Changes in NPK uptake of Calluna vulgaris as a
consequence of climate change
Mick den Ouden 12303615 8 april 2021, Amsterdam
Supervised by:
Abstract
Because of rising average global temperatures and summer drought, many terrestrial ecosystems, not used to such extreme weather conditions, are endangered. The Dutch heathland ecosystem is one of such. Typical plants like Calluna vulgaris are already losing numbers due to high nitrogen deposition and now face another threat; climate change. This research aimed to discover influences of climate change on NPK dynamics of C. vulgaris. Two independent variables, elevated minimum temperature and summer drought, were used to mimic climate change. Three plots per treatment (drought, warming, control) were marked on a heathland in Oldebroek and studied for 5 consecutive years. We found that neither of the treatments significantly impacted shoot biomass (p-values: drought-control 0.97, warming-control 0.23, drought-warming 0.33) or NPK uptake (p-values: D-C 0.41, W-C 0.98, D-W 0.53). Total NPK concentrations in the shoots of the drought treatment were significantly lower than the concentrations in the shoots of the control group (p-value: D-C NPK 0.006). Of the three nutrients only K concentration was not significantly lower in the D-C comparison (p-values: N 0.006, P 0.003, K 0.09). Based on these findings there has to be concluded that both summer drought and nighttime warming does negatively influence NPK uptake of C. vulgaris shoots but over a 5 year time span this change is not significant.
================================================================================ Keywords: Nitrogen, Phosphorus, Potassium, Nutrient uptake, Heathland, the Netherlands, Calluna vulgaris.
Abstract………. 1
Introduction……….. 3
Methods and Data………... 4
Research Design……… 4 The Experiment……….. 4 Warming Treatment………. 4 Drought Treatment……….. 4 Control Group………... 5 Data Collection……… 5 Calculations………. 5 Statistical Analysis………..5 Results………... 7 Treatment results……… 7 Plant results………. 8 Discussion……….. 10 Drought……….. 10 Warming……….11
Flaws and follow-up research……….11
Conclusion………. 12
Literature list……….. 13
Acknowledgements………. 14
Appendices……… 15
Appendix I. Calluna plant category division………. 15
Introduction
Average global temperatures are rising as a result of increasing anthropogenic greenhouse gas levels in the atmosphere. In 2015, The Netherlands, together with 195 other parties, signed the Paris agreement (United Nations, 2015). This states that all its participants aim to limit global warming to a maximum of 2 °C compared to preindustrial levels (United Nations, 2015). Increased temperature causes heat waves and droughts during the summer to happen more frequently and more intensely (D’amato & Akdis, 2020). Higher temperatures are likely to extend the growth season (Beier, 2004) and boost net mineralization (Rustad et al., 2001), resulting in increased nutrient availability and plant nutrient absorption. Drought, on the other hand, can cause nutrient deficiencies in plants due to reduced mobility and absorbance of nutrients, resulting in a slower rate of mineral diffusion from the soil to the root systems (Bista et al., 2018).
In 1998 the CLIMOOR (climate driven changes in the functioning of heath and moorland ecosystems) research project started studying the influence of warming and drought on heath and moorland ecosystems in five different countries, including the Netherlands (Beier et al., 2004). From 1998-2002 the study site in Oldebroek, the Netherlands, collected data on plant biomass and chemical composition of the plants. The dominant plant species on heathlands like this is Calluna vulgaris.
Like all plants, C. vulgaris needs nitrogen (N), phosphorus (P) and potassium (K) as three of its most important nutrients. N is such an important nutrient for plant growth due to its major presence in chlorophyll, the compound that allows plants to conduct photosynthesis (Leghari et al., 2016). N is also a big constructor of amino acids, which are the building blocks of proteins. Proteins are extremely important in metabolic processes and multiplication. This is why plants wither and die when they are deprived of them (Leghari et al., 2016).
P and K are often limiting factors for plant growth due to their relatively low availability (Agren et al., 2012; Prajapati & Modi, 2012). Every living plant cell contains P and it plays a role in several key functions such as energy transfer, nutrient movement and photosynthesis. K is essential for the activation of more than 80 enzymes in plants. It is necessary for the ability to survive cold, heat, drought and pests. K also boosts water efficiency and converts carbohydrates to starch. P and K also come from different sources than N, namely weathering (Manning, 2010).
