Spatial patterns of zooplankton communities In Swedish mountain and boreal lakes.

(1)

Examensarbete, 15 credits Bachelor’s Degree Thesis in Biology

Vt 2020

SPATIAL PATTERNS OF

ZOOPLANKTON COMMUNITIES

In Swedish mountain and boreal lakes.

Juan Manuel de la Quintana.

(2)

1

Abstract :

Zooplankton is crucial for lake ecosystems as it is an important component in most of the food webs of these ecosystems. However, environmental changes have caused reductions in ecosystems nutrients and stoichiometry, which directly affects phytoplankton’s development and quality. Studies were carried out in 33 different lakes throughout Sweden, divided into 4 regions (Abisko, Jämtland, Västerbotten and Värmland) to assess whether differences in environmental characteristics induced differences in phytoplankton abundance and zooplankton communities. Using data from 3 different seasons of the year, I found that zooplankton composition differed between northern and southern regions, and greater differences were found between the mountain regions than between the boreal regions.

Dissolved inorganic N (DIN) and dissolved inorganic N to total P ratio (DIN:TP) concentrations were lower in northern regions than in southern regions. Phytoplankton biomass increased with dissolved organic carbon (DOC), likely through the positive effects of DOC on overall nutrient availability. DOC concentrations were higher in boreal regions than in mountain regions, as also happens with chlorophyll-a. Positive correlations between DOC and 3 different zooplankton genera were proven, whereas 3 different zooplankton taxa were correlated with DIN:TP (two negatively and one positively). Lakes with lower DIN:TP ratios had higher abundances of calanoids, which were the major contributor of the dissimilarity in zooplankton composition among the regions. Therefore, the DIN:TP ratio possibly has stronger effects than DOC on zooplankton composition in Swedish oligotrophic lakes. But further increases in DOC concentrations will likely reduce the differences in zooplankton composition caused by the declines in lake DIN:TP observed in this study.

Key words: Lake, zooplankton, phytoplankton, DOC and DIN:TP.

(3)

2

Table of Contents : Chapters

:

1.

Introduction

……….……3

2.

Material and methods

………..……….4

a.

Field sampling, nutrients, chlorophyll-a and zooplankton analyses

……….…………..….4

b.

Data analyses

……….……….6

3.

Results

………..……….…….7

a.

Overall zooplankton composition

………..….7

b.

Zooplankton abundance

……….……..…...…….8

c.

Lake environmental characteristics

.………..…..8

d.

Chlorophyll-a concentration

………….…………..…….………..10

e.

Correlations of chlorophyll a and zooplankton abundances with lake DOC and DIN:TP

………..….………10

4.

Discussion

………...11

5.

Acknowledgements

……….……….…..12

6.

References

……….……..…...12

(4)

3

Introduction:

Global environmental changes, such as climate warming and decreases in atmospheric nitrogen (N) and sulfur deposition, have caused reductions in ambient nutrients – i.e. N and phosphorus (P) – in northern high-latitude freshwater ecosystems (Huser et al. 2018; Isles et al. 2018). Consequently, widespread declines in dissolved inorganic N to total P ratio (DIN:TP) have been observed in Swedish lakes (Isles et al. 2018), which potentially affect the nutrient limitation regimes of the lakes (Bergström et al. 2020; Isles et al. 2018). Contrarily, climate change and recent recovery from sulfur deposition have enhanced the export of terrestrial colored dissolved organic carbon (DOC) from catchments, resulting in brownification of the northern surface freshwaters (Monteith et al. 2007; Larsen et al. 2011; Isles et al. 2018). For example, a warmer and wetter climate could promote vegetation growth and decomposition in catchments, subsequently enhancing the DOC supply to lakes (Finstad et al, 2016; de Wit et al.

2016). Thus, Swedish lakes are challenged by declines in ambient nutrients and DIN:TP ratio combined with increases in DOC, but the impacts of these combined ongoing changes on the biological communities, particularly those in the oligotrophic lakes, are largely unknown.

