Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/csdms/sedhyd-2019

Resources for the SEDHYD 2019 conference
https://github.com/csdms/sedhyd-2019

Last synced: 4 days ago
JSON representation

Resources for the SEDHYD 2019 conference

Awesome Lists containing this project

README 

        
# sedhyd-2019 💧



Resources for the SEDHYD 2019 conference (24-28 June 2019).



## Links 🔗



* [Community Surface Dynamics Modeling System

  (CSDMS)](http://csdms.colorado.edu)

* [Basic Model Interface (BMI)](http://bmi.readthedocs.io)

* [Python Modeling Toolkit (pymt)](http://pymt.readthedocs.io)

* [SEDHYD 2019](http://www.sedhyd.org/2019)



## Get Started 🚀



Install *pymt* using the *conda* package manager,



    $ conda env create --file=environment.yml



# Extended Abstract 📃



**Exploring Surface Processes Using the Community Surface Dynamics

Modeling System Modeling Tools**



**Jordan Adams**, Postdoctoral Research Associate, The Institute of

Arctic and Alpine Research, University of Colorado, Boulder, CO,

jordan.adams\@colorado.edu **Irina Overeem**, Associate Professor,

Department of Geological Sciences, University of Colorado, Boulder, CO,

irina.overeem\@colorado.edu **Eric Hutton**, Chief Software Engineer,

Community Surface Dynamics Modeling System, University of Colorado,

Boulder, CO, eric.hutton\@colorado.edu **Gregory Tucker**, Professor,

Department of Geological Sciences, University of Colorado, Boulder, CO,

gtucker\@colorado.edu **Albert Kettner**, Research Scientist II, The

Institute of Arctic and Alpine Research, University of Colorado,

Boulder, CO, albert.kettner\@gmail.com



**Extended Abstract**



Hydrological and sediment transport processes operate over a range of

temporal and spatial scales. Flood events can cause catastrophic erosion

rates over short timescales, reshaping floodplains and catchments in

hours or over days. Over longer timeframes, from decades to millennia,

the cumulative effect of these erosional events sculpt watershed

morphologies, driving changes in drainage area, density and relief. The

feedbacks between hydrologic events and sediment transport can shape

areas as small as millimeter-scale hillslope rills or as large as

continental-scale river basins.



The unpredictable or unobservable nature of flood events makes it

difficult to study the interaction between hydrologic and

sedimentological proccesses. The stochastic nature of floods make it

challenging to predict when an event will occur and how to best measure

erosional and depositional changes. Morphological responses to climate

change occur on timescales too long to make meaningful observations.

Moreover, at many relevant natural hazard scales, hydrology and sediment

transport are entangled with ecosystem and human dynamics, complex

interactions that are even less understood.



Improved understanding, and ultimately, improved resiliency in the face

of hydrologically- driven Earth surface change, require computational

models that bridge boundaries and link process mechanics. Many

hydrological models exist and have a variety of uses: forecasting floods

or water resource availability (e.g. WRF-Hydro, Gochis et al., 2018);

channel and floodplain engineering (e.g. HEC-RAS, Brunner, 2016) or

groundwater resource assessment (SWAT, USDA ARS Grassland Soil and Water

Research Laboratory, 2018). As computational resources become more

efficient, more researchers are adding numerical modeling skills to

their repertoire. Yet, as more models are built, questions remain: can

the surface processes community work together to share these

ever-improving tools? Is there a way to standardize both existing and

new modeling components so that they can be coupled flexibly and

effectively?



The Community Surface Dynamics Modeling System (CSDMS) is a NSF-funded

initiative that supports the open software efforts of the surface

processes community. CSDMS sets modeling standards and protocols, hosts

a Model Repository to distribute models and modeling tools, and provides

cyberinfrastructure to an interdisciplinary set of community members.

The CSDMS Repository contains over 200 tools and models that simulate

lithosphere, hydrosphere, atmosphere or cryosphere dynamics. The goal of

CSDMS is simple: to expedite scientific discovery and eliminate

duplication of effort by sharing computational resources.



As part of these efforts, CSDMS has designed a new tool for

hypothesis-driven modeling; the CSDMS Python Modeling Tool (PyMT)

provides a unified framework that allows users to interactively run and

couple numerical models written in a variety of programming languages.

These coupled models can operate on disparate time and space scales,

which are then resolved by PyMT. Principally, the PyMT is three things:

(1) a collection of Earth-surface models wrapped with a common

interface, (2) an extensible plug-in framework into which new models can

be incorporated, and (3) tools for coupling models that operate on a

variety of spatial grids and time steps.



Currently, the PyMT model collection consists of several dozen

Earth-surface models that cover a variety of process domains that range

from land, to coast, to ocean. Each of these models were contributed by

CSDMS community members to the CSDMS Model Repository as standalone

models. There was no initial intent for these models to be part of a

larger framework. The heterogeneity of the collection is represented not

just in the variety of programming languages but also by idiosyncratic

user interfaces (e.g. model specific input and output file formats).



All models within the PyMT collection are wrapped in a single, unified,

interface within the Python programming language. An overriding tenet of

the PyMT is: *if you know how to use one pymt model, you know how to use

all pymt models*. The PyMT model interface allows users to interactively

run models within a Python kernel such that they can advance models

through time while dynamically changing their state variables. This

allows users to become model composers by orchestrating different model

functionality within a script, while being able to leverage the power of

the Python programming language and its powerful collection of

third-party packages (e.g. numpy, scipy, matplotlib, xarray, dask).



