pith. sign in

arxiv: 2606.24563 · v1 · pith:MLIOTTV2new · submitted 2026-06-23 · ⚛️ physics.ao-ph

An observationally constrained probabilistic trigger for organized deep convection in an NWP ensemble

Mark R. Muetzelfeldt , Robert S. Plant , Hannah M. Christensen , Zhixiao Zhang , Tim Woollings , Alison Stirling , Warren Tennant , Mitch Moncrieff This is my paper

Pith reviewed 2026-06-25 21:46 UTC · model grok-4.3

classification ⚛️ physics.ao-ph
keywords stochastic parametrizationorganized convectionmesoscale convective systemsensemble spreadtropical precipitationnumerical weather predictionprobabilistic triggertotal column water vapour
0
0 comments X

The pith

A probabilistic trigger based on total column water vapour improves ensemble spread for tropical precipitation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a stochastic parametrization for organized deep convection that activates a mesoscale scheme probabilistically only where total column water vapour is enhanced and according to chosen spatiotemporal patterns. It compares this new version to the original MCSP scheme and a control in an NWP ensemble. The new trigger maintains gains in precipitation scales while increasing forecast spread and reducing underdispersion. A reader would care because more realistic ensemble spread supports better probability statements about tropical rainfall.

Core claim

The observationally constrained probabilistic trigger for the MCSP scheme, activated in regions of enhanced total column water vapour using given spatiotemporal scales, boosts ensemble spread for tropical precipitation relative to the original MCSP and thereby improves the spread-error relationship.

What carries the argument

The probabilistic trigger for the multiscale coherent structure parametrization (MCSP) that activates in areas of enhanced environmental total column water vapour.

If this is right

  • Both the original and new MCSP schemes improve the spatiotemporal scales of tropical precipitation compared to a control.
  • The new scheme boosts spread compared to original MCSP.
  • This improves the underdispersion of the ensemble seen with the original MCSP.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar environmental constraints could be tested on other stochastic schemes that currently lack observational anchors.
  • The trigger may affect the timing and location statistics of extreme rainfall events beyond the scales already examined.
  • Running the scheme in different ensemble sizes or resolutions could reveal whether the spread gain scales with model configuration.

Load-bearing premise

That enhanced total column water vapour provides a reliable observational basis for probabilistically triggering the representation of mesoscale convective systems.

What would settle it

If the new trigger produces no measurable increase in ensemble spread or leaves the spread-error relationship underdispersed for tropical precipitation.

Figures

Figures reproduced from arXiv: 2606.24563 by Alison Stirling, Hannah M. Christensen, Mark R. Muetzelfeldt, Mitch Moncrieff, Robert S. Plant, Tim Woollings, Warren Tennant, Zhixiao Zhang.

Figure 1
Figure 1. Figure 1: Instantaneous precipitation rates over the Indian Ocean in (a) IMERG coarsened to the resolution of the model grid and (b) Control, (c) PRIME-MCSP, and (d) Stoch-PRIME￾MCSP simulations at 08:40 local solar time (LST; from centre of domain) on 4 February 2020. We assess the degree of spatiotemporal organization using ASoP analysis (Section 2.4). The results are from a single, non-deterministic ensemble memb… view at source ↗
Figure 2
Figure 2. Figure 2: Fractional contribution of binned rate of precipitation to the total over the tropics for IMERG (four top-left panels) and Control, PRIME-MCSP and Stoch-PRIME-MCSP (one per quadrant). The bin ranges are above each panel, and are referred to in the text as low, moderate, heavy, and extreme. The sum for each grid cell in a four-panel quadrant is unity. We investigate how well the different simulations match … view at source ↗
Figure 3
Figure 3. Figure 3: Spatiotemporal correlation of precipitation over the tropics for (a) IMERG and (b) Control, (c) PRIME-MCSP, and (d) Stoch-PRIME-MCSP simulations. The spatial correlation of each grid cell at a given distance from the reference grid cell is shown in the x direction, and the temporal correlation of each grid cell with the reference grid cell is shown in the y direction. A 3-hour time average is used, as in K… view at source ↗
Figure 4
Figure 4. Figure 4: Precipitation spread and error over the tropics for the three configurations, at three scales of spatial smoothing (σ). Spread is given by dRMSE (dashed lines), and error by eRMSE (solid lines; Section 2.5). (a) σ = 0 is no spatial smoothing (i.e., at the grid scale), (b) σ = 2 ap￾plies Gaussian smoothing with a kernel that has an e-folding distance of 2 times the grid spacing in the zonal direction, and s… view at source ↗
read the original abstract

