1 of 48

An Intro to SAIL and the Weather Research and Forecasting (WRF) Model��2024 ARM Summer School�May 20, 2024�

1

Monday, May 20, 2024

Dan Feldman, drfeldman@lbl.gov PI of SAIL,

Lawrence Berkeley National Laboratory

^��

Will Rudisill (LBNL), Zexuan Xu (LBNL), and many others

2 of 48

Presentation Outline

2

Overview and motivation for the SAIL field campaign.

What we learn from field campaigns.

The importance of process models for advancing science.

Overview of the Weather Research and Forecasting (WRF) Model.

Running WRF and analyzing WRF data.

3 of 48

Who I am and What I do

3

Staff Scientist at Berkeley Lab �2021 - Present�
Research Scientist at Berkeley Lab 2015 - 2021�
Project Scientist at Berkeley Lab, 2011 - 2015

Principal Investigator of SAIL.

Principal Investigator of Climate Model Downscaling for the 5th National Climate Assessment.

4 of 48

Water Resources are Threatened

4

The Colorado River has been drying up.

Discharge decreasing 5.2% per degree F warming.

Major implications for water resources across the West, including Southern California.

Figures from Woodburn, Rhoades, Feldman et al, 2021

5 of 48

A Range of Projections of Water Availability in the Future

This is a big range!

Figures from Woodburn, Rhoades, Feldman et al, 2021

The Department of Energy supports the development of long-range computer models to predict how much water will be available in the future. To capture the dynamics that impact western water, these models need to cover the whole-earth, since it really is all inter-connected.

A recent paper from my group looked at some of the computer models that DOE is developing along with essentially all of the other models that have been developed by scientists in the US and abroad to develop a probable range of where we think water resources are heading. We looked at snow-water equivalent, which indicates how much water is available in the snowpack to melt out into reservoirs. Across the West, we expect to lose about 25% of snow water in the first half of the 21^st Century and that will increase to about 50%. I want to also point out the low end of the range. We should expect to lose at least 20% of snow in the first half of the 21^st Century and at least 40% by the second half of the 21^st Century. In the Rockies, it’s a little bit better than that … we should expect to lose at least 5% of snow by mid-century and at least 30% of snow by end-of-century.

The path to get there will be bumpy. Some years will be better than others. Some will be a bit worse. In general though, it’s time to tighten our belts when it comes to snow.

However, this is a very large range of projections, and I don’t show it here, but there is also a large range in how much water will make it downstream.

So let’s try to narrow that gap!

6 of 48

The Water Resource Threat Can Be Acute

6

The Bureau of Reclamation declared its first ever water shortage in August, 2021.

But the Colorado mountains received about average precipitation in Winter 2020/2021.

Where did the water go?!?!

Is this the new normal?

7 of 48

The Water Resource Threat Can Be Acute

7

The Bureau of Reclamation declared its first ever water shortage in August, 2021.

But the Colorado mountains received about average precipitation in Winter 2020/2021.

Where did the water go?!?!

Is this the new normal?

8 of 48

Berkeley Lab is Studying the Water Cycle to Find Out

P = ET + R + I + S

Precipitation Evapotranspiration Runoff Infiltration Storage

9 of 48

Berkeley Lab is Going to Where the Water Is

Figures from Western Water Assessment, 2020

10 of 48

Observational Gaps Lead to Understanding and Prediction Gaps

10

Maddox et al., 2002, BAMS

Photo Credit: billy barr (2021)

�

Photo Credit: Ryan Currier (2021)

Well, ok, so how do we narrow that gap?

More science! That really starts with more observations though. When it comes to understanding and predicting water in the West, unfortunately we start out with one hand tied behind our back. Here is a map showing the coverage of our best precipitation observations from the operational radars that NOAA runs. These radars are where the people are, not where the water is!

The headwaters of the Colorado River, that serves 40 million people is a radar black hole. I want to let that sink in for a minute. Our ability to build predictive models for the Colorado River relies on observations from a few hundred data points from weather stations. We have only a rough understanding of how much rain and snow the Colorado Rockies actually extract from the atmosphere.