Northern lake food webs generally depend on the trophic support by phytoplankton (Jansson et al. 2007; Sterner & Hessen, 1994) although algae’s role is also of vital importance (Taiplae et al. 2013) However, phytoplankton may become more N-limited and their biomass may be reduced upon more intensified nutrient depletion and declines in lake DIN:TP ratio, as phytoplankton are usually nutrient limited (Deininger et al. 2017). Increases in DOC, i.e.

browning, can increase or reduce phytoplankton biomass, as DOC has positive effects on nutrient availability (stimulates primary production) and negative effects on light climate (inhibits primary production) in lakes. Based on whole-lake studies in arctic and boreal Sweden, Bergström & Karlsson (2019) found a unimodal relationship between phytoplankton biomass and lake DOC concentrations. In low-DOC lakes, under a certain threshold (i.e. 10.6 mg/L), browning will stimulate the phytoplankton biomass by increasing the nutrient availability. But in high-DOC lakes, browning will reduce the phytoplankton biomass by limiting light and photosynthesis thus reducing the availability of oxygen due to the increase of the bacterial activity (Zwart et al., 2016). Deininger et al. (2017) found that phytoplankton biomass in Swedish unproductive boreal lakes increases with N fertilization and DIN:TP, but phytoplankton biomass in the N-fertilized lakes decreases with increasing DOC concentrations. Hence, there are likely interacting effects of DIN:TP and DOC on the plankton community. Declines in DIN:TP combined with increases in DOC may negatively affect zooplankton in oligotrophic lakes by hampering phytoplankton development. With all this, it is to be expected that with the changes that have occurred from the deposition of DIN and dissolved organic matter (DOM), changes will arise in lake stoichiometry in terms of N:P ratio (Isles et al. 2017).

Zooplankton composition may change with the phytoplankton community, as zooplankton taxa differ in feeding modes and nutritional requirements (e.g. essential fatty acids [EFA];

Persson et al. 2007). Edible flagellates dominate the phytoplankton communities in northern lakes (Deininger et al. 2017), and they contain high food quality (i.e. high EFA amounts) for zooplankton. A reduction in the phytoplankton biomass, as a result of the combination between the decline in DIN:TP and the increase in DOC, will potentially reduce the dietary EFA availability for zooplankton. For example, copepods have a higher demand for EFA than cladocerans (Persson et al. 2007). So the relative abundance of copepods may be negatively affected in lakes with low DIN:TP and high DOC, where the relative abundance of EFA-rich phytoplankton taxa is also expected lower. Nevertheless, the changes in zooplankton composition resulted from the decline in DIN:TP and intensified lake browning are yet to be quantified.

(5)

4

In this study, the zooplankton communities of a total of 33 lakes from mountain and boreal regions (each with one region in the north and one in the south) in Sweden were investigated.

Information about each lake (i.e. region, lake name, coordinates, lake area and catchment characteristics) can be found in Table 1. Mountain lakes generally have lower DOC concentrations than lakes in lower elevations in Sweden (Bergström et al. 2013). There is a north-south gradient in increasing DIN:TP ratio in Swedish lakes attributed to the increasing atmospheric N deposition towards the south (Bergström et al. 2005). These regions were Abisko (Mountain-North; 9 lakes), Jämtland (Mountain-South; 6 lakes), Västerbotten (Boreal- North; 9 lakes) and Värmland (Boreal-South; 9 lakes). Within each region, lakes with different levels of DOC were used, therefore the lakes in this study covered a gradient at both local and regional scales. My goal was to test the following hypotheses:

1. The overall zooplankton composition differed between regions, with greater differences between the northern and the southern regions than between mountain and boreal regions.

2. There are regional differences in environmental characteristics, which could explain the different concentrations of chlorophyll-a.

3. Chlorophyll-a concentration is positively correlated with DOC if its concentration is below 10.6 mg/L, but negatively correlated if DOC concentrations are over the threshold. Furthermore, zooplankton abundance is negatively correlated with DOC concentration and both zooplankton and chlorophyll-a are positively correlated with lake DIN:TP ratio.

4. There are regional differences in chlorophyll-a concentration that follow the spatial patterns in lake water chemistry.

Materials and methods :

Lakes were sampled once each in early- (beginning of the growing season), mid- (after the onset of stratification), and late-summer (before circulation). Environmental characteristics of the lakes, e.g. DIN, total phosphorus (TP), DIN:TP, DOC, phytoplankton biomass (i.e.

chlorophyll-a concentration), carbon-specific ultraviolet absorption (SUVA), vertical light extinction coefficient (Kd), water temperature, etc., were measured according to Bergström et al. (2020). This project will focus on the dominant zooplankton taxa including copepods and cladocerans.