The PyMT provides an extensible plugin framework that allows additional

models to be easily incorporated into the PyMT framework. This allows

new models, written by domain experts, to become PyMT models usable by a

broad community in potentially novel ways. To be incorporated into PyMT,

new models must be written to expose a Basic Model Interface (BMI). At

its core, the BMI is simply a specification that defines the necessary

functions a model must provide to make it coupleable. These functions

control how a model is initialized, updates through time, as well as how

a model provides its output variables or ingests externally provided

input variables. The CSDMS modeling stack additionally provides tools

for automatically wrapping models written in several programming

languages (currently C, C++, Fortran, and Python) into PyMT. These four

languages cover a significant majority of the models in the CSDMS Model

Repository.



The PyMT contains a collection of tools useful for model coupling

(either model-to-model or model-to-data coupling). As mentioned

previously, the CSDMS Model Repository is a heterogeneous collection of

models that were not necessarily written with the intent of coupling to

other models. As such, a significant obstacle when coupling models is

transferring values from one model's solution grid to another. Included

with the PyMT are a set of grid mappers, based on the Earth System

Modeling Framework (ESMF) grid mapping library. This allows for

efficient mapping of values between large grids. Other coupling tools

include:



*  Unit conversion utilities, which conform to the cfunits conventions,

   when models provide the same values but with different units,

*  Time interpolators that estimate state variables between model

   timesteps,

*  Output writers that write model output to standardized netCDF files

   that conform to the UGRID/SGRID specifications



PyMT is designed to expedite the process of exploring ideas, testing

hypotheses, and comparing models with data, and make Earth-surface

process models more accessible.



For proof of concept, we present an example of a coupled hydrodynamic

and bedrock incision model in PyMT. This work uses two components from

the Landlab model, OverlandFlow and DetachmentLtdErosion, as they are

implemented in PyMT (Adams et. al, 2017; Hobley et al., 2017). The

OverlandFlow component was originally developed to bridge the gap

between fully hydrodynamic models used to model single hydrograph events

and simplified 'steady-state' hydrology components used in long-term

fluvial geomorphology models. Figure 1 illustrates the difference

between these two model types: 'steady-state' models often simplify

rainfall and runoff into steady, constant values that drive steady,

constant incision rates (Figure 1a), while non-steady models can take

rainfall stochasticity into account and drive individual event

hydrographs and changing incision rates through time (Figure 1b). As

computational efficiency and speed have improved over the last decade,

more efforts have been made to bring hydrodynamics into models of

landscape evolution. The Landlab OverlandFlow model is an open-source

tool designed to achieve that goal.






**Figure 1**. Comparison of steady (**a**) and nonsteady (**b**)

hydrology models. The latter case illustrates how the OverlandFlow model

is implemented. A sample OverlandFlow workflow is also shown.



In this presentation, we run several test cases in PyMT to illustrate

landscape sensitivity to rainfall parameters, hydrograph shape and basin

orientation. These results are compared against traditional steady-state

model results. Landscapes eroded and evolved using nonsteady methods are

characterized by greater relief and increased channel concavities when

compared to steady results, suggesting that hydrodynamics should be

considered when studying the impact of bedrock river incision on

topographic evolution over long timescales.



The implementation of OverlandFlow and DetachmentLtdErosion is just one

example of coupled hydrology-sedimentology modeling available through

PyMT. This presentation will provide detailed background on how models

can be brought into the PyMT

framework, how PyMT resolves grid and temporal differences across

models, the existing hydrologic and sedimentologic tools in PyMT, and

examples of model output from the OverlandFlow and DetachmentLtdErosion

models.



**References**



*  Adams, J.M., Gasparini, N.M., Hobley, D.E., Tucker, G.E., Hutton, E.,

   Nudurupati, S.S. and Istanbulluoglu, E., 2017. The Landlab v1.0 OverlandFlow

   component: a Python tool for computing shallow-water flow across watersheds.

   Geoscientific Model Development, 10(4).

*  Brunner, G. W. (2016). HEC-RAS river analysis system 2D modeling user's

   manual. US Army Corps of Engineers---Hydrologic Engineering Center, 1-171.

*  Gochis, D.J., M. Barlage, A. Dugger, K. FitzGerald, L. Karsten, M.  McAllister,

   J. McCreight, J.  Mills, A. RafieeiNasab, L. Read, K. Sampson, D. Yates,

   W. Yu, 2018.  The WRF-Hydro modeling system technical description,

   (Version 5.0).  NCAR Technical Note. 107 pages. DOI:10.5065/D6J38RBJ

*  Hobley, D.E., Adams, J.M., Nudurupati, S.S., Hutton, E.W., Gasparini, N.M.,

   Istanbulluoglu, E.  and Tucker, G.E., 2017. Creative computing with Landlab:

   an open-source toolkit for building, coupling, and exploring two-dimensional

   numerical models of Earth-surface dynamics. Earth Surface Dynamics, 5(1), p.21.

*  USDA ARS Grassland Soil and Water Research Laboratory; Texas A&M AgriLife

   Research, 2018\. SWAT - Soil and Water Assessment Tool. USDA Agricultural

   Research Service; Texas A&M AgriLife Research.

   https://data.nal.usda.gov/dataset/swat-soil-and-water- assessment-tool