So, what can we do about that? Well, first of all, it’s important to remember that mountains are difficult places to do measurements. Here’s a picture of what Christmas Eve looked like one of these headwater areas. It can get pretty wild.

Speaking of wild, there’s wildlife too. When we make measurements of snow (which is what those orange poles are doing), we have to harden our instruments to things like moose fights.

But measurements are what we need to improve those models.

11 of 48

Review paper, published in 2020 in Bulletin of the American Meteorological Society.
Borne of many papers published over at least last decade. This paper is meant to be provocative.

Implications of Observational Gaps

12 of 48

At first glance, this paper’s title is shocking.

Observationalists may find it objectionable.

How can a model inform us better than observational systems?

What if purported uber-models disagree?
What is the path forward for improved understanding if the basis for model evaluation is inferior to the model itself?

But the authors do not make these claims. Rather, they recognize that models are simplifications of real-world systems, and that measurements can reflect processes of interest but also can under-sample.

What Does it Mean for the State of the Science?

13 of 48

SAIL and the Watershed Function Scientific Focus Area

�Field Work to Inform Climate and Hydrology Modeling

And that’s where SAIL comes in. SAIL stands for the Surface Atmosphere Integrated Field Laboratory and is a major infrastructure investment in the headwaters of the Upper Colorado River. This effort is supported by an atmospheric research program in the Department of Energy called ARM and takes place near Crested Butte, Colorado. It started about a year ago to measure basically everything we can about the atmosphere in this area, while at the same time making measurements of streamflow, groundwater, snowpack, really all of those pieces of the water cycle I showed you.

Here’s what the instruments look like.

The DOE efforts are focused on using this data to improve climate forecasts, and we are working directly with NOAA to use our data and theirs to improve their weather forecasts.

The logo soup, shown below, indicates that SAIL has seeded, a large number of partnerships, across the National Labs and university ecosystems. It even precipitated, so to speak, two whole other field campaign in the same area, supported by NOAA to improve the weather and water forecasts in the mountains, and by NSF to measure just how much snow we are losing in the winter to sublimation.

Here are some pictures of the SAIL team going out into the field to measure snow with folks from NOAA, NSF, and our local partners in the area.

14 of 48

SAIL: The Surface Atmosphere Integrated Field Laboratory

14

SAIL deployed the AMF2to the East River Watershed near Crested Butte, Colorado from 09/2021 – 06/2023.

Dozens of measurements of the atmosphere and surface state simultaneously.

East River Watershed covers 300 km² of the Upper Colorado River Basin. Elevation 2500-3500 masl, ~500-1000 cm snowfall/yr, ~66-124 cm liquid equivalent precip/yr, DJF DTR (-20°C,-1°C), JJA DTR (3°C, 23°C)

This has motivated the Surface Atmosphere Integrated Field Laboratory, or SAIL campaign. Mountainous hydrology research has long recognized the important of characterizing the atmosphere along with the surface and subsurface. SAIL provides this link by partnering with a large group of researchers from the Watershed Function SFA to achieve atmosphere through bedrock observations.

The campaign started last September and extend nearly 2 years to the East River Watershed, near the town of Crested Butte, Colorado, and will seek to characterize atmospheric processes and their impacts to energy and mass budgets in the Upper Colorado River.

I included this picture, as I particularly like it since it shows both the scenery of the area and important processes. For scenery, I sincerely hope you have a change to visit, since it is not just picturesque, but also the site of a significant amount of research activity. For atmospheric processes, the figure shows that discharge is a strong function of, among other processes, precipitation, land-atmosphere interactions, and heterogeneous radiation, which models need to capture.

