Department Systems Analysis, Integrated Assessment and Modelling

This paper presents some of the reasons for studying the history of hydrology and for the formation of the International Association of Hydrological Sciences (IAHS) History of Hydrology Working Group. In particular, we consider the importance of recording the histories of hydrological data, catchments, diversity in hydrology (of both people and topics), and what can be gained from the historical literature in hydrology. We also consider why the classical concepts of catchment response to rainfall have evolved slowly in hydrology, despite identified limitations, and how the major impetus for change in the 1970s, in the form of techniques for measuring environmental tracers, effectively came from developments outside the discipline. We conclude by speculating whether the questions posed by recent machine learning studies might lead to further change and understanding.

To what extent is biodiversity shaped by environmental conditions, and to what extent is it the result of self-organization? Both natural processes and organismal properties may contribute to promoting diversity. Here, we show that one such process, namely natural selection, and an organismal property, namely life history, interact in a feedback mechanism that promotes the emergence of diversity. We illustrate how this mechanism operates using various models of ecological diversification driven by intraspecific resource competition, in which both a niche trait that determines resource use and a life history trait can evolve. We find that natural selection acting on life history traits leads to increased competition, which, in the presence of ecological opportunity, facilitates niche diversification. As a consequence, the environmental conditions for diversification are more restrictive in the absence of life history evolution than in its presence. Our findings indicate a strong influence of life history evolution on ecological processes that in turn shape the origin of biodiversity. Our results call for a better integration of life history evolution and niche diversification in both theoretical and empirical realms.

The Indian Ocean plays a crucial role in water cycle dynamics around the Indian Ocean Basin, yet the relationships between the Indian Walker Circulation (IWC) and hydroclimate variability remain poorly understood. We examine the IWC-hydroclimate linkages throughout the Indian Ocean Basin during two study periods (instrumental era and the last millennium), using the ECHAM5-wiso, iCESM-iLME, and ERA5 global data products. We leverage the stable isotopic composition of precipitation ((Formula presented.) ¹⁸O_P) as an integrative water cycle tracer to elucidate the link between large-scale atmospheric processes and regional hydroclimate characteristics. We find a strong statistical relationship between the East African short rain (Formula presented.) ¹⁸O_P and IWC in Eastern Africa and Horn of Africa. In Indonesia, the relationship is weaker due to the strong influence of the Pacific Walker Circulation (PWC). These patterns arise over multiple time scales and are consistent with the strong tie between the IWC and precipitation variability during the short rain season in Eastern Africa and Horn of Africa that had previously been reported in the instrumental era but not yet for the last millennium. We document strong fluctuations in the strength of the IWC during the last millennium, with periods of stronger or weaker interactions with the PWC that influenced (Formula presented.) ¹⁸O_P variability. Finally, our study underscores the ability of (Formula presented.) ¹⁸O_P to capture major atmospheric circulation signals, demonstrating its potential to examine the impact of interacting Walker Circulations on regional hydroclimate even during periods of low precipitation.

Pro-environmental decisions, such as rejecting pesticide use in agriculture, may stem from both environmental and health concerns. Identifying which concerns are more decisive for pro-environmental decisions, and whether this varies between people, depending on their value orientations, could offer valuable insights into how to best promote pro-environmental decisions across different audiences. While biospheric values likely underlie environmental concerns, it is unclear which value orientation underlies health concerns. In a preregistered online experiment (N = 823), we explored whether egoistic or personal safety values—a subtype of personal security values developed for this study—underlie health concerns regarding pesticide use in agriculture. Participants reported on their opposition to the use of a fictitious fungicide in potato cultivation, based on information about its risks to human health (relevant for egoistic and personal safety values) and/or the environment (relevant for biospheric values). Stronger biospheric values were consistently associated with stronger opposition to the fungicide's use, regardless of the risk information. Egoistic values interacted with risk information, but these interactions contradicted our assumption that egoistic values reflect health concerns. Personal safety values showed no interaction with risk information and were not independently associated to opposition to the fungicide's use. Our findings suggest that neither egoistic nor personal safety values serve as the basis for health concerns driving pro-environmental decisions. This underscores the need to identify an additional value orientation that reflects health concerns and develop measures to assess it.

Neural likelihood estimation methods for simulation-based inference can suffer from performance degradation when the modeled data is very high-dimensional or lies along a lower-dimensional manifold, which is due to the inability of the density estimator to accurately estimate a density function. We present Surjective Sequential Neural Likelihood (SSNL) estimation, a novel member in the family of methods for simulation-based inference (SBI). SSNL fits a dimensionality-reducing surjective normalizing flow model and uses it as a surrogate likelihood function, which allows for computational inference via Markov chain Monte Carlo or variational Bayes methods. Among other benefits, SSNL avoids the requirement to manually craft summary statistics for inference of high-dimensional data sets, since the lower-dimensional representation is computed simultaneously with learning the likelihood and without additional computational overhead. We evaluate SSNL on a wide variety of experiments, including two challenging real-world examples from the astrophysics and neuroscience literatures, and show that it either outperforms or is on par with state-of-the-art methods, making it an excellent off-the-shelf estimator for SBI for high-dimensional data sets.

The interplay between abiotic (resource supply, temperature) and biotic (grazing) factors determines growth and loss processes in phytoplankton through resource competition and trophic interactions, which are mediated by morphological traits like size. Here, we study the relative importance of grazers, water physics, and chemistry on the daily net accumulation rates (ARs) of individual phytoplankton from natural communities, grouped into six size classes from circa 10 to 500 μm. Using a Random Forest modelling approach and 4 years of daily data from a lake, we find that water temperature is generally a pivotal control of all phytoplankton ARs. At the same time, nutrients and light are important for the smallest and the largest classes. Mesozooplankton abundance is a key predictor of the AR for small phytoplankton, with microzooplankton being important for the middle-size range. In our data, large and small phytoplankton have different (seasonal) blooming patterns: small forms are favoured by low temperature and grazing, and high phosphorus levels. Larger forms show positive ARs at high temperatures and low phosphorus (being relatively insensitive to zooplankton grazing). These results help us understand the opportunities and limitations of using size to explain and model phytoplankton responses to biotic and abiotic environmental change.