Field sampling, nutrients, chlorophyll-a and zooplankton analyses:

Methods for measurements of the lake environmental characteristics are described in Bergström et al. (2020). In brief, an YSI Pro-DO sensor was utilized to measure the temperature profiles. Photosynthetically active radiation (PAR) was measured using a LiCOR Li-250 spherical quantum sensor. Kd was calculated as the slope of the linear regression of the logarithm of PAR versus depth. Depending on the stratification of the lakes, they were sampled differently: in Abisko it was made at a depth of one meter, since most of the lakes were not stratified during the summer, while in Jämtland, where they were stratified during the summer, samples were taken at a depth of 0.5 m. Samples of both ammonium (NH4+) and nitrate (NO3-), DOC and total dissolved N (TDN) were taken from the water and then filtered after sampling. The NH4+ and NO3- samples were used to calculate the DIN by the sum of both values. Chlorophyll-a samples were kept in the dark before filtration and the DOC samples were acidified.

(6)

5

RegionLake nameLongitudeLake area (ha) Catchment area (ha) Catchment mean slope (°) % Deciduous% Coniferous % Mixedwood Wetland area

VärmlandÖrtjärnen13,3315,33124,2814,449,7444,428,620,00VärmlandHemsjön13,3424,6864,1913,073,9357,6321,432,18

VärmlandIgeltjärnen13,299,4753,2812,390,0089,644,8910,78VärmlandHotlamm12,6232,66263,0017,100,1169,648,072,94VärmlandDjupen12,597,38463,5911,024,2660,8410,8514,95

VärmlandIsakstjärn12,641,5676,7316,170,0076,794,605,22VärmlandStora Abbortjärnet12,651,78277,219,950,4182,732,197,76

VärmlandMarkustjärnet12,702,15175,5811,730,0082,390,5710,44VärmlandGöptjärnet12,751,90125,8713,770,0079,755,836,11VästerbottensNästjärn18,801,034,4411,200,0099,970,0011,11

VästerbottensÖvre Btj18,785,00292,828,075,2659,7614,0726,24VästerbottensByxrivarlidvägen18,761,23178,4910,987,8561,5813,4423,25

VästerbottensGålgotjärn18,715,9583,4412,423,9551,4215,4722,16VästerbottensEnhörningen19,0511,9166,1710,313,9270,373,3615,50VästerbottensGäddtjärn19,0635,37395,2410,835,0361,659,9919,23

Västerbottensunknown18,761,9017,539,750,0097,600,0011,34VästerbottensNedre Skarda18,771,3721,419,320,0075,850,0026,73

VästerbottensÖvre Skarda18,852,0513,114,710,0063,120,0059,85AbiskoSolbackasjön18,913,697,7512,3151,590,000,000,00AbiskoHästskosjön18,975,6941,3013,3848,360,000,000,00

AbiskoVouskojavri19,1069,301285,8419,5029,090,080,001,34AbiskoLångsjön19,1511,4747,7214,2734,480,000,000,00

AbiskoAlmberga19,155,7928,4712,3437,590,000,000,00AbiskoLillsjön bredvid19,150,573,0712,0515,630,000,000,00AbiskoKaisepaktesjön19,4114,64932,8323,2524,690,000,370,15

AbiskoKoukkelsjön18,581,648,0110,8119,350,000,000,00AbiskoBanansjön18,637,7541,719,4839,280,000,000,00

JämtlandsHägglidstjärnarna12,234,2261,149,4996,392,521,094,95JämtlandsBergtjärnen12,2613,5345,0311,9685,567,067,200,00JämtlandsVargtjärnen12,2712,15439,036,1796,451,831,550,72

JämtlandsLång-Björsjön12,2613,72921,574,9099,260,410,330,54JämtlandsDörrstjärnen12,305,12615,404,8999,800,180,020,48

JämtlandsGravatjärnen12,264,59154,725,7095,042,812,150,00JämtlandsHästbäckstjärnen12,284,82110,285,0695,042,312,640,42 Table 1: Summary of the catchment’s characteristics. This table shows different characteristics of the basin of each lake analyzed such as: the name of the lake, itscoordinates, its area, the area of the catchment and its slope, the proportion of riparian vegetation (percentage of deciduous vegetation, percentage of conifers and percentage of mixed forests) and the wetlands area.

(7)

6

TDN and DOC measurement were done thanks to an IL-550 TOC / TN analyzer. To measure total P, it was necessary to expose it to sulfate peroxide and make it pass through an autoclave and then treated as soluble reactive phosphate. Using a spectrophotometer, the absorbance was measured at 250, 365, and 440 nm. SUVA was estimated using an absorbance of 250 nm and divided by the DOC concentration. Chlorophyll-a samples could be measured thanks to a spectrophotometer with excitation and emission wavelengths of 433 and 673 nm.