15 of 48

Precipitation, Clouds, Winds, Aerosols, Radiation, Temperature, Humidity …

15

Numerous datastreams are collected by ARM.
https://sail.lbl.gov/what-we-measure

Images courtesy of ARM Flickr Account

16 of 48

SAIL and the Watershed Function Scientific Focus Area

�Field Work to Inform Climate and Hydrology Modeling

And that’s where SAIL comes in. SAIL stands for the Surface Atmosphere Integrated Field Laboratory and is a major infrastructure investment in the headwaters of the Upper Colorado River. This effort is supported by an atmospheric research program in the Department of Energy called ARM and takes place near Crested Butte, Colorado. It started about a year ago to measure basically everything we can about the atmosphere in this area, while at the same time making measurements of streamflow, groundwater, snowpack, really all of those pieces of the water cycle I showed you.

Here’s what the instruments look like.

The DOE efforts are focused on using this data to improve climate forecasts, and we are working directly with NOAA to use our data and theirs to improve their weather forecasts.

The logo soup, shown below, indicates that SAIL has seeded, a large number of partnerships, across the National Labs and university ecosystems. It even precipitated, so to speak, two whole other field campaign in the same area, supported by NOAA to improve the weather and water forecasts in the mountains, and by NSF to measure just how much snow we are losing in the winter to sublimation.

Here are some pictures of the SAIL team going out into the field to measure snow with folks from NOAA, NSF, and our local partners in the area.

17 of 48

SAIL and the Watershed Function Scientific Focus Area

�Field Work to Inform Climate and Hydrology Modeling

And that’s where SAIL comes in. SAIL stands for the Surface Atmosphere Integrated Field Laboratory and is a major infrastructure investment in the headwaters of the Upper Colorado River. This effort is supported by an atmospheric research program in the Department of Energy called ARM and takes place near Crested Butte, Colorado. It started about a year ago to measure basically everything we can about the atmosphere in this area, while at the same time making measurements of streamflow, groundwater, snowpack, really all of those pieces of the water cycle I showed you.

Here’s what the instruments look like.

The DOE efforts are focused on using this data to improve climate forecasts, and we are working directly with NOAA to use our data and theirs to improve their weather forecasts.

The logo soup, shown below, indicates that SAIL has seeded, a large number of partnerships, across the National Labs and university ecosystems. It even precipitated, so to speak, two whole other field campaign in the same area, supported by NOAA to improve the weather and water forecasts in the mountains, and by NSF to measure just how much snow we are losing in the winter to sublimation.

Here are some pictures of the SAIL team going out into the field to measure snow with folks from NOAA, NSF, and our local partners in the area.

18 of 48

Atmospheric Radiation Measurement (ARM) Program: Global Climate Measurements with a Local Focus

18

maintains the world’s largest global infrastructure for obtaining observations of the natural atmosphere: Clouds, Aerosol, Precipitation, and Energy

Houston, TX

Crested Butte, CO

3 Fixed Sites:

Southern Great Plains (SGP)

North Slope Alaska (NSA)

Eastern North Atlantic (ENA)

3 Mobile Facilities:

ARM Mobile Facility 1 (AMF1)

ARM Mobile Facility 2 (AMF2)

ARM Mobile Facility 3 (AMF3)

1 Aerial Facility:

ARM Aerial Facility (AAF)

19 of 48

The ARM Data Center

19

Continuous 24/7 operations deliver quality

assured data to the

and is distributed freely worldwide within 24-48 hours of collection and processing

Data Center

20 of 48

Lots of SAIL Data!

20

21 of 48

SAIL is a Team Sport!

�Field Work to Inform Climate and Hydrology Modeling

And that’s where SAIL comes in. SAIL stands for the Surface Atmosphere Integrated Field Laboratory and is a major infrastructure investment in the headwaters of the Upper Colorado River. This effort is supported by an atmospheric research program in the Department of Energy called ARM and takes place near Crested Butte, Colorado. It started about a year ago to measure basically everything we can about the atmosphere in this area, while at the same time making measurements of streamflow, groundwater, snowpack, really all of those pieces of the water cycle I showed you.

Here’s what the instruments look like.

The DOE efforts are focused on using this data to improve climate forecasts, and we are working directly with NOAA to use our data and theirs to improve their weather forecasts.