While the influence of elevated temperatures and drought on C. vulgaris root systems has been studied (Arndal et al., 2013), research on the nutrient dynamics within aboveground biomass of this particular heathland species is lacking. Knowledge on chemical compositions of these plants could help preserve the biodiversity of these landscapes. This is especially important because Dutch heathlands are currently also under pressure due to excessive N deposition.
In the Netherlands a surplus of N has developed caused by traffic, industries and livestock. Nitrogen oxides (NOx) are emitted by traffic and industries while livestock produces ammonia
(NH3) through the evaporation of livestock manure, used as fertilizer. NOxand NH3 can both
enter the soil through precipitation (wet deposition) or passive uptake by plants (dry deposition) (RIVM, 2019). These elevated N depositions are threatening heathland ecosystems (Jones & Power, 2011). Like many terrestrial ecosystems, heathlands are used to conditions of low N availability (Bobbink et al., 2010). Its biodiversity is decreasing due to an increased availability of N. Heathland plants like C. vulgaris can be expelled by grasses which thrive under the richer N conditions (Dueck et al., 1991).
This research is based on the findings of the CLIMOOR study. While most results are new, some findings from Beier et al (2004) will also be provided when necessary for the understanding of our results. The final aim is to discover influences of climate change on NPK dynamics of the heathland plant Calluna vulgaris. In this study, climate change is studied as two independent factors; increased minimum (nighttime) temperature and summer drought. It is hypothesized that increased temperature enhances the uptake of nutrients while drought impedes nutrient uptake.
Methods and Data Research Design
The effect of climate change on aboveground uptake of NPK was studied using a fieldscale manipulation experiment. This means that multiple groups were compared based on measurements over time. Plant chemical composition (response variable) was tested for three different climatological scenarios (explanatory variable). Scenario 1 was a warming treatment, scenario 2 had summer drought and scenario 3 was under ambient conditions and served as a control group.
The experiment took place on a heathland in Oldebroek, The Netherlands. This heather is dominated by Calluna vulgaris. Yearly measurements in August, when plant biomass is at its highest, provided data on plant shoot biomass and chemical composition of C. vulgaris which were used together with data on aboveground litter production to calculate total shoot NPK uptake for each scenario. Shoots are the parts of the plants that play the most active role in photosynthesis, therefore shoots were studied rather than stems, flowers or roots. The calculations were then provided with a statistical analysis to see whether the effects of climate change have a significant impact on C. vulgaris NPK uptake.
The Experiment Warming Treatment
Three plots of 20m2(5m by 4m) were marked out with galvanized steel scaffoldings, which also
prevented trampling the plants in the plots. The steel tubes were wrapped in polyethylene plastic in order to prevent pollutants from leaching in the soil. In order to increase the minimum average temperature in this warming treatment, the scaffolding was equipped with an infrared reflecting curtain. This curtain reflected almost all direct and diffuse radiation but allowed water vapor to get through. The curtain was open during the day and automatically closed when light intensity got below 0.4 W/m2. Since the hydrological conditions were to remain unchanged, a
rain detector allowed the curtain to open when there was rainfall during the night. Lastly, when wind speeds exceeded 10 m/s at night the curtain also opened in order to prevent damage (Beier et al., 2004).
The curtains opened and closed in sequential order, resulting in a 4 minute delay between the first and last curtain in the series. The curtains were placed about 0.6 to 1.0 m above the soil surface depending on the height of the vegetation. The research plots were open on all sides. Drought Treatment
For the drought treatment there were three 20m2 plots as well. The drought plots had the same
design as the warming plots, only for this scenario the curtains were made out of transparent polyethylene, allowing solar and surface radiation to pass through. The plots were equipped with a rain sensor which automatically closed the curtain when it started raining. The
accumulated water from the curtain was removed from the plot with a gutter. Again, when wind speeds exceeded 10 m/s the curtain was opened to prevent it from getting damaged. The drought treatment was done over a timespan of about 60 to 70 days during the growing season, May, June and July, each year. When the drought treatment period ended, the plots were treated the same as the control plots.
Control Group
Three untreated 20m2 control plots were run parallel to the other treatments for comparison.
These plots were also provided with a scaffolding same as the warming and drought treatment, but they did not have a curtain.
Data Collection
Data was collected for five consecutive years, starting in 1998. The 1998 data was pre-treatment but the uptake in 1998/99 was treated. Total aboveground biomass was weighed at the end of every growing season and expressed in gram/m2. Annual litter production was
determined with 6 collectors (100cm2) per plot, sampled every month. Plant and litter samples
were taken to a lab in order to establish the concentrations of N, P and K. The experimental setup was not designed for this research in particular. This caused raw data to be lacking for some years or plots. Educated guesses, based on average values of other years, were made where possible whenever this was the case.