To collect the zooplankton samples a net of 100 µm mesh size was used at the deepest part of the lake, carrying out a vertical haul of the net, starting approximately 1 m above lake bottom (Deininger et al. 2017). Samples were kept with Lugol’s iodine in the dark at 6ºC to better preserve them (Deininger et al. 2017). Zooplankton were counted using inverted microscopy and dry weight (DW) was estimated thanks to length measurements and length-dry weight regressions (Deininger et al. 2017). Once the zooplankton taxa were identified and their dry weight was calculated, we were able to estimate the relative percent dry weight of specific genera. For those nauplii larvae that could not be identified to genus level they were classified simply as nauplii. The rest of them were classified in two different levels, which are: genera of brachiopods (Bosmina, Chydorus, Daphnia, Diaphanosoma, Ceriodaphnia, Holopedium, Limnosida and Polyphemus) and copepod orders (Calanoida and Cyclopoida).

Data analyses:

Seasonal means of the environmental characteristics and the zooplankton abundance data were used for all statistical analyses, as in this study I focused on the regional comparisons but not the seasonal patterns. The seasonal means of TP, DIN, DIN:TP, DOC, TDN, Kd, surface temperature and chlorophyll-a), were log10-transformed to approximate a normal distribution and for homoscedasticity. SUVA data was normally distributed and so it was not log10- transformed. Both the absolute abundances and the relative abundances of the zooplankton taxa were compared among regions by using Kruskal-Wallis tests and post hoc Dunn’s tests (for hypothesis 1), since these zooplankton data, even after log10-transformation, were not normally distributed.

I conducted non-metric multi-dimensional scaling (nMDS) based on Bray-Curtis dissimilarity to analyse how similar or different of the overall zooplankton compositions were among the regions (again for hypothesis 1). NMDS is an ordination technique that analyses multiple variables (or taxa in this case) and visualizes the results into a representation with a reduced number of dimensions. Within the nMDS ordination plot, lake samples closer to each other are more similar in their zooplankton composition. The Bray-Curtis dissimilarity takes into account not only abundance, but also the presence or absence of taxa. The analysis of similarity (ANOSIM) test was then used to compare the zooplankton composition among regions. Lastly I used similarity percentages (SIMPER) for pairwise comparisons of regions and identifying the zooplankton taxa that were most responsible for the regional differences in zooplankton composition. SIMPER calculates the average contribution of each zooplankton taxon to the average overall Bray-Curtis dissimilarity.

One-Way ANOVA and post hoc Tukey tests were then used to compare the environmental variables among the four study regions (for testing hypothesis 2). The concentrations of nutrients (i.e. TP, DIN and TDN) were also compared among regions, as they might have affected chlorophyll-a concentrations and also the abundance of zooplankton in the lakes.

Furthermore, another One-Way ANOVA and post hoc Tukey test were carried out to compare

(8)

7

if the means of chlorophyll-a concentrations were similar or different across regions in order to test hypothesis 4.

Bergström et al. (2020) found that the DOC concentrations could be negatively correlated with the DIN:TP ratio in Swedish mountain lakes. Increasing DOC concentrations could limit light and productivity (Ask et al. 2009; Karlsson et al. 2009), and declines in DIN:TP ratio could enhance N-limitation that reduces phytoplankton development in northern oligotrophic lakes (Deininger et al. 2017). Thus, I tested whether chlorophyll-a concentrations and the zooplankton abundances were associated to DOC and DIN:TP by using Pearson correlation (for chlorophyll-a) or Spearman’s rank correlation analyses (for zooplankton abundances, which were not normally distributed) for hypothesis 3.

Results :

Overall zooplankton composition:

The nMDS and ANOSIM results (Fig. 1) showed apparent regional differences in lake zooplankton composition (nMDS: 2D stress = 0.15; ANOSIM: Global R = 0.45, p = 0.001). The differences were greater between the northern and southern regions, rather than between mountain and boreal regions in the north nor in the south. Differences in zooplankton composition were greatest between Abisko lakes and Jämtland lakes based on the nMDS ordination.

Figure 1: Results of the nMDS analysis on lake zooplankton composition. Open symbols are individual lakes. Solid symbols are centroids of lakes in individual regions. Solid and broken lines indicate hull areas of individual mountain and boreal regions, respectively, on the nMDS ordination space. Black, northern regions; grey, southern regions. A greater distance between the points indicates greater dissimilarity in zooplankton composition between the lakes. The nMDS results show that there were overall differences in lake zooplankton composition between the northern and the southern regions. Also, differences between the northern and the southern mountain regions were greater than those between the northern and the southern boreal regions.