The logo soup, shown below, indicates that SAIL has seeded, a large number of partnerships, across the National Labs and university ecosystems. It even precipitated, so to speak, two whole other field campaign in the same area, supported by NOAA to improve the weather and water forecasts in the mountains, and by NSF to measure just how much snow we are losing in the winter to sublimation.

Here are some pictures of the SAIL team going out into the field to measure snow with folks from NOAA, NSF, and our local partners in the area.

22 of 48

Collaborative Resources: Watershed Function SFA

22

SAIL is leveraging resources and expertise from the Watershed Function SFA, which is an SBR-funded research program to characterize surface and sub-surface processes in mountainous watersheds.

23 of 48

Collaborative Resources: Snow Surveys

23

SAIL has catalyzed the collection of remote-sensing snow surveys from the Airborne Snow Observatory.

ASO Snow Depth, East River, April 2019

24 of 48

Collaborative Resources: NOAA SPLASH

24

NOAA Physical Sciences Laboratory is leading the Study of Precipitation, the Lower Atmosphere, and Surface for Hydrometeorology (SPLASH). It has deployed to the East River at different, nearby locations from SAIL and ran concurrently.

It has deployed 2 surface flux stations, several disdrometers, a couple of vertically-pointing snow-level radars, and an another X-band radar to test weather and water forecasting in the Upper Colorado (UFS, HRRR, NWM)

25 of 48

The Density of Observations at the East River Watershed

25

Graphic Courtesy of Jessica Lundquist �and the SOS Campaign

To give a bit of context for the location of SAIL and the surrounding area, the East River Watershed, which is where SAIL takes place, is part of the Upper Colorado and covers 300 km2. The study area is at high elevation, ranging from 2500 to 3500 masl, with some significant gradients in precipitation. From station observations, there’s a range of about a factor of 2 in snowfall and liquid equivalent precipitation, with about 2/3 coming from snow and 1/3 from rain. It’s quite cold in the winter and pleasant in the summer, which makes nearby Crested Butte a destination not just for scientists but for skiers, hikers, mountain bikers, hunters and people taking family vacations.

The East River watershed has peaks and valleys and was chosen because it is part of this major source of water, it experiences the processes that are important for this water source, it exhibits strong spatial gradients, even at the same elevation, and, most importantly, is heavily-studied. Through the Watershed Function SFA, there is a major surface and sub-surface research component underway. The focus of modeling and research efforts at the East River, to interrogate process models and develop a thorough understanding of the processes governing this watershed, has parallels to how ARM explores atmospheric processes in intensive ways. SAIL and the Watershed Function SFA will enable a vertical integration of observations to explore the processes that govern hydrology and their interactions. There are also additional resources from the Rocky Mountain Biological Laboratory, which has been collecting weather and biological observations for many decades.

The SAIL instruments will be going to 2 locations

26 of 48

Some Cool Stuff that the Density Enables

26

To give a bit of context for the location of SAIL and the surrounding area, the East River Watershed, which is where SAIL takes place, is part of the Upper Colorado and covers 300 km2. The study area is at high elevation, ranging from 2500 to 3500 masl, with some significant gradients in precipitation. From station observations, there’s a range of about a factor of 2 in snowfall and liquid equivalent precipitation, with about 2/3 coming from snow and 1/3 from rain. It’s quite cold in the winter and pleasant in the summer, which makes nearby Crested Butte a destination not just for scientists but for skiers, hikers, mountain bikers, hunters and people taking family vacations.

The East River watershed has peaks and valleys and was chosen because it is part of this major source of water, it experiences the processes that are important for this water source, it exhibits strong spatial gradients, even at the same elevation, and, most importantly, is heavily-studied. Through the Watershed Function SFA, there is a major surface and sub-surface research component underway. The focus of modeling and research efforts at the East River, to interrogate process models and develop a thorough understanding of the processes governing this watershed, has parallels to how ARM explores atmospheric processes in intensive ways. SAIL and the Watershed Function SFA will enable a vertical integration of observations to explore the processes that govern hydrology and their interactions. There are also additional resources from the Rocky Mountain Biological Laboratory, which has been collecting weather and biological observations for many decades.

