of this work, the constructive comments from which strengthened the final product. This work was funded by a University of Otago International Postgraduate Scholarship awarded to TAS, by postgraduate research funding awarded to TAS and MJD, and by supplemental departmental funding provided to CDH.

Supplementary material

10750_2018_3788_MOESM1_ESM.docx (35 kb)
Supplementary material 1 (DOCX 35 kb)


  1. Arkema, K. K., D. C. Reed & S. C. Schroeter, 2009. Direct and indirect effects of giant kelp determine benthic community structure and dynamics. Ecology 90(11): 3126–3137.CrossRefGoogle Scholar
  2. Barrales, H. L. & C. S. Lobban, 1975. The comparative ecology of Macrocystis pyrifera, with emphasis on the forests of Chubut, Argentina. The Journal of Ecology 63: 657–677.CrossRefGoogle Scholar
  3. Benjamini, Y. & Y. Hochberg, 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society. Series B (Methodological) 57: 289–300.Google Scholar
  4. Brady-Campbell, M. M., D. B. Campbell & M. M. Harlin, 1984. Productivity of kelp (Laminaria spp.) near the southern limit in the northwestern Atlantic Ocean. Marine Ecology Progress Series 18(1): 79–88.CrossRefGoogle Scholar
  5. Connell, J. H., 1989. Some processes affecting the species composition in forest gaps. Ecology 70(3): 560–562.CrossRefGoogle Scholar
  6. Dayton, P. K., 1985. Ecology of kelp communities. Annual Review of Ecology and Systematics 16: 215–245.CrossRefGoogle Scholar
  7. Dayton, P. K. & M. J. Tegner, 1984. Catastrophic storms, El Niño, and patch stability in a southern California kelp community. Science 224(4646): 283–285.CrossRefGoogle Scholar
  8. Dayton, P. K., V. Currie, T. Gerrodette, B. D. Keller, R. Rosenthal & D. V. Tresca, 1984. Patch dynamics and stability of some California kelp communities. Ecological Monographs 54(3): 253–289.CrossRefGoogle Scholar
  9. Dayton, P. K., M. J. Tegner, P. E. Parnell & P. B. Edwards, 1992. Temporal and spatial patterns of disturbance and recovery in a kelp forest community. Ecological Monographs 62(3): 421–445.CrossRefGoogle Scholar
  10. Dayton, P. K., M. J. Tegner, P. B. Edwards & K. L. Riser, 1999. Temporal and spatial scales of kelp demography: the role of oceanographic climate. Ecological Monographs 69(2): 219–250.CrossRefGoogle Scholar
  11. Dean, T. A. & F. R. Jacobsen, 1984. Growth of juvenile Macrocystis pyrifera (Laminariales) in relation to environmental factors. Marine Biology 83(3): 301–311.CrossRefGoogle Scholar
  12. Desmond, M. J., D. W. Pritchard & C. D. Hepburn, 2015. Light limitation within southern New Zealand kelp forest communities. PLoS ONE 10(4): e0123676.CrossRefGoogle Scholar
  13. Deysher, L. E. & T. A. Dean, 1986. In situ recruitment of sporophytes of the giant kelp, Macrocystis pyrifera (L.) CA Agardh: effects of physical factors. Journal of Experimental Marine Biology and Ecology 103(1): 41–63.CrossRefGoogle Scholar
  14. Falter, J. L., M. J. Atkinson & C. F. Coimbra, 2005. Effects of surface roughness and oscillatory flow on the dissolution of plaster forms: evidence for nutrient mass transfer to coral reef communities. Limnology and Oceanography 50(1): 246–254.CrossRefGoogle Scholar
  15. Fram, J. P., H. L. Stewart, M. A. Brzezinski, B. Gaylord, D. C. Reed, S. L. Williams & S. MacIntyre, 2008. Physical pathways and utilization of nitrate supply to the giant kelp, Macrocystis pyrifera. Limnology and Oceanography 53(4): 1589.CrossRefGoogle Scholar
  16. Gaylord, B., J. H. Rosman, D. C. Reed, J. R. Koseff, J. Fram, S. MacIntyre & S. G. Monismith, 2007. Spatial patterns of flow and their modification within and around a giant kelp forest. Limnology and Oceanography 52(5): 1838–1852.CrossRefGoogle Scholar
  17. Gaylord, B., K. J. Nickols & L. Jurgens, 2012. Roles of transport and mixing processes in kelp forest ecology. The Journal of Experimental Biology 215(6): 997–1007.CrossRefGoogle Scholar
  18. Gerard, V. A., 1984. The light environment in a giant kelp forest: influence of Macrocystis pyrifera on spatial and temporal variability. Marine Biology 84(2): 189–195.CrossRefGoogle Scholar
  19. Goldberg, N. A. & G. A. Kendrick, 2004. Effects of island groups, depth, and exposure to ocean waves on subtidal macroalgal assemblages in the Recherche Archipelago, Western Australia. Journal of Phycology 40(4): 631–641.CrossRefGoogle Scholar
  20. Graham, M. H., J. A. Vasquez & A. H. Buschmann, 2007. Global ecology of the giant kelp Macrocystis: from ecotypes to ecosystems. Oceanography and Marine Biology 45: 39.Google Scholar
  21. Heath, R. A., 1975. Oceanic circulation and hydrology off the southern half of South Island, New Zealand, Vol. 72. New Zealand Oceanographic Institute, Wellington.Google Scholar
  22. Hepburn, C. D., J. D. Holborow, S. R. Wing, R. D. Frew & C. L. Hurd, 2007. Exposure to waves enhances the growth rate and nitrogen status of the giant kelp Macrocystis pyrifera. Marine Ecology Progress Series 339: 99.CrossRefGoogle Scholar