From ANOSIM I obtained the R statistic, with a value of 0.45, a significance level of 0.001 and a total of 999 permutations, statistically denoting that indeed there are differences between

(9)

8

the studied regions. Based on SIMPER, Calanoida contributed most to the dissimilarity in lake zooplankton composition between regions, especially between Abisko and Jämtland (Table 2).

Table 2: Cumulative contributions of the 4 most influential organisms to make significant differences in zooplankton abundance between two compared regions based on the value of their mean contributions to the average overall Bray-Curtis dissimilarity.

Zooplankton abundance:

In total, two copepod orders (Calanoida and Cyclopoida) and eight cladoceran genera (Bosmina, Chydorus, Daphnia, Diaphanosoma, Ceriodaphnia, Holopedium, Limnosida and Polyphemus) were present in the lake zooplankton samples. The absolute and relative abundance of Calanoida, Diaphansoma and Ceriodaphnia differed between the regions (Kruskal-Wallis tests: all p ≤ 0.011). The relative abundance of Cyclopoida also differed between regions (Kruskal-Wallis tests: all p ≤ 0.044) (Fig. 2 and Fig. 3). But no differences in absolute and relative abundance of Bosmina, Chydorus, Daphnia, Holopedium, Limnosida and Polyphemus among regions were found (Kruskal-Wallis tests: all p > 0.05).

Lake environmental characteristics:

To test whether or not if there are significant differences between the means of the environmental variables of the lakes One Way ANOVAs were again used. The following environmental parameters differed across regions: SUVA, TP, DIN, DIN:TP, DOC, TDN, Kd and surface temperature (one way ANOVA: all p = 0.0001 (one way ANOVA: all p = 0.0001, F ratio rank = 83.98-10.19) . Table 3 summarizes what the differences between regions are like for each particular parameter.

The results of the Spearman correlation between the surface temperature and the zooplankton data had the following results: On one hand we have those taxa with a p ≥ 0.174, which were Bosmina, Calanoida, Chydorus, Daphnia, Holopedium, Limnosida and Polyphemus.

Whereas, on the other, we have only a pair of organisms with significant p values which were

Compared regions Taxa Mean Abisko-Jämtland Calanoida 0,54

Bosmina 0,10 Holopedium 0,05 Daphnia 0,05 Abisko-Värmland Calanoida 0,37 Bosmina 0,08 Holopedium 0,06 Cyclopoida 0,05 Abisko-Västerbottenn Calanoida 0,26 Cyclopoida 0,07 Bosmina 0,07 Daphnia 0,04 Jämtland-Värmland Calanoida 209,00 Holopedium 133,80 Cyclopoida 86,85 Bosmina 67,54 Jämtland-Västerbotten Calanoida 341,30 Cyclopoida 150,00 Daphnia 61,65 Holopedium 58,36 Värmland-Västerbotten Calanoida 199,00 Cyclopoida 101,60 Holopedium 72,29 Daphnia 44,42

(10)

9

Diaphansoma and Ceriodaphnia (Spearman correlation: all p ≤ 0.009, Spearman’s rank r = 0.443-0.603).

Figure 2: Boxplot that illustrates the results of the Kruskal-Wallis test and post hoc Dunn’s test for the zooplankton abundance: (a) Calanoida, (b) Diaphanasoma and (c) Ceriodaphnia. Regions with the same letters are not significantly different based on Dunn’s test (p>0.05). Circles and stars represent outliers of different magnitude, circles being out values not as big as stars.

Figure 3: Boxplot that illustrates the results of the Kruskal-Wallis test and post hoc Dunn’s test for percent dry weight of zooplankton. Regions with the same letters are not significantly different based on Dunn’s test (p>0.05).

(11)

10

Circles and stars represent outliers of different magnitude, circles being out values not as big as stars. a) is percent dry weight of cladocerans, b) is percent dry weight of Cyclopoida, c) is percent dry weight of Diaphanasoma and d) is percent dry weight of Ceriodaphnia.

Chlorophyll-a concentration:

Results from ANOVA showed that the differences in chlorophyll-a concentration across regions (Table 3) were marginally significant (F = 2.90, DF = 33, p = 0.0512). Based on Tukey comparisons, the chlorophyll-a concentrations were higher in Värmland lakes than the Abisko lakes (p <0.05). But the chlorophyll-a concentrations in Västerbotten lakes and Jämtland lakes did not differ from those in Värmland lakes and Abisko lakes (p > 0.05).

Table 3: Results of ANOVA and post hoc Tukey’s tests on regional comparisons of lake environmental characteristics. For each variable, regions with different letters are significantly different (Tukey’s test: p < 0.05).

Parameter Region Mean SD

SUVA Abisko B 0,02 0,01

Västerbotten A 0,04 0,01

Jämtland A 0,05 0,01

Värmland A 0,04 0,01

TP Abisko B 0,65 0,15

Jämtland B 0,65 0,06

Värmland B 0,81 0,24

DIN Abisko B C 0,74 0,19

Västerbotten C 0,55 0,27

Värmland A B 1,01 0,26

DIN:TP Abisko A 0,41 0,25

Västerbotten B -0,32 0,37

DOC Abisko B 0,72 0,16

Jämtland B 0,69 0,20

TDN Abisko B 257,41 103,81

Jämtland C 142,56 41,23

Värmland A 399,69 79,66

Kd Abisko C -0,23 0,18

Jämtland B C -0,01 0,19

Värmland A B 0,25 0,28

Surface temperature Abisko D 11,76 0,83

Jämtland C 13,99 0,69

Värmland B 16,05 1,01

Chl-a Abisko B 0,00 0,32

Västerbotten A B 0,21 0,21

Jämtland A B 0,18 0,15

(12)

11

Correlations of chlorophyll a and zooplankton abundances with lake DOC and DIN:TP

There was an overall negative correlation between DOC and DIN:TP in the 33 study lakes (r = -0.690, p < 0.001). Chlorophyll-a concentrations were positively correlated with the DOC concentrations ((p = 0.009, r = 0. 440), but not correlated with the DIN:TP ratios in the lakes (r = -0.197, p = 0.263).

On one hand, the absolute abundances of Diaphanasoma, Ceriodaphnia, and Holopedium were positively associated with lake DOC concentrations (Spearman’s rank r = 0.346-0.678, all p ≤ 0.045), but no significant correlation was found between the absolute abundances of other zooplankton taxa and DOC (Spearman’s rank r = -0.124-0.678, all p > 0.05).

Whereas on the other, a Spearman correlation was again done between DIN:TP and the zooplankton abundance. Most of the taxa showed no correlation at all with DIN:TP levels:

Bosmina, Calanoida, Cyclopoida, Daphnia, Holopedium and Polyphemus (Spearman correlation: all p ≥ 0.056). However, those taxa that showed to be correlated were:

Diaphanasoma, Ceriodaphnia and Limnosida (Spearman correlation: all p ≤ 0.028) with the following Spearman correlation coefficients: -0.566, -0.434, and 0.376 respectively.

Discussion:

My results showed that there were overall regional differences in zooplankton composition and environmental characteristics of the lakes, supporting hypothesis 1 and 2. The zooplankton composition differed between the northern and the southern regions, and greater differences were found between the mountain regions (i.e. Abisko vs Jämtland) than between the boreal regions (i.e. Västerbotten vs Värmland). Calanoida was the major taxon responsible for these north-south differences, despite that Bosmina, Holopedium, Daphnia and Cyclopoida were also important.

The regional patterns in zooplankton composition were similar to those in DIN and DIN:TP concentrations, where the Abisko lakes and Västerbotten lakes had lower concentrations compared to the Jämtland lakes and Värmland lakes. According to Bergström et al. 2020, seston has higher concentrations of EFA in lakes with higher DIN:TP (i.e. Jämtland), although phytoplankton biomass does not differ between Abisko lakes and Jämtland lakes. This indicates that the food for zooplankton would be of a better quality in Jämtland lakes as they have higher DIN:TP ratios. However, based on my results, calanoid abundance was lower in these lakes. This might be connected to the higher P demand of calanoids compared to the cladocerans. Indeed, my results corroborate those in Deininger et al. 2017, where they found no effects of N-fertilization on zooplankton abundance (especially copepods) in boreal lakes, likely because of the intensified P limitation in phytoplankton after the N-fertilization. Thus, the declines in lake DIN:TP ratio might have favored the zooplankton development.

Regional differences in zooplankton communities can be connected to those in phytoplankton communities. My results showed that phytoplankton biomass (i.e. chlorophyll a concentration) increased with DOC, likely through the positive effects of DOC on overall nutrient availability (Bergström & Karlsson 2019). This result supports hypothesis 3 because most lakes in this study had DOC concentrations lower than the threshold level (i.e. 10.6 mg/L; Bergström &

Karlsson 2019), above which the phytoplankton biomass would be limited by light instead of nutrient availability. The lake DOC and chlorophyll-a concentrations also differed across regions, being higher in boreal regions than in mountain regions. This was probably due to the greater forest cover in the boreal regions which has enhanced the export of terrestrial DOC into the lakes (Table 1).

