Add remaining Genslip parameters to SRFConfig by Copilot · Pull Request #93 · ucgmsim/workflow

Copilot · 2026-03-02T21:10:28Z

The codebase exposed only a small subset of genslip v5.6.2 parameters. This adds the full parameter set with correct defaults extracted from the genslip source.

Changes

defaults.yaml: ~90 new parameters added under srf:, grouped by category (rise time, hybrid correlation length, rupture velocity, tapering, time shift factors, kinematic models, roughness/wavelength, miscellaneous, aseismic). rvfrac_slip_sig added to rupture_velocity section.
schemas.py: SRF_SCHEMA extended with validators for all new fields. Added KModel IntEnum for kinematic model index. rvfrac_slip_sig moved to RUPTURE_VELOCITY_SCHEMA. Fields deferred to future CLI arguments (roughnessfile, init_slip_file, read_slip_file) excluded.
realisations.py: SRFConfig dataclass extended with typed fields. kmodel/hyb_corlen_kmodel use the KModel enum; kord is int. rvfrac_slip_sig moved to RuptureVelocity.
realisation_to_srf.py: _build_genslip_command handles Enum values by serialising to their integer value.
Tests: test_srf_config_example and test_build_genslip_command_static_args updated to cover all new fields with correct types.

Design decisions

Genslip type	Python type	Notes
`0`/`1` flags	`bool`	Serialised as `0`/`1` in command
`-1`, `-999`, `1e15` sentinels	`None`	Skipped by `_build_genslip_command`
Kinematic model index (`kmodel`, `hyb_corlen_kmodel`)	`KModel` (IntEnum)	Serialised as integer value
Filter order (`kord`)	`int`	Validated with `And(int, _is_positive)`
Optional file paths	`str \| None`	Skipped when `None`
Rupture velocity slip sigma (`rvfrac_slip_sig`)	`float \| None` in `RuptureVelocity`	Moved from `SRFConfig` for future HF sync

Original prompt

This section details on the original issue you should resolve

<issue_title>Add remaining Genslip parameters into the codebase</issue_title>
<issue_description>Currently we expose a limited subset of genslip parameters in the code base, but we could do to add the rest. Here are the default values in YAML format that I have extracted from genslip_v5.6.2.c:

# Rise time and source time function values
beta_asp: 0.3
beta_deep: 0.13
beta_mid: 0.13
beta_mid_depth: 6.5
beta_mid_depth_range: 1.5
beta_shal: 0.5
beta_shal_depth: 2.0
beta_shal_depth_range: 1.0
beta_subevt: 0.1
deep_risetimedep: 17.5
deep_risetimedep_range: 2.5
deep_risetimefac: 2.0
risetime_coef: 1.6
risetimedep: DEFAULT_DEPTH_SCALING_LEVEL
risetimedep_range: DEFAULT_DEPTH_SCALING_RANGE
risetimefac: 2.0
rt_rand: 0.0
rt_scalefac: DEFAULT_RT_SCALEFAC
stype: null

# Hybrid Correlation Length (hyb_corlen) Parameters
hyb_corlen_deep_wt_end: 1.0
hyb_corlen_deep_wt_start: 0.0
hyb_corlen_dep: DEFAULT_DEPTH_SCALING_LEVEL
hyb_corlen_dep_range: DEFAULT_DEPTH_SCALING_RANGE
hyb_corlen_fac: 2.0
hyb_corlen_flag: 0
hyb_corlen_kmodel: 5
hyb_corlen_shal_wt_end: 0.0
hyb_corlen_shal_wt_start: 1.0
hyb_corlen_side_taper: 0.08

# Rupture Velocity & Fault Dimensions
fdrup_scale_slip: 0
fdrup_time: 0
rupture_delay: 0.0
rvfmax: 1.414
rvfmin: 0.25
rvfrac_slip_sig: -1.0

# Tapering & Slip Level Adjustments
truncate_zero_slip: DEFAULT_TRUNCATE_ZERO_SLIP
slip_water_level: -1.0
slip_sigma: DEFAULT_SLIP_SIGMA
rake_sigma: 15.0
fractal_rake: 0

# Time Shift Factors (tsfac)
tsfac1_scor: 0.8
tsfac1_sigma: 1.0
tsfac2_lambda_max: 5.0
tsfac2_lambda_min: -1.0
tsfac2_scor: 0.5
tsfac2_sigma: 1.0
tsfac_bzero: -0.1
tsfac_coef: 1.1
tsfac_main: -1.0e+15
tsfac_slope: -0.5