Calculations
The data on the NPK concentrations within the plant aboveground biomass and litter as well as the total amount of aboveground biomass and litter was used to calculate the total uptake of the NPK using the following equation:
Equation 1: Uptake 1998/99 = (B 1999 * [NPK] 1999) - (B 1998 * [NPK] 1998) + (LP 1998/99 * [NPK] L 1998/99) [g NPK/m2/year] B = biomass [NPK] = concentration NPK LP = litter production
[NPK] L = concentration NPK in the litter
The calculation will be repeated for all studied years, starting from 1998/99.
Data on total plant biomass can be compared to the total uptake of nutrients to see if different chemical compositions due to warming or drought have any effect on the growth of the plants. Besides this we will also look at the average temperatures of the warming treatment compared to the control group and the moisture content of the drought treatment compared to the control group to see if the experimental design has actually given the desired effect.
Statistical Analysis
In order to determine whether the treatments differed significantly or not, statistical analysis was executed in the integrated development environment Rstudio. A one way analysis of variance (ANOVA) compared the three groups, warming treatment against control, drought treatment against control and warming treatment against drought treatment. Requirements for
this ANOVA were normally distributed and homogenous data. A post-hoc Tukey test showed exactly which groups differed.
Results
Treatment results
If we are to interpret the results it is important to know to what degree the treatments changed the conditions in which the plants grew. Figure 1 shows the mean annual temperature at soil surface for the control and warming treatment. During the night, when the curtain was closed on the warming treatment plots, mean temperatures rose to a maximum of almost 1.0 °C above that of the control group. When the curtain opened, surface temperatures started to congregate resulting in equal temperatures during most of the day. Changes in temperature were strongly influenced by ambient temperatures. At lower ambient temperatures the curtain tended to conserve the soils heat better resulting in above average warming. The opposite was true at higher ambient temperature, resulting in below average warming compared to the control group (Beier et al., 2004).
Figure 1: Mean annual temperature at soil surface for the warming treatment and control group.
The drought treatment showed even larger differences when comparing it to the control group than the warming treatment (figure 2). On an annual basis there was a blocking of about 20% of the rainwater, going into effect as soon as the drought period started. The soil moisture content dropped rapidly during this period, resulting in less available water at the end of the drought period. Though the annual water budgets for the warming and drought treatments were similar, the warming treatment removed only a small fraction of the water throughout the year while the drought treatment took this amount in a period of 2,5 month (Beier et al., 2004).
Plant results
Now that the results of the treatments are known, we can look at how this influenced C. vulgaris growth and nutrient uptake. Figure 3 shows the mean biomass of the shoots per treatment. Control group biomass increased with 129 g/m2 from 1999 till 2002. Biomass of the
warming treatment increased even more during this period of time (224g/m2) while the drought
treatment only increased with 57 g/m2. However, none of these increases were significantly
different (p-values: D-C 0.97, W-C 0.23, D-W 0.34). In 1999 most plots were affected by a heather beetle plague causing shoot biomass to reduce compared to 1998.
Figure 3: Calluna vulgaris shoot biomass per treatment.
The mean total uptake of NPK in C. vulgaris shoots follows a rather constant pattern for most of the years (figure 4). Regardless of the amount of NPK that is taken up, the drought treatment seems to have the least uptake of the three treatments for all years except 2001-2002. With 5330 g NPK/m2/year, the warming treatment has the highest uptake in 1998-1999. This is
115% of the uptake of the control group during that same period of time. However, this percentage dropped over the years and in 2001-2002 the plants on the warming treatment only took up 62% compared to the control group. Nevertheless, the total uptake of the treatments did not differ significantly from the control group or from each other (p-values: D-C 0.41, W-C 0.98, D-W 0.53). A year by year analysis may have been insightful but unfortunately there were too few plots per treatment to conduct this kind of statistical analysis.
The amount of NPK that is taken up (figure 4) does follow the growth in biomass logically (figure 3). In the years 1998-1999 and 2000-2001 the uptake was high and biomass increased, while 1999-2000 and 2001-2002 had lower uptake and smaller increase (even decreased in 1999) of biomass.
Figure 4: Total combined NPK uptake per treatment.
When looking at the division between the three elements (figure 5), N uptake is clearly the highest of the three for all years and all treatments. The origin of the negative K uptake for some years and treatments could be found in nutrient allocation but there has been no previous report of K allocation. Negative findings could also be because of minor measurement mistakes.