A novel stochastic parametrization scheme representing organized convection is described. The effects of mesoscale convective systems (MCSs) are represented in an observationally constrained manner, by probabilistically triggering an MCS scheme in regions of enhanced environmental total column water vapour. In combination with the probabilistic trigger, patterns with given spatiotemporal scales determine where and when the scheme is active. Our scheme builds on the multiscale coherent structure parametrization (MCSP), which represents the top-heavy heating structure associated with MCSs. The original and new MCSP schemes are tested in a numerical weather prediction (NWP) ensemble. Both MCSP schemes improve the spatiotemporal scales of tropical precipitation compared to a control. When the spread-error relationship of tropical precipitation is analysed, the new scheme successfully boosts spread compared to original MCSP, improving the underdispersion of the ensemble seen with the original MCSP.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper describes a novel stochastic parametrization for organized deep convection that augments the existing multiscale coherent structure parametrization (MCSP) with a probabilistic trigger active in regions of enhanced total column water vapour (TCWV), combined with prescribed spatiotemporal patterns. The scheme is tested within an NWP ensemble; both original and new MCSP versions improve the spatiotemporal scales of tropical precipitation relative to a control, while the new version additionally increases ensemble spread and reduces underdispersion in the spread-error relationship for tropical precipitation.

Significance. If the reported diagnostics hold after full verification, the work offers a practical route to increase ensemble spread for tropical precipitation by representing mesoscale convective system effects in a manner informed by environmental observations. The probabilistic-trigger approach could be extensible to other stochastic schemes where underdispersion is an issue.

major comments (2)
  1. [Abstract] Abstract: the claim of an 'observationally constrained' scheme is central to the novelty, yet the abstract (and by extension the scheme description) provides no quantitative details on how the probability function, TCWV threshold, or spatiotemporal patterns were derived from or validated against observations. Without this, it is impossible to judge whether the reported spread improvement is independent of the constraint or follows by construction from the chosen parameters.
  2. [Results / ensemble experiments] Scheme description and experimental setup: the spread-error diagnostics are presented as the key result demonstrating improvement over original MCSP, but no error bars, sample sizes, or explicit description of the ensemble configuration (member count, resolution, verification period) are supplied. This information is load-bearing for the claim that the new trigger 'successfully boosts spread'.
minor comments (2)
  1. Define the exact functional form of the probabilistic trigger (e.g., the TCWV dependence and any scaling constants) in an equation or pseudocode block.
  2. Clarify whether the spatiotemporal patterns are fixed a priori or also tuned; if the latter, state the tuning procedure.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments and positive assessment of the work. We address each major comment below and will revise the manuscript to incorporate the requested clarifications.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim of an 'observationally constrained' scheme is central to the novelty, yet the abstract (and by extension the scheme description) provides no quantitative details on how the probability function, TCWV threshold, or spatiotemporal patterns were derived from or validated against observations. Without this, it is impossible to judge whether the reported spread improvement is independent of the constraint or follows by construction from the chosen parameters.

    Authors: We agree that the abstract and scheme description would benefit from explicit quantitative details on the observational basis for the probability function, TCWV threshold, and spatiotemporal patterns. In the revised manuscript we will add these specifics (including the exact TCWV threshold value and its observational source, the functional form of the probability, and the derivation of the pattern scales) to both the abstract and the methods section. This will make clear how the constraint was applied and allow assessment of whether the spread gains are independent of the parameter choices. revision: yes

  2. Referee: [Results / ensemble experiments] Scheme description and experimental setup: the spread-error diagnostics are presented as the key result demonstrating improvement over original MCSP, but no error bars, sample sizes, or explicit description of the ensemble configuration (member count, resolution, verification period) are supplied. This information is load-bearing for the claim that the new trigger 'successfully boosts spread'.

    Authors: We accept that the current manuscript omits error bars, sample sizes, and a full description of the ensemble configuration. These details are required to substantiate the spread-error results. In the revision we will insert a complete description of the ensemble setup (member count, resolution, verification period) in the experimental setup section and add error bars or confidence intervals to the spread-error diagnostics in the results. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The provided abstract and description introduce a new probabilistic MCS trigger based on enhanced TCWV combined with spatiotemporal patterns, building on the existing MCSP scheme. Ensemble tests report improved precipitation scales and spread-error diagnostics versus control and original MCSP. No equations, parameter-fitting steps, or self-citation chains are quoted that reduce any central prediction or result to the inputs by construction. The reported outcomes derive from numerical experiments rather than definitional equivalence or fitted renaming, making the derivation self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit information on free parameters, axioms, or invented entities; the probabilistic trigger and observational constraint are described at a conceptual level only.