# Kinematic Models & Corner Frequencies
circular_average: DEFAULT_CIRCULAR_AVERAGE
kmodel: MAI_FLAG
kord: 4
kx_corner: null
ky_corner: null
magC: 6.3
mag_area_Acoef: -1.0
mag_area_Bcoef: -1.0
mai_wt: 0.5
modified_corners: DEFAULT_MODIFIED_CORNERS
somerville_wt: 0.5
stretch_kcorner: DEFAULT_STRETCH_KCORNER
use_gaus: 1
use_median_mag: 0
xmag_exponent: 0.5
ymag_exponent: 0.5

# Roughness & Wavelength Limits
alpha_rough: -1.0
lambda_max: -1.0
lambda_min: -1.0
roughnessfile: null
wavelength_max: 1.0e+15
wavelength_min: -1.0

# Miscellaneous Slip/Rupture/Rake Vars
asp_taper_fac: 0.05
extend_fac: -1.0
flen_max: -1.0
fwid_max: -1.0
init_slip_file: null
moment_fraction: -1.0
perturb_subfault_location: 1
rand_rake_degs: DEFAULT_RAND_RAKE_DEGS
read_slip_file: 0
rtime1_depth: 2.0
rtime1_depth_range: 1.0
rtime1_scor: 0.8
rtime1_sigma: DEFAULT_SLIP_SIGMA
rtime2_scor: 0.5
rtime2slip_exp: 0.5
rtime_rand: -1.0
set_rake: -999.0
svr_wt: 0
target_savg: -1.0
use_Mw: 1

# Aseismic & Segment Settings
aseis_flag: 0
aseis_smooth: 0
aseis_dep: 10.0
aseis_fac: -1.0
xshift: 0.0
yshift: 0.0

The tasks to complete this feature would be as follows:

Add this YAML block into the root defaults.yaml file: here.
Update the schema in schemas.py: here.
Check that genslip is correctly supplied with these values by running realisation-to-srf with an input realisation. You could probably steal one from Cesar or Ayushi and update it using the migration tool on your new branch. The genslip command is built automatically from the realisation config in _build_genslip_command.
Update the test cases in tests/test_realisation_to_srf.py and tests/test_realisation.py to include these new defaults.
</issue_description>

<agent_instructions>You may use these defaults for named constants
#define DEFAULT_DHYPO_FRAC 0.75 /* hypo at 0.75 down-dip width /
#define DEFAULT_SHYPO_STEP 20.0 / hypo spacing at 20 km along strike /
#define DEFAULT_SHYPO_MIN_OFF 1.0 / hypos start at 1.0 km along strike /
#define DEFAULT_SLIPS_TO_HYPOS 2 / #slip models = 2 times #hypos */

#define DEFAULT_DT 0.1
#define NTMAX 10000
#define SOMERVILLE_FLAG 1
#define MAI_FLAG 2
#define INPUT_CORNERS_FLAG -1
#define MINSLIP 1.0e-02

#define DEFAULT_KMODEL 2
#define DEFAULT_FLIP_AT_SURFACE 1
#define DEFAULT_STRETCH_KCORNER 0
#define DEFAULT_CIRCULAR_AVERAGE 0
#define DEFAULT_MODIFIED_CORNERS 0
#define DEFAULT_TRUNCATE_ZERO_SLIP 1
#define DEFAULT_RAND_RAKE_DEGS 60.0
#define DEFAULT_SLIP_SIGMA 0.85

#define DEFAULT_TSFAC -0.5
#define DEFAULT_TSFAC_COEF 1.8
#define DEFAULT_TSFAC_FACTOR 1

#define DEFAULT_RT_SCALEFAC 1

#define DEFAULT_VR_TO_VS_FRAC 0.8 /* vrup = 0.8 times local Vs /
#define DEFAULT_SHAL_VRUP_FRAC 0.7 / shallow_vrup...

Fixes Add remaining Genslip parameters into the codebase #92

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

…tests Co-authored-by: lispandfound <12835929+lispandfound@users.noreply.github.com>

lispandfound · 2026-03-02T22:00:26Z

Testing Instructions

To test that your defaults are validated by your schema:

Install the "test" optional dependencies,
Run pytest tests/test_realisation.py::test_defaults_are_loadable.

There are a couple of tests that use the SRFConfig block, the relevant tests are test_realisation_to_srf.py and test_realisation.py::test_srf_config_example.

workflow/schemas.py

lispandfound · 2026-03-02T21:43:24Z

workflow/schemas.py

+        Literal(
+            "roughnessfile",
+            description="Path to roughness file (null = no file).",
+        ): Or(str, None),


Remove this, it should be passed as a command line argument (at a later date). Make sure it is removed from the tests and yaml defaults as well.