Figure 5: Mean division of NPK uptake per treatment.
Although NPK uptake does not show any significant differences between treatments, the concentrations of NPK in the C. vulgaris shoots do. NPK concentrations stayed rather stable throughout the research (figure 6) but N and P concentrations in the drought treatment were significantly lower than those of the control group (p-values: N 0.006, P 0.003) also resulting in an significantly lower total NPK concentration in the D-C comparison (p-value: 0.006).
Figure 6: Mean concentrations of NPK within the C. vulgaris shoots per treatment.
Discussion
Both treatments changed the conditions under which the C. vulgaris grew. The warming treatment raised nighttime temperatures with 1 °C and the drought treatment annually deprived the plots of 20% of the rainwater. However, neither C. vulgaris shoot biomass nor shoot NPK uptake was significantly impacted by the treatments over a period of 5 year. Shoot N and P concentrations were significantly lower in the drought treatment compared to the control group while K did not show significant change.
Drought
Drought conditions are commonly not beneficial for plant nutrient uptake (Da Silva et al., 2014) and this is no different for the C. vulgaris. As expected, the drought treatment had a lower nutrient uptake in all studied years but one. This outlier (2001-2002) caused the drought treatment nutrient uptake to be insignificantly different from the control group while the graph (figure 4) clearly indicates lower uptake for all other years. A lower uptake in the drought treatment can easily be explained by impaired nutrient uptake due to lower moisture content in the soil (Da Silva et al., 2014).
Negative influences of drought stress on N and P concentrations in a range of plants are readily confirmed by previous studies. He and Dijkstra (2014) did a meta-analysis on 155 observations and found on average 3.73% and 9.18% lower concentrations of respectively N and P. Our studied drought effects on C. vulgaris were slightly stronger with an average of 9.44% lower concentration of NPK in drought treatment compared to the control group. Arndal et al (2013) also found decreased (but not significant) uptake of N in the root systems of C. vulgaris. A drought treatment as well as a combined drought and warming treatment both showed lower uptake of N in the roots. This is in line with the lower (but not significant) uptake in the C. vulgaris shoots.
Warming
In contrast to the expectations, the warming treatment also seemed to be slightly affected by a smaller water balance. Higher temperature means more evaporation of water, causing the NPK uptake of the plants on the warming plots to get proportionally smaller compared to the control group over time. Extended research would provide more insight on the existence of this pattern, for the duration of this research overall uptake was again not significant.
Shoot biomass on the warming plots showed the largest increase of all treatments even when NPK uptake became lower than that of the control group. This seems very unlikely and no other studies found similar patterns. Andresen et al (2010) also found increased biomass on warming plots, however this was accompanied by increased N and P levels.
Flaws and follow-up research
The most noteworthy flaw of this study lies in the duration of the research. Since we had to work with data that was already collected years ago, the data was only provided for the years 1998 till 2002, making it so that we only had the required data to calculate the uptake of NPK for the years 1999 till 2002. Discovering patterns in uptake due to certain environmental changes over such short timespan is near impossible. In addition a large section of the plots was tormented by a heather beetle infestation in 1999 and because this pest did not affect all plots equally, the data on plant biomass for this year is uncertain.
Also, even for the calculated years not all data was provided forcing us to make assumptions based on mean values of the provided data. Assumptions were made on two occasions. The first assumption was made for the NPK concentrations of the litter. N litter concentration was only given for the years 1999 till 2001 and P and K concentrations were only given for the years 1999 and 2000. Mean values of these concentrations were used for the other years in order to calculate the yearly nutrient uptake. The second assumption concerns the division of plant biomass into shoots, stems and flowers. Since the data lacked plant biomass NPK concentrations of the stems and flowers for some years, we chose to calculate NPK uptake within the shoots of the plants. However, plant biomass was measured as total aboveground biomass. Therefore plant biomass had to be divided in plausible percentages of shoots, stems and flowers (appendix I).
Follow-up research with an extended study period and additional plots per treatment is highly recommended. Significance through statistical analysis is determined over the complete period of the research (1998-2002). This causes years with positive correlations to even out years with negative correlations, resulting in insignificant results. Year by year statistical analysis would have been extremely insightful in this study. However, this was not possible due to the relatively low amount of plots per treatment (3 plots per treatment, 9 plots per year). Most statistical tests need a sample size of at least 30 to provide valid and meaningful results.
Conclusion
Results of the 5 year field experiment showed that the uptake of NPK in the drought treatment was lower than that of both the control group and the warming treatment. This was in line with the hypothesis that drought would cause impaired uptake of nutrients, and confirmed by other published studies. N and P concentrations in the C. vulgaris shoots of the drought treatment were significantly lower than those of the control group. Warming treatment nutrient uptake also seemed to be decreasing over time although statistical analysis found almost no difference over the entire studied period. The decreasing uptake in the warming treatment goes against the hypothesis that an increased minimum temperature enhances nutrient uptake. All results considered, we have to conclude that both drought and warming negatively influence the uptake of NPK in C. vulgaris shoots. However, C. vulgaris shoot biomass increased in the warming treatment and significant influences of both drought and warming were not confirmed by statistical analysis. Some findings remain unexplained, therefore follow-up research with an extended study period and additional plots per treatment is highly recommended.
Literature list
Ågren, G. I., Wetterstedt, J. Å. M., & Billberger, M. F. K. (2012). Nutrient limitation on terrestrial plant growth - modeling the interaction between nitrogen and phosphorus. New Phytologist, 194(4), 953–960. https://doi.org/10.1111/j.1469-8137.2012.04116.x
Andresen, L. C., Michelsen, A., Jonasson, S., Schmidt, I. K., Mikkelsen, T. N., Ambus, P., & Beier, C. (2010). Plant nutrient mobilization in temperate heathland responds to elevated CO 2, temperature and drought. Plant and Soil, 328(1), 381-396.
Arndal, M. F., Merrild, M. P., Michelsen, A., Schmidt, I. K., Mikkelsen, T. N., & Beier, C. (2013). Net root growth and nutrient acquisition in response to predicted climate change in two contrasting heathland species. Plant and Soil, 369(1), 615-629.
Arndal, M. F., Schmidt, I. K., Kongstad, J., Beier, C., & Michelsen, A. (2014). Root growth and N dynamics in response to multi-year experimental warming, summer drought and elevated CO2 in a mixed heathland-grass ecosystem. Functional Plant Biology, 41(1), 1-10.
Beier, C. (2004). Climate Change and Ecosystem Function: Full-Scale Manipulations of CO 2 and Temperature. New Phytologist, 243-245.
Beier, C., Emmett, B., Gundersen, P., Tietema, A., Penuelas, J., Estiarte, M., ... & Williams, D. (2004). Novel approaches to study climate change effects on terrestrial ecosystems in the field: drought and passive nighttime warming. Ecosystems, 7(6), 583-597.
Bista, D., Heckathorn, S., Jayawardena, D., Mishra, S., & Boldt, J. (2018). Effects of Drought on Nutrient Uptake and the Levels of Nutrient-Uptake Proteins in Roots of Drought-Sensitive and -Tolerant Grasses. Plants, 7(2), 28. https://doi.org/10.3390/plants7020028
Bobbink, R., Hicks, K., Galloway, J., Spranger, T., Alkemade, R., Ashmore, M., . . . De Vries, W. (2010). Global assessment of nitrogen deposition effects on terrestrial plant diversity: a synthesis. Ecological Applications, 20(1), 30–59. https://doi.org/10.1890/08-1140.1
D’Amato, G., & Akdis, C. A. (2020). Global warming, climate change, air pollution and allergies. Allergy, 75(9), 2158–2160. https://doi.org/10.1111/all.14527
da Silva, E. C., Nogueira, R. J. M. C., da Silva, M. A., & de Albuquerque, M. B. (2011). Drought stress and plant nutrition. Plant stress, 5(Special Issue 1), 32-41.
Dueck, T. A., Van Der Eerden, L. J., Beemsterboer, B., & Elderson, J. (1991). Nitrogen uptake and allocation byCalluna vulgaris(L.) Hull andDeschampsia flexuosa(L.) Trin. exposed to15NH3. Acta Botanica Neerlandica, 40(4), 257–267. https://doi.org/10.1111/j.1438-8677.1991.tb01557.x
He, M., & Dijkstra, F. A. (2014). Drought effect on plant nitrogen and phosphorus: a meta-analysis. New Phytologist, 204(4), 924-931.
Jones, A. G., & Power, S. A. (2011). Field-scale evaluation of effects of nitrogen deposition on the functioning of heathland ecosystems. Journal of Ecology, 100(2), 331–342.
https://doi.org/10.1111/j.1365-2745.2011.01911.x
Leghari, S. J., Wahocho, N. A., Laghari, G. M., Laghari, A. H., Bhabhan, G. M., & Talpur, K. H. (2016). Role of nitrogen for plant growth and development: A review. American-Eurasian Network for Scientific Information, 10(9). Geraadpleegd van
https://go.gale.com/ps/i.do?v=2.1&it=r&sw=w&id=GALE%7CA472372583&prodId=AONE&si d=googleScholarFullText&userGroupName=amst
Manning, D. A. C. (2010). Mineral sources of potassium for plant nutrition. A review. Agronomy for Sustainable Development, 30(2), 281–294. https://doi.org/10.1051/agro/2009023
Prajapati, K., & Modi, H. A. (2012). THE IMPORTANCE OF POTASSIUM IN PLANT GROWTH – a REVIEW. Indian Journal of Plant Sciences, 1. Published.
Rijksinstituut voor Volksgezondheid en Milieu. (2019). Stikstof | RIVM. Geraadpleegd op 21 maart 2021, van https://www.rivm.nl/stikstof
Rustad, L. E. J. L., Campbell, J., Marion, G., Norby, R., Mitchell, M., Hartley, A., ... & Gurevitch, J. (2001). A meta-analysis of the response of soil respiration, net nitrogen mineralization, and aboveground plant growth to experimental ecosystem warming. Oecologia, 126(4), 543-562. Schulten, H.-R., & Schnitzer, M. (1997). The chemistry of soil organic nitrogen: a review. Biology and Fertility of Soils, 26(1), 1–15. https://doi.org/10.1007/s003740050335
The United Nations. (2015). The Paris Agreement. Geraadpleegd op 22 maart 2021, van
https://unfccc.int/process-and-meetings/the-paris-agreement/the-paris-agreement
Acknowledgements
This research was made possible by the efforts of Albert Tietema on collecting data since 1998 from the CLIMOOR experimental setup in Oldebroek. Therefore, a special thanks to Albert Tietema for providing the essential data and supervising this study. Anne Uilhoorn is also thanked for her supervision and help with the statistical analysis of the data.
Appendices
Appendix I. Calluna plant category division
Percentage (%) of Calluna biomass, constant in all plots, years and treatments
current year Calluna shoots
22
current year Calluna stems
18
current year Calluna flowers
11
older Calluna shoots
9
older Calluna stems
32
dead Calluna
8
100
Appendix II. Script for statistical analysis #install packages for statistical analysis #not all packages were used
install.packages("car") install.packages("tidyverse") install.packages("rstatix") install.packages("ggpubr") install.packages("outliers") #load the dataset in Rstudio library(readxl)
thesis_dataset <- read_excel("Thesis_Data_R.xlsx", col_types = c("text", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric"))
View(thesis_dataset)
#removing NA's from the dataset dataNAomit <- na.omit(thesis_dataset) View(dataNAomit)
#check the homoscedascity and normality of the data preliminary to ANOVA library(car)
leveneTest(y = dataNAomit$Total_Uptake, group = dataNAomit$Treatment) shapiro.test(thesis_dataset$Total_Uptake)
#one way analysis of variance total uptake of NPK
aovTOT <- aov(Total_Uptake ~ Treatment, data = dataNAomit) summary(aovTOT)
#one way analysis of variance nitrogen uptake
aovN <- aov(Uptake_N ~ Treatment, data = dataNAomit) summary(aovN)
#one way analysis of variance phosphorus uptake aovP <- aov(Uptake_P ~ Treatment, data = dataNAomit) summary(aovP)
#one way analysis of variance potassium uptake
aovK <- aov(Uptake_K ~ Treatment, data = dataNAomit) summary(aovK)
#one way analysis of variance shoot biomass
aovBiomass <- aov(Biomass ~ Treatment, data = thesis_dataset) summary(aovBiomass)
#one way analysis of variance NPK concentrations
aovConcN <- aov(Nitrogen ~ Treatment, data = thesis_dataset) aovConcP <- aov(Phosphorus ~ Treatment, data = thesis_dataset) aovConcK <- aov(Potassium ~ Treatment, data = thesis_dataset)
aovConctot <- aov(NPK ~ Treatment, data = thesis_dataset) summary(aovConctot)
#post hoc tests TukeyHSD(aovTOT) TukeyHSD(aovN) TukeyHSD(aovP) TukeyHSD(aovK) TukeyHSD(aovBiomass) TukeyHSD(aovConcN) TukeyHSD(aovConcP) TukeyHSD(aovConcK) TukeyHSD(aovConctot)
