Community Workflows to Advance Reproducibility in Hydrologic Modeling: Separating model-agnostic and model-specific configuration steps in applications of large-domain hydrologic models
“…Recently, in fact, it has become clear that ML and deep learning (DL) techniques can be interpreted and explained (Gharari et al, 2021); thus, they can be used as a tool for understanding (the process) (Arrieta et al, 2020) or model parameter learning (Tsai et al, 2021), instead of primarily for predictive purposes. In any case, ML growth has been mostly driven by a large variety of problems, for instance, computer vision applications, and speech and natural language processing in a way that has to be harmonized with the practices of more traditional ways of conceiving hydrological models.…”
Section: From Models To Darthsmentioning
confidence: 99%
“…That is to say, promoting diversity should be accompanied by the sharing of standards for the parts. The way to do it has been traced, for instance, in Knoben et al (2021), who argued that the whole models informatics can be separated into model-agnostic parts, which potentially can be shared, and model-specific parts, which could be differentiated among the various developers or research groups.…”
Section: From Models To Darthsmentioning
confidence: 99%
“…At present, both in research and in applications, the action of matching data and models is done offline by the researchers, but the DARTH vision would require that these data be automatically ingested and processed. Therefore, DARTHs should take care of these aspects by design and be able to abstract data from the algorithms, as also explained in Knoben et al (2021).…”
Section: A Necessary First Step: Making Data and Formats Openmentioning
confidence: 99%
“…Traditionally engineered models tend to be "applications", meaning that they bundle together all of the features that are required to have an all-round modeling experience, but these are exactly the opposite of what is needed by DARTHs, where everything must be provided as a service and loosely tied. Knoben et al (2021) give a further clear description of MaaA. Most current models fall into this category.…”
Section: Four Plus One Steps Towards Darthsmentioning
Abstract. The “Digital Earth” (DE) metaphor is very useful for both end users and hydrological modelers (i.e., the coders). In this opinion paper, we analyze different categories of models with the view of making them part of Digital eARth Twin Hydrology systems (DARTHs). We stress the idea that DARTHs are not models, rather they are an appropriate infrastructure that hosts (certain types of) models and provides some basic services for connecting to input data. We also argue that a modeling-by-component strategy is the right one for accomplishing the requirements of the DE.
Five technological steps are envisioned to move from the current state of the art of modeling. In step 1, models are decomposed into interacting modules with, for instance, the agnostic parts dealing with inputs and outputs separated from the model-specific parts that contain the algorithms. In steps 2 to 4, the appropriate software layers are added to gain transparent model execution in the cloud, independently of the hardware and the operating system of computer, without human intervention. Finally, step 5 allows models to be selected as if they were interchangeable with others without giving deceptive answers. This step includes the use of hypothesis testing, the inclusion of error of estimates, the adoption of literate programming and guidelines to obtain informative clean code. The urgency for DARTHs to be open source is supported here in light of the open-science movement and its ideas. Therefore, it is argued that DARTHs must promote a new participatory way of performing hydrological science, in which researchers can contribute cooperatively to characterize and control model outcomes in various territories. Finally, three enabling technologies are also discussed in the context of DARTHs – Earth observations (EOs), high-performance computing (HPC) and machine learning (ML) – as well as how these technologies can be integrated in the overall system to both boost the research activity of scientists and generate knowledge.
“…Recently, in fact, it has become clear that ML and deep learning (DL) techniques can be interpreted and explained (Gharari et al, 2021); thus, they can be used as a tool for understanding (the process) (Arrieta et al, 2020) or model parameter learning (Tsai et al, 2021), instead of primarily for predictive purposes. In any case, ML growth has been mostly driven by a large variety of problems, for instance, computer vision applications, and speech and natural language processing in a way that has to be harmonized with the practices of more traditional ways of conceiving hydrological models.…”
Section: From Models To Darthsmentioning
confidence: 99%
“…That is to say, promoting diversity should be accompanied by the sharing of standards for the parts. The way to do it has been traced, for instance, in Knoben et al (2021), who argued that the whole models informatics can be separated into model-agnostic parts, which potentially can be shared, and model-specific parts, which could be differentiated among the various developers or research groups.…”
Section: From Models To Darthsmentioning
confidence: 99%
“…At present, both in research and in applications, the action of matching data and models is done offline by the researchers, but the DARTH vision would require that these data be automatically ingested and processed. Therefore, DARTHs should take care of these aspects by design and be able to abstract data from the algorithms, as also explained in Knoben et al (2021).…”
Section: A Necessary First Step: Making Data and Formats Openmentioning
confidence: 99%
“…Traditionally engineered models tend to be "applications", meaning that they bundle together all of the features that are required to have an all-round modeling experience, but these are exactly the opposite of what is needed by DARTHs, where everything must be provided as a service and loosely tied. Knoben et al (2021) give a further clear description of MaaA. Most current models fall into this category.…”
Section: Four Plus One Steps Towards Darthsmentioning
Abstract. The “Digital Earth” (DE) metaphor is very useful for both end users and hydrological modelers (i.e., the coders). In this opinion paper, we analyze different categories of models with the view of making them part of Digital eARth Twin Hydrology systems (DARTHs). We stress the idea that DARTHs are not models, rather they are an appropriate infrastructure that hosts (certain types of) models and provides some basic services for connecting to input data. We also argue that a modeling-by-component strategy is the right one for accomplishing the requirements of the DE.
Five technological steps are envisioned to move from the current state of the art of modeling. In step 1, models are decomposed into interacting modules with, for instance, the agnostic parts dealing with inputs and outputs separated from the model-specific parts that contain the algorithms. In steps 2 to 4, the appropriate software layers are added to gain transparent model execution in the cloud, independently of the hardware and the operating system of computer, without human intervention. Finally, step 5 allows models to be selected as if they were interchangeable with others without giving deceptive answers. This step includes the use of hypothesis testing, the inclusion of error of estimates, the adoption of literate programming and guidelines to obtain informative clean code. The urgency for DARTHs to be open source is supported here in light of the open-science movement and its ideas. Therefore, it is argued that DARTHs must promote a new participatory way of performing hydrological science, in which researchers can contribute cooperatively to characterize and control model outcomes in various territories. Finally, three enabling technologies are also discussed in the context of DARTHs – Earth observations (EOs), high-performance computing (HPC) and machine learning (ML) – as well as how these technologies can be integrated in the overall system to both boost the research activity of scientists and generate knowledge.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.