The SAIL instruments will be going to 2 locations

27 of 48

Connections to Process Modeling

27

It is impossible to measure everything, everywhere, all at once.

There is a significant role for process modeling to interpret observations and support predictive understanding.

We are going to be talking about atmospheric process modeling here, with a focus on the Weather Research and Forecasting (WRF) model.

From Noblis WRF Tutorial

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

28 of 48

Weather Research and Forecasting (WRF) Structure

28

From WSU BioEarth

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

29 of 48

Weather Research and Forecasting (WRF) Structure

29

The Weather Research and Forecasting (WRF) model is a widely-used atmospheric process model.

Serves both research and numerical weather prediction needs.

It is a computer model that solves conservation equations to predict the evolution of the atmosphere and surface from one time step to the next.

It is a freely-available community tool.

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

30 of 48

Weather Research and Forecasting (WRF) Structure

30

From WSU BioEarth

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

31 of 48

Weather Research and Forecasting (WRF) Solver

31

Equations: Fully compressible, Euler nonhydrostatic with a run-time hydrostatic option available. Conservative for scalar variables.

Prognostic Variables: Velocity components u and v in Cartesian coordinate, vertical velocity w, perturbation moist potential temperature, perturbation geopotential, and perturbation surface pressure of dry air.

Vertical Coordinate: Terrain-following, mass-based, hybrid sigma-pressure vertical coordinate based on dry hydrostatic pressure, with vertical grid stretching permitted.

Initial Conditions: Three dimensional for real-data, and one-, two- and three-dimensional for idealized data.

Lateral Boundary Conditions: Periodic, open, symmetric, and specified options available.

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

32 of 48

WRF Physics Schemes

32

WRF parameterizes physical processes that don’t’ have closed-form and/or analytical representations.

Microphysics: represent the micrometer-scales of how clouds and precipitation actually form �different schemes that support mixed-phase, multi-moment, and aerosol-aware process studies.

Cumulus parameterizations: represent the evolution of clouds�adjustment, mass-flux, and scale-aware schemes available

Surface physics: represents vegetation, soil-moisture, snow-cover and sea-ice effects on weather�simple model to full land-surface model possible

Planetary boundary layer physics: represents the transition between the surface and the free troposphere�turbulent kinetic energy prediction or non-local K schemes available

Atmospheric radiation physics: represents the propagation of shortwave and longwave energy through the atmosphere�prediction of surface, top-of-atmosphere, and intra-atmospheric fluxes with clouds �

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

33 of 48

WRF Data Assimilation

33

WRF has a built-in data-assimilation capability

Quasi-Newtonian and conjugate gradient methods available�
3DVar, 4DVar, hybrid-ensemble 3DVar available

Global, regional, and multi-model DA capability.

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

34 of 48

WRF-Chem

34

Online model with conservative transport�
Dry deposition coupled with soil/vegetation scheme

Aqueous phase chemistry coupled to some microphysics and aerosol schemes

Biogenic emissions parameterizations

Anthropogenic emissions are user-specified

Gas-phase chemical reaction calculation representations are available

Several choices of photolysis schemes

Choices of aerosol schemes

Tracer transport option available.

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

35 of 48

Initial and Boundary Conditions for WRF

35

Atmospheric process models are sensitive to their initial and boundary conditions.�

From Skamarock et al, 2019

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

36 of 48

Initial and Boundary Conditions for WRF

36

Boundary conditions are established at the coarse grid boundary.

From Skamarock et al, 2019

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

37 of 48

37

Example: Using Radar and Gauges to Evaluate WRF

As an example, in the Swiss Alps, WRF showed significant spatial pattern biases. Sometimes accumulated snowfall across the domain agrees with radar, sometimes it is biased. Gauge and radar comparisons difficult to interpret.
Different WRF simulations with and without terrain smoothing reveal that for some storms, the effects of terrain impact precipitation, but for some storms, they do not.

Gerber et al, The Cryosphere, 2018

January 31, 2016

February 4, 2016

March 5, 2016

38 of 48

38

Example: Testing WRF with Radar Fingerprints of Aggregation and Riming

With radar vertical scans, we can take a deeper dive into how precipitation is forming.
Three observed storms from the Front Range (FROST campaign) produce three different fingerprints.
The blue and brown lines correspond to storms where aggregation occurred, while the red line shows a storm with negative Z_DR gradients and changing fall velocities (V_R), indicating riming.
Radar scans enable characterization of micro and macrophysical precipitation processes to develop statistics over nearly two water years of where models are succeeding and failing and why.

Shrom and Kumjian, JAMC, 2016

39 of 48

39

SAIL WRF: An Integrated Integrated Process Model

Three specific WRF configurations were run over the SAIL domain.

Integrated Process Models (WRF + ParFlow-CLM).

Xu et al, 2023, HESS

40 of 48

40

SAIL WRF: An Integrated Integrated Process Model

Three specific WRF configurations were run over the SAIL domain.

Integrated Process Models (WRF + ParFlow-CLM).

Xu et al, 2023, HESS

41 of 48

41

SAIL WRF: An Integrated Integrated Process Model

Choices of model physics and forcings based on published WRF configurations in complex terrain

Xu et al, 2023, HESS

42 of 48

42

SAIL WRF: An Integrated Integrated Process Model

Integrated Process Models (WRF + ParFlow-CLM) show what specific atmospheric information is relevant to watershed discharge.

Modeling uncertainties in atmospheric processes within the East River Watershed dominate uncertainty in discharge.

Therefore, advancing hydrology prediction in the UCRB requires a deep-dive into atmospheric physical processes … like we are doing with SAIL.

Xu et al, 2023, HESS

43 of 48

43

Diving Into SAIL WRF: Dimensions

We will not be running WRF for the ARM Summer School (but let’s follow up if you want to).

We will be diving into WRF outputs and using Jupyter notebooks to display and analyze simulations and show you how to compare to observations.

44 of 48

44

Diving Into SAIL WRF: Surface Variables

Directly, continuously measured

Indirectly or occasionally measured

45 of 48

45

Diving Into SAIL WRF: Atmospheric Variables

Directly, continuously measured

Indirectly or occasionally measured

46 of 48

46

Diving Into SAIL WRF: Flux Variables

Directly, continuously measured

Indirectly or occasionally measured

47 of 48

Summary Discussion !!!

47

SAIL and ARM has built and is looking to build more partnerships to understand how water is produced in the West and use this as a basis to predict how it will change. �
We look forward to working with you this week on analyzing SAIL and other ARM datasets and also diving into WRF analysis.

Look to the Intro to xwrf tutorial�
Email me! drfeldman@lbl.gov

Visit our website: https://sail.lbl.gov

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.

48 of 48

Acknowledgements

48

This work was supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research and the Atmospheric System Research under U.S. Department of Energy Contract No. DE-AC02-05CH11231.

The National Oceanic and Atmospheric Administration (NOAA) supported SPLASH observations and analysis.

The National Science Foundation through the Earth Observing Laboratory supported SOS observations and analysis.

Image courtesy of ARM Flickr Account

So before describing what the SAIL campaign is and what science it intends to pursue, I would like to give you a quick background. Approximately have of the world depends on water resources derived from mountains. However, these resources are threatened. You can see here that in the Mountain West, many station observations show negative trends in the seasonal snow they receive.

It would be really helpful to know how these resources will be evolving in the future, but there is a fundamental question as to what Earth System Models actually need to represent in mountains to be able to have predictive power for these resources. There need to be datasets and modeling to clarify which physical processes, broadly speaking by which mountains extract moisture from the air and deliver water, that need to be represented in these models to have confidence in their projections.