(13)

12

There was no association between lake DIN:TP ratios and the biomass of phytoplankton, which does not support hypothesis 3. I found regional differences in lake DIN:TP (Table 2), suggesting that the lakes in these regions differed in nutrient limitation regimes for phytoplankton. This, in addition to our findings about DOC, could explain the regional patterns in chlorophyll-a concentrations and environmental characteristics of the lakes. The role of phytoplankton is crucial in the trophic food web of all lakes since they act as primary producers and therefore depending on their abundance or composition this will likely also have impacts on higher trophic levels (Deininger et al. 2017). My results showed that the Värmland lakes had higher chlorophyll-a concentrations than the Abisko lakes, which support that there were bigger differences in phytoplankton biomass between northern and southern regions (supporting hypothesis 4), and might underlie the regional differences in zooplankton abundances. These differences could be related to DIN and DIN:TP concentrations, as they are lower in northern regions but higher in southern regions.

My results suggest that an increase in DOC concentrations would promote higher abundances of the cladocerans Diaphanasoma, Ceriodaphnia, and Holopedium. There were also correlations between DIN:TP concentrations and zooplankton taxa, being negative for Diaphanasoma and Ceriodaphnia but positive for Limnosida. Thus, hypothesis 3 was almost entirely supported by my results. Zooplankton could consume considerable amounts of terrestrial DOC in oligotrophic systems where primary production is limited (Tanentzap et al.

2017). When the availability of terrestrial DOC increases, the less-selective filter feeders, such as cladocerans, particularly consume more of this terrestrial resource than do the other zooplankton taxa such as the copepods (Tanentzap et al. 2017). This terrestrial support could explain the higher abundances of the cladoceran taxa in lakes with higher DOC in this study.

Increasing lake DIN:TP might have enhanced the P-limitation for zooplankton, but this is less likely for cladocerans (Deininger et al. 2017). This potentially underlies the variable effects of increasing DIN:TP on different cladoceran taxa in my study.

Overall, I found that the zooplankton communities in Swedish lakes differed between northern and southern regions, and greater differences were found between the mountain regions than between the boreal regions. These differences were likely related to the lake environmental characteristics including DIN, DIN:TP and DOC, which had also affected the phytoplankton biomass (i.e. chlorophyll-a concentrations, which tended to be higher in southern regions than in northern regions). Lakes with lower DIN:TP ratios had higher abundances of calanoids, while lakes with higher DOC concentrations had higher abundances of certain cladoceran taxa.

Yet, calanoids were the major contributor of the dissimilarity in zooplankton composition among the regions. Therefore, the DIN:TP ratio possibly has stronger effects than DOC on zooplankton composition in Swedish oligotrophic lakes. But DOC could moderate these effects.

Increases in lake DOC concentrations would reduce the differences in zooplankton composition caused by the declines in lake DIN:TP ratios (i.e. smaller differences in zooplankton composition between boreal regions than between mountain regions).

Acknowledgment:

I would like to thank Danny Lau, Ann-Kristin Bergström and all the scientists involved during both lake sampling and lab work. Furthermore, I would like to specially thank Danny Lau because without his help this thesis would not have been possible, being his support fundamental throughout the entire development of the project.

(14)

13

References

:

Ask, J., Karlsson, J., Persson, L., Ask, P., Byström, P., & Jansson, M. (2009). Terrestrial organic matter and light penetration: Effects on bacterial and primary production in lakes. Limnology and Oceanography, 54(6), 2034-2040.

Bergström, A. K., Blomqvist, P., & Jansson, M. (2005). Effects of atmospheric nitrogen deposition on nutrient limitation and phytoplankton biomass in unproductive Swedish lakes. Limnology and Oceanography, 50(3), 987-994.

Bergström, A. K., Faithfull, C., Karlsson, D., & Karlsson, J. (2013). Nitrogen deposition and warming–effects on phytoplankton nutrient limitation in subarctic lakes. Global Change Biology, 19(8), 2557-2568.

Bergström, A. K. & Karlsson, J. (2019) Light and nutrient control phytoplankton biomass responses to global change in northern lakes. Global Change Biology Volume 25, Issue 6.

25:2021–2029.

Clarke, K. (1993). Nonparametric Multivariate Analyses of Changes in Community Structure.

Austral Ecology. 18. 117-143.

Deininger, A., Faithfull, C. L., Karlsson, J., Klaus, M. & Bergström, A. K. (2017) Pelagic food web response to whole lake N fertilization. Limnology and Oceanography Volume 62, Issue 4. 1498–1511.

de Wit H. A., Valinia, S., Weyhenmeyer, G. A., Futter, M. N., Kortelainen, P., Austnes, K., Hessen, D. O., Räike, A., Laudon, H. & Vuorenmaa, J.(2016) Current browning of surface waters will be further promoted by wetter climate. Environ Sci Technol Lett 3:430–435.

Finstad, A., Andersen, T., Larsen, S., Tominaga, K., Blumentrath, S., de Wit, H. A., Tømmervik, H. & Hessen, D. O. (2016) greening to browning: Catchment vegetation development and reduced S-deposition promote organic carbon load on decadal time scales in Nordic lakes. Sci Rep 6, 31944.

Huser, B. J., Futter, M. N., Wang, R., & Fölster, J. (2018). Persistent and widespread long-term phosphorus declines in Boreal lakes in Sweden. Science of the total environment, 613, 240-249.

Isles, P. D., Creed, I. F., & Bergström, A. K. (2018). Recent synchronous declines in DIN: TP in Swedish lakes. Global Biogeochemical Cycles, 32(2), 208-225.

Isles, P. D. F., Jonsson, A., Creed, I. F. & Bergström, A. K. (2020) Does browning affect the identity of limiting nutrients in lakes? Aquatic Sciences 82, 45.

Jansson, M., Persson, L., DeRoos, A., Jones, R. I., & Tranvik, L. J. (2007). Terrestrial carbon and intraspecific size‐variation shape lake ecosystems. Trends in Ecology and Evolution, 22, 316–322.

Jones, R. I. 1992. The inﬂuence of humic substances on lacustrine planktonic food-chains.

Hydrobiologia 229: 73–91.

Karlsson, J., Byström, P., Ask, J., Ask, P., Persson, L., & Jansson, M. (2009). Light limitation of nutrient-poor lake ecosystems. Nature, 460(7254), 506-509.

(15)

14

Larsen, S., Andersen, T. O. M., & Hessen, D. O. (2011). Climate change predicted to cause severe increase of organic carbon in lakes. Global Change Biology, 17(2), 1186-1192.

Monteith, D., Stoddard, J., Evans, C., de Wit, H., Forsius, M., Høgåsen, T., Wilander, A., Skjelkvåle, B. L., Jeffries, D., Vuorenmaa, J., Keller, B., Kopáček, J. & Vesely, J. (2007).

Dissolved organic carbon trends resulting from changes in atmospheric deposition chemistry. Nature. 450. 537-40.

Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D. & Stevens, M.

H. H. (2017). vegan: Community Ecology Package. R package version 2.4–2. 2017.

Persson, J., Brett, M. T., Vrede, T. & Ravet, J. L. (2007). Food qquantity and quality regulation of trophic transfer between primary producers and keystone grazer (Daphnia) in pelagic freshwater food webs. Oikos 116:1152-1163.

Persson, J. & Vrede, T. (2006) Polyunsaturated fatty acids in zooplankton: variation due to taxonomy and trophic position. Freshwater Biology. 51, 887–900.

R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/

Sterner, R. W., & Hessen, D. O. (1994). Algal nutrient limitation and the nutrition of aquatic herbivores. Annual Review of Ecology and Systematics, 25, 1–29.

Taipale S., Strandberg U., Peltomaa E., Galloway A. W. E., Ojala A. & Brett, M. T. (2013) Fatty acid composition as biomarkers of freshwater microalgae: analysis of 37 strains of microalgae in 22 genera and in seven classes. Aquat Microb Ecol 71(2):165–178.

Tanentzap, A. J., Kielstra, B. W., Wilkinson, G. M., Berggren, M., Craig, N., del Giorgio, P. A., Grey, J., Gunn, J. M., Jones, S. E., Karlsson, J., Solomon, C. T., Christopher, T. & Pace, M. L.(2017). Terrestrial support of lake food webs: Synthesis reveals controls over cross- ecosystem resource use. Science advances, 3(3), e1601765.

Zwart, J. A., Craig, N., Kelly, P. T., Sebestyen, S. D., Solomon, C. T., Weidel, B. C., & Jones, S.

E. (2016). Metabolic and physiochemical responses to a whole-lake experimental increase in dissolved organic carbon in a north-temperate lake. Limnology and Oceanography, 61(2), 723–734.