  23. Hurd, C. L., 2000. Water motion, marine macroalgal physiology, and production. Journal of Phycology 36(3): 453–472.CrossRefGoogle Scholar
  24. Irving, A. D. & S. D. Connell, 2006. Predicting understorey structure from the presence and composition of canopies: an assembly rule for marine algae. Oecologia 148(3): 491–502.CrossRefGoogle Scholar
  25. Jackson, G. A. & C. D. Winant, 1983. Effect of a kelp forest on coastal currents. Continental Shelf Research 2(1): 75–80.CrossRefGoogle Scholar
  26. Jokiel, P. L. & J. I. Morrissey, 1993. Water motion on coral reefs: evaluation of the ‘clod card’ technique. Marine Ecology Progress Series 93: 175–181.CrossRefGoogle Scholar
  27. Kain, J. M., 1989. The seasons in the subtidal. British Phycological Journal 24(3): 203–215.CrossRefGoogle Scholar
  28. Long, M. H., J. E. Rheuban, P. Berg & J. C. Zieman, 2012. A comparison and correction of light intensity loggers to photosynthetically active radiation sensors. Limnology and Oceanography: Methods 10(6): 416–424.Google Scholar
  29. Lonsdale, W. M. & A. R. Watkinson, 1982. Light and self-thinning. New Phytologist 90(3): 431–445.CrossRefGoogle Scholar
  30. Lowe, R. J., J. R. Koseff, S. G. Monismith & J. L. Falter, 2005. Oscillatory flow through submerged canopies: 2. Canopy mass transfer. Journal of Geophysical Research: Oceans 110(C10): C10017.CrossRefGoogle Scholar
  31. Mann, K. H., 1972. Ecological energetics of the seaweed zone in a marine bay on the Atlantic coast of Canada. I. Zonation and biomass of seaweeds. Marine Biology 12(1): 1–10.Google Scholar
  32. North, W. J., 1971. Growth of individual fronds of the mature giant kelp, Macrocystis. Nova hedwigia, Beihefte.Google Scholar
  33. Pearse, J. S. & A. H. Hines, 1979. Expansion of a central California kelp forest following the mass mortality of sea urchins. Marine Biology 51(1): 83–91.CrossRefGoogle Scholar
  34. Porter, E. T., L. P. Sanford & S. E. Suttles, 2000. Gypsum dissolution is not a universal integrator of ‘water motion’. Limnology and Oceanography 45(1): 145–158.CrossRefGoogle Scholar
  35. Reed, D. C. & M. S. Foster, 1984. The effects of canopy shadings on algal recruitment and growth in a giant kelp forest. Ecology 65(3): 937–948.CrossRefGoogle Scholar
  36. Reed, D. C., A. Rassweiler & K. K. Arkema, 2008. Biomass rather than growth rate determines variation in net primary production by giant kelp. Ecology 89(9): 2493–2505.CrossRefGoogle Scholar
  37. Rosman, J. H., S. G. Monismith, M. W. Denny & J. R. Koseff, 2010. Currents and turbulence within a kelp forest (Macrocystis pyrifera): insights from a dynamically scaled laboratory model. Limnology and Oceanography 55(3): 1145.CrossRefGoogle Scholar
  38. Scheffer, M., M. R. de Redelijkheid & F. Noppert, 1992. Distribution and dynamics of submerged vegetation in a chain of shallow eutrophic lakes. Aquatic Botany 42(3): 199–216.CrossRefGoogle Scholar
  39. Schiel, D. R. & M. S. Foster, 1986. The structure of subtidal algal stands in temperate waters. Oceanography and Marine Biology Annual Review 24: 265–307.Google Scholar
  40. Spies, T. A. & J. F. Franklin, 1989. Gap characteristics and vegetation response in coniferous forests of the Pacific Northwest. Ecology 70(3): 543–545.CrossRefGoogle Scholar
  41. Stephens, T. A. & C. D. Hepburn, 2014. Mass-transfer gradients across kelp beds influence Macrocystis pyrifera growth over small spatial scales. Marine Ecology Progress Series 515: 97–109.CrossRefGoogle Scholar
  42. Stewart, H. L. & R. C. Carpenter, 2003. The effects of morphology and water flow on photosynthesis of marine macroalgae. Ecology 84(11): 2999–3012.CrossRefGoogle Scholar
  43. Stewart, H. L., J. P. Fram, D. C. Reed, S. L. Williams, M. A. Brzezinski, S. MacIntyre & B. Gaylord, 2009. Differences in growth, morphology and tissue carbon and nitrogen of Macrocystis pyrifera within and at the outer edge of a giant kelp forest in California, USA. Marine Ecology Progress Series 375: 101–112.CrossRefGoogle Scholar
  44. Wernberg, T., G. A. Kendrick & B. D. Toohey, 2005. Modification of the physical environment by an Ecklonia radiata (Laminariales) canopy and implications for associated foliose algae. Aquatic Ecology 39(4): 419–430.CrossRefGoogle Scholar
  45. Wing, S. R. & M. R. Patterson, 1993. Effects of wave-induced lightflecks in the intertidal zone on photosynthesis in the macroalgae Postelsia palmaeformis and Hedophyllum sessile (Phaeophyceae). Marine Biology 116(3): 519–525.CrossRefGoogle Scholar
  46. Wing, S. R., J. J. Leichter & M. W. Denny, 1993. A dynamic model for wave-induced light fluctuations in a kelp forest. Limnology and Oceanography 38(2): 396–407.CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Department of Marine ScienceUniversity of OtagoDunedinNew Zealand
  2. 2.College of Fisheries and Ocean SciencesUniversity of Alaska FairbanksJuneauUSA

Personalised recommendations