Removed roughnessfile from schemas, realisations, defaults, and tests in commit 5401095. Also removed init_slip_file and read_slip_file which were similarly removed from SRFConfig in earlier commits.

workflow/schemas.py

tests/test_realisation.py

tests/test_realisation_to_srf.py

Co-authored-by: Jake Faulkner <jakefaulkn@gmail.com>

…ad_slip_file, rvfrac_slip_sig

lispandfound · 2026-03-04T22:18:02Z

tests/test_realisation_to_srf.py

+        "hyb_corlen_dep_range=1.5",
+        "hyb_corlen_fac=2.0",
+        "hyb_corlen_flag=0",
+        "hyb_corlen_kmodel=schemas.KModel.SUZUKI",


deserialise this like so

elif isinstance(value, Enum): return value.value

in realisation-to-srf

Added elif isinstance(value, Enum): serialised_value = str(value.value) to _build_genslip_command in commit 5401095. Enum values now serialize to their integer values (e.g., kmodel=2, hyb_corlen_kmodel=5).

lispandfound · 2026-03-04T22:22:38Z

tests/test_realisation.py

+                "tsfac_slope": -0.5,
+                "circular_average": False,
+                "kmodel": 2.0,
+                "kord": 4.0,


This should be an integer, as discussed the test is not smart enough to know this and I will update later. But for now, change this to

Suggested change

"kord": 4.0,

"kord": 4,

Fixed in commit 5401095 - "kmodel": 2 (int, from IntEnum serialization).

Co-authored-by: lispandfound <12835929+lispandfound@users.noreply.github.com>

lispandfound

I am happy to merge this once ruff is happy, the tests pass and Cesar/Felipe has had a chance to once over all the defaults and check they are reasonable or match the genslip defaults.

Copilot

Pull request overview

This PR expands the SRF/genslip configuration surface area so the workflow can drive (nearly) the full genslip v5.6.2 parameter set from realisation configs, with defaults aligned to genslip’s source defaults.

Changes:

Added many missing genslip SRF parameters (with defaults) under srf: in the root defaults YAML, and added rvfrac_slip_sig to rupture_velocity.
Extended SRF_SCHEMA with validators for the new SRF fields and introduced KModel as an IntEnum.
Extended SRFConfig/RuptureVelocity and updated genslip command building to correctly serialise Enum values; updated tests accordingly.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
`workflow/scripts/realisation_to_srf.py`	Serialises `Enum`/`IntEnum` SRFConfig values into genslip CLI args; adds `rvfrac_slip_sig` handling.
`workflow/schemas.py`	Adds `KModel` enum and expands SRF/RV schema with validators for many new genslip fields.
`workflow/realisations.py`	Extends `SRFConfig` and `RuptureVelocity` dataclasses with typed fields for the new parameters.
`workflow/default_parameters/root/defaults.yaml`	Adds the expanded SRF defaults and `rvfrac_slip_sig` default.
`tests/test_realisation_to_srf.py`	Updates genslip command expectations and config construction to include new fields.
`tests/test_realisation.py`	Updates SRF config example expectations to include the new fields/types.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-05T21:01:52Z

workflow/schemas.py

+        Literal(
+            "rvfrac_slip_sig",
+            description="Rupture velocity fraction slip sigma (null = disabled)."): Or(
+            NUMBER, None


rvfrac_slip_sig is a sigma, but the schema allows any NUMBER (including negatives) when not null. Since the sentinel disable value is represented as null/None, it would be safer to validate the numeric case as non-negative (e.g., Or(And(NUMBER, _is_non_negative), None)) to prevent passing invalid negative sigmas through to genslip.

Suggested change

NUMBER, None

And(NUMBER, _is_non_negative), None

Copilot · 2026-03-05T21:01:53Z

workflow/realisations.py

+    stype: str | None
+    """Slip time function type for genslip (None = use genslip default)."""


stype is annotated as str | None, but the schema validates it with Use(Stype), which will produce a schemas.Stype (a StrEnum). For consistency (and clearer typing for callers), consider changing the field type to schemas.Stype | None (and updating the docstring accordingly).

Suggested change

stype: str | None

"""Slip time function type for genslip (None = use genslip default)."""

stype: schemas.Stype | None

"""Slip time function type for genslip as a schemas.Stype (None = use genslip default)."""

Copilot · 2026-03-05T21:01:53Z

workflow/realisations.py

+    hyb_corlen_flag: bool
+    """Enable hybrid correlation length model."""
+    hyb_corlen_kmodel: schemas.KModel
+    """Hybrid correlation length k-model index. Stored as float for schema compatibility."""


The hyb_corlen_kmodel docstring says it's "Stored as float for schema compatibility", but the field is now typed as schemas.KModel. Updating the docstring to reflect the enum (and/or the schema coercion) would avoid confusion for future readers.

Suggested change

"""Hybrid correlation length k-model index. Stored as float for schema compatibility."""

"""Hybrid correlation length k-model (schemas.KModel), parsed/coerced from the schema value."""

Copilot · 2026-03-05T21:01:53Z

workflow/schemas.py

+        Literal(
+            "hyb_corlen_kmodel",
+            description="Hybrid correlation length k-model index.",
+        ): And(NUMBER, _is_positive),


hyb_corlen_kmodel is typed as KModel in SRFConfig, but the schema currently validates it as And(NUMBER, _is_positive). This both (a) rejects the valid KModel.INPUT_CORNERS = -1 value and (b) allows non-integer floats. Consider validating/coercing with Use(KModel) (or an explicit membership check against KModel) so YAML integers are converted to the enum and invalid numeric values are rejected.

Suggested change

): And(NUMBER, _is_positive),

): Use(KModel),

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated no new comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Initial plan

3dcde70

Copilot AI assigned Copilot and lispandfound Mar 2, 2026

Copilot started work on behalf of lispandfound March 2, 2026 21:10 View session

Add remaining genslip parameters to defaults, schema, SRFConfig, and …

c6744a3

…tests Co-authored-by: lispandfound <12835929+lispandfound@users.noreply.github.com>

Copilot AI changed the title ~~[WIP] Add remaining Genslip parameters into the codebase~~ Add remaining Genslip parameters to SRFConfig Mar 2, 2026

Copilot finished work on behalf of lispandfound March 2, 2026 21:24

lispandfound self-requested a review March 2, 2026 21:51

lispandfound assigned sungeunbae Mar 2, 2026

lispandfound reviewed Mar 3, 2026

View reviewed changes

sungeunbae and others added 4 commits March 4, 2026 16:37

Apply suggestion from @lispandfound

6834147

Co-authored-by: Jake Faulkner <jakefaulkn@gmail.com>

Updated handlings of KModel, kord, roughtnessfile, init_slip_file, re…

5968e3e

…ad_slip_file, rvfrac_slip_sig

Updated handlings of KModel, kord, roughtnessfile, init_slip_file, re…

3a99d04

…ad_slip_file, rvfrac_slip_sig

kord is integer 4

beee5ef

lispandfound requested changes Mar 4, 2026

View reviewed changes

Copilot started work on behalf of lispandfound March 4, 2026 22:23 View session

Fix enum serialization, remove stale fields, fix kmodel/kord in tests

5401095

Co-authored-by: lispandfound <12835929+lispandfound@users.noreply.github.com>

Copilot finished work on behalf of lispandfound March 4, 2026 22:32

serialise rvfrac slip sigma

017d3b2

lispandfound marked this pull request as ready for review March 5, 2026 20:59

Copilot AI review requested due to automatic review settings March 5, 2026 20:59

Copilot started reviewing on behalf of lispandfound March 5, 2026 20:59 View session

fix(test_hf): set rvfrac_slip_sig to None

f0cfeeb

lispandfound reviewed Mar 5, 2026

View reviewed changes

Copilot AI reviewed Mar 5, 2026

View reviewed changes

lispandfound requested review from capajaro and Copilot March 5, 2026 21:02

Copilot started reviewing on behalf of lispandfound March 5, 2026 21:02 View session

Copilot AI reviewed Mar 5, 2026

View reviewed changes

		stype: str \| None
		"""Slip time function type for genslip (None = use genslip default)."""

	"""Hybrid correlation length k-model index. Stored as float for schema compatibility."""
	"""Hybrid correlation length k-model (schemas.KModel), parsed/coerced from the schema value."""

Conversation

Copilot AI commented Mar 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes

Design decisions

Uh oh!

lispandfound commented Mar 2, 2026

Testing Instructions

Uh oh!

Uh oh!

lispandfound Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 4, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

lispandfound Mar 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 4, 2026

Choose a reason for hiding this comment

Uh oh!

lispandfound Mar 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 4, 2026

Choose a reason for hiding this comment

Uh oh!

lispandfound left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Copilot AI commented Mar 2, 2026 •

edited

Loading