pith-pipeline@v0.9.1-grok · 5708 in / 959 out tokens · 25641 ms · 2026-06-25T21:46:37.079634+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

26 extracted references · 20 canonical work pages

  1. [1]

    and Colangeli, Matteo and Coleman, Danielle R

    Berner, J and Achatz, U and Batte, L and Bengtsson, Lisa and De La Camara, A and Christensen, H M and Colangeli, M and Coleman, D R B and Crommelin, D and Dolaptchiev, S I and others. Stochastic parameterization: towards a new view of weather and climate models. Bulletin of the American Meteorological Society. 2017. doi:10.1175/BAMS-D-15-00268.1

    work page doi:10.1175/bams-d-15-00268.1 2017

  2. [2]

    Stochastic representation of model uncertainties in the ECMWF ensemble prediction system

    Buizza, Roberto and Milleer, M and Palmer, Tim N. Stochastic representation of model uncertainties in the ECMWF ensemble prediction system. Quarterly Journal of the Royal Meteorological Society. 1999. doi:10.1002/qj.49712556006

    work page doi:10.1002/qj.49712556006 1999

  3. [3]

    Quarterly Journal of the Royal Meteorological Society

    Bush, Stephanie J and Turner, Andrew G and Woolnough, Steven J and Martin, Gill M and Klingaman, Nicholas P. Quarterly Journal of the Royal Meteorological Society. 2015. doi:10.1002/qj.2371

    work page doi:10.1002/qj.2371 2015

  4. [4]

    2023 , publisher=

    Bush, Mike and Boutle, Ian and Edwards, John and Finnenkoetter, Anke and Franklin, Charmaine and Hanley, Kirsty and Jayakumar, Aravindakshan and Lewis, Huw and Lock, Adrian and Mittermaier, Marion and others , journal=. 2023 , publisher=

    2023

  5. [5]

    Effects of organized convection parameterization on the MJO and precipitation in E3SMv1

    Chen, C C and Richter, J H and Liu, C and Moncrieff, M W and Tang, Q and Lin, W and Xie, S and Rasch, P J. Effects of organized convection parameterization on the MJO and precipitation in E3SMv1. Part I: Mesoscale heating. Journal of Advances in Modeling Earth Systems. 2021. doi:10.1029/2020MS002401

    work page doi:10.1029/2020ms002401 2021

  6. [6]

    A spatial view of ensemble spread in convection permitting ensembles

    Dey, Seonaid RA and Leoncini, Giovanni and Roberts, Nigel M and Plant, Robert S and Migliorini, Stefano. A spatial view of ensemble spread in convection permitting ensembles. Monthly Weather Review. 2014

    2014

  7. [7]

    A global high-resolution mesoscale convective system database using satellite-derived cloud tops, surface precipitation, and tracking

    Feng, Zhe and Leung, L Ruby and Liu, Nana and Wang, Jingyu and Houze Jr, Robert A and Li, Jianfeng and Hardin, Joseph C and Chen, Dandan and Guo, Jianping. A global high-resolution mesoscale convective system database using satellite-derived cloud tops, surface precipitation, and tracking. Journal of Geophysical Research: Atmospheres. 2021. doi:10.1029/20...

    work page doi:10.1029/2020jd034202 2021

  8. [8]

    Keller Jordan, Yuchen Jin, Vlado Boza, You Jiacheng, Franz Cesista, Laker Newhouse, and Jeremy Bernstein

    Hersbach, Hans and Bell, Bill and Berrisford, Paul and Hirahara, Shoji and Hor \'a nyi, Andr \'a s and Muñoz-Sabater, Joaqu \' n and Nicolas, Julien and Peubey, Carole and Radu, Raluca and Schepers, Dinand and others. The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society. 2020. doi:10.1002/qj.3803

    work page doi:10.1002/qj.3803 2020

  9. [9]

    Mesoscale convective systems

    Houze Jr., Robert A. Mesoscale convective systems. Reviews of Geophysics. 2004. doi:10.1029/2004RG000150

    work page doi:10.1029/2004rg000150 2004

  10. [10]

    Integrated Multi-satellitE Retrievals for GPM (IMERG) technical documentation

    Huffman, George J and Bolvin, David T and Nelkin, Eric J and Tan, Jackson. Integrated Multi-satellitE Retrievals for GPM (IMERG) technical documentation. NASA/GSFC Code. 2020

    2020

  11. [11]

    Met Office MOGREