Refine
Document Type
- Article (7)
- Part of Periodical (2)
- Monograph/Edited Volume (1)
- Doctoral Thesis (1)
- Postprint (1)
- Review (1)
Keywords
- acute kidney injury (2)
- end-stage kidney disease (2)
- genome-wide association (2)
- polymictic lakes (2)
- rapid eGFRcrea decline (2)
- spectral analysis (2)
- study (2)
- variability (2)
- Lake number (1)
- Prozessidentifikation (1)
Institute
- Hasso-Plattner-Institut für Digital Engineering GmbH (3)
- Institut für Umweltwissenschaften und Geographie (3)
- Institut für Biochemie und Biologie (2)
- MenschenRechtsZentrum (2)
- Institut für Geowissenschaften (1)
- Institut für Informatik und Computational Science (1)
- Referat für Presse- und Öffentlichkeitsarbeit (1)
The central aim of this thesis is to demonstrate the benefits of innovative frequency-based methods to better explain the variability observed in lake ecosystems. Freshwater ecosystems may be the most threatened part of the hydrosphere. Lake ecosystems are particularly sensitive to changes in climate and land use because they integrate disturbances across their entire catchment. This makes understanding the dynamics of lake ecosystems an intriguing and important research priority. This thesis adds new findings to the baseline knowledge regarding variability in lake ecosystems. It provides a literature-based, data-driven and methodological framework for the investigation of variability and patterns in environmental parameters in the time frequency domain.
Observational data often show considerable variability in the environmental parameters of lake ecosystems. This variability is mostly driven by a plethora of periodic and stochastic processes inside and outside the ecosystems. These run in parallel and may operate at vastly different time scales, ranging from seconds to decades. In measured data, all of these signals are superimposed, and dominant processes may obscure the signals of other processes, particularly when analyzing mean values over long time scales. Dominant signals are often caused by phenomena at long time scales like seasonal cycles, and most of these are well understood in the limnological literature. The variability injected by biological, chemical and physical processes operating at smaller time scales is less well understood. However, variability affects the state and health of lake ecosystems at all time scales. Besides measuring time series at sufficiently high temporal resolution, the investigation of the full spectrum of variability requires innovative methods of analysis.
Analyzing observational data in the time frequency domain allows to identify variability at different time scales and facilitates their attribution to specific processes. The merit of this approach is subsequently demonstrated in three case studies. The first study uses a conceptual analysis to demonstrate the importance of time scales for the detection of ecosystem responses to climate change. These responses often occur during critical time windows in the year, may exhibit a time lag and can be driven by the exceedance of thresholds in their drivers. This can only be detected if the temporal resolution of the data is high enough. The second study applies Fast Fourier Transform spectral analysis to two decades of daily water temperature measurements to show how temporal and spatial scales of water temperature variability can serve as an indicator for mixing in a shallow, polymictic lake. The final study uses wavelet coherence as a diagnostic tool for limnology on a multivariate high-frequency data set recorded between the onset of ice cover and a cyanobacteria summer bloom in the year 2009 in a polymictic lake. Synchronicities among limnological and meteorological time series in narrow frequency bands were used to identify and disentangle prevailing limnological processes.
Beyond the novel empirical findings reported in the three case studies, this thesis aims to more generally be of interest to researchers dealing with now increasingly available time series data at high temporal resolution. A set of innovative methods to attribute patterns to processes, their drivers and constraints is provided to help make more efficient use of this kind of data.
Portal Transfer
(2023)
Liebe Leserinnen und Leser, kein Nachrichtentag vergeht, an dem nicht die Expertise aus der Wissenschaft gefragt ist: Ob zum russischen Angriffskrieg in der Ukraine, zur UNKlimakonferenz in Ägypten, zur Flutkatastrophe in Pakistan, zum Dürresommer, zur Energiekrise, selbst zur umstrittenen Fußballweltmeisterschaft in Katar standen und stehen Expertinnen und Experten in den Medien Rede und Antwort. Auch aus der Universität Potsdam. Wir haben sie gefragt, wie sie damit umgehen, wie es ihnen gelingt, aus der laufenden Forschung heraus aktuelle Probleme zu bewerten. Und was davon bleibt, wenn das öffentliche Interesse abebbt. Für die Potsdamer Politik- und Verwaltungswissenschaftlerin Sabine Kuhlmann besteht die Kunst darin, „außerhalb der Krise Ideen und Lösungsansätze zu verstetigen und sie tatsächlich in die Praxis umzusetzen“.
In unserem Alumni- und Transfermagazin berichten wir davon, was und wie die Universität Potsdam dazu beiträgt. Wir erzählen, wie Erfindungen zu Innovationen in der Wirtschaft werden und sich Start-ups auf den Weg machen, ihr Produkt selbst zu vermarkten. Das Spektrum reicht von Meeresfrüchten auf Pflanzenbasis bis zu einer App, mit der sich Frühformen der Demenz erkennen lassen. Neben neuen Technologien kommt es aber vor allem darauf an, das an der Universität erzeugte Wissen in die Praxis zu transferieren. Deshalb stellen wir ein Programm zur Bekämpfung von Hassrede in der Schule vor oder auch eine Klettertherapie zur Behandlung von Skoliose. Und wir zeigen, wie eine Studie zur sportlichen Leistungskraft von Kindern helfen kann, den Sportunterricht zu verbessern.
Den größten Teil des an der Universität produzierten Wissens tragen die Studierenden in die Welt, wenn sie nach ihrem Abschluss als Musiklehrerin in einer Schule arbeiten oder als Software-Ingenieur im eigenen Unternehmen, als Geologin nach Seltenen Erden schürfen, als Ökologe ausgelaugte Böden wieder fruchtbar machen oder als Politikerin ein Ministerium leiten. Sie alle kommen in diesem Magazin zu Wort. Oder in unserem neuen Podcast „Listen.UP“, in dem Studierende, Forschende und Alumni von ihren Transferprojekten erzählen. Von der Gründerin Ulrike Böttcher erfährt man dort zum Beispiel, wie sie mit Schnallen, Ösen und Knöpfen aus Bio- Materialien die Modeindustrie in diesem Bereich nachhaltig verändern will. Nachzulesen ist das auch in diesem Heft. Immer dort, wo das „Listen. UP“-Logo erscheint, lohnt es, zusätzlich in den Podcast hineinzuhören.
Many lakes exhibit seasonal stratification, during which they develop strong thermal and chemical gradients. An expansion of depth-integrated monitoring programs has provided insight into the importance of organic carbon processing that occurs below the upper mixed layer. However, the chemical and physical drivers of metabolism and metabolic coupling remain unresolved, especially in the metalimnion. In this depth zone, sharp gradients in key resources such as light and temperature co-occur with dynamic physical conditions that influence metabolic processes directly and simultaneously hamper the accurate tracing of biological activity. We evaluated the drivers of metalimnetic metabolism and its associated uncertainty across 10 stratified lakes in Europe and North America. We hypothesized that the metalimnion would contribute highly to whole-lake functioning in clear oligotrophic lakes, and that metabolic rates would be highly variable in unstable polymictic lakes. Depth-integrated rates of gross primary production (GPP) and ecosystem respiration (ER) were modelled from diel dissolved oxygen curves using a Bayesian approach. Metabolic estimates were more uncertain below the epilimnion, but uncertainty was not consistently related to lake morphology or mixing regime. Metalimnetic rates exhibited high day-to-day variability in all trophic states, with the metalimnetic contribution to daily whole-lake GPP and ER ranging from 0% to 87% and < 1% to 92%, respectively. Nonetheless, the metalimnion of low-nutrient lakes contributed strongly to whole-lake metabolism on average, driven by a collinear combination of highlight, low surface-water phosphorous concentration and high metalimnetic volume. Consequently, a single-sensor approach does not necessarily reflect whole-ecosystem carbon dynamics in stratified lakes.