Lifelines for a drowning science ‐ improving findability and synthesis of hydrologic publications

Abstract

Increasing publication numbers make it difficult to keep up with knowledge evolution in a science like hydrology. Here we give recommendations to authors and journals for writing future-proof articles that contribute to knowledge accumulation and synthesis.

Catchment characterization: current descriptors, knowledge gaps and future opportunities

“…This ambiguity limits the synthesis of the results from different studies (McMillan, 2022). As suggested by Wagener et al (2021) and Stein et al (2022), providing a clear description and aggregation method for catchment descriptors preferably in a tabular form together with the metadata about the study catchments might help to consolidate valuable information available from local and regional studies that is currently lacking at the global scale. This might be especially important for subsurface catchment characterization where the resolution and accuracy of the global data sources does not allow us to capture the vast subsurface heterogeneity and complex vertical structure of the critical zone.…”

Machine Learning for Bayesian Experimental Design in the Subsurface

“…The unifying concept of these terms seems to be deriving optimal (under a specific criterion) locations for an experimental campaign or strategy for observing a phenomenon, be it spatial observations, space-based observations, or field experiments such as those aimed at data assimilation. The issue is that the terms are not used consistently, and the literature is scattered across a variety of disciplines, which make it difficult to find relevant information (see also Rahman et al 2022;Stein et al 2022). It is most likely due to the fact that it is a multidisciplinary field, with experiment goals frequently varying greatly from one problem to the next.…”

GroMoPo: A Groundwater Model Portal for Findable, Accessible, Interoperable, and Reusable (FAIR) Modeling

“…While data has become more FAIR, in our experience, the sharing of groundwater models is lagging behind in the academic community and rarely meets these standards (with some organizations, like the U.S. Geological Survey, the Geological Survey of Denmark and Greenland, and the consortium of the Dutch Hydrological Instrument as notable exceptions). We speculate that there are multiple interconnected reasons for this, including that: (1) many models are developed for site‐specific investigations or without a scientific research focus, for example, those developed for environmental assessment projects, and therefore are never published in the peer‐reviewed literature; (2) models may also contain proprietary intellectual property and/or private information (Zipper et al 2019); (3) models often require large input/output files that are challenging to archive; (4) some numerical models require proprietary software (Zipper et al 2022); (5) incentive structures in the academic system are not designed for common‐good activities such as model sharing (Verbeke 2023); and (6) those models that do get published are challenging to find due to the ever‐increasing rate of publication (Stein et al 2022). In sum, we estimate that globally between 330 and 540 journal articles describing groundwater models are published (findable) every year (Supporting information), but almost none of these models are fully FAIR.…”

“…(3) models often require large input/output files that are challenging to archive; (4) some numerical models require proprietary software (Zipper et al 2022); (5) incentive structures in the academic system are not designed for common-good activities such as model sharing (Verbeke 2023); and (6) those models that do get published are challenging to find due to the ever-increasing rate of publication (Stein et al 2022). In sum, we estimate that globally between 330 and 540 journal articles describing groundwater models are published (findable) every year (Supporting information), but almost none of these models are fully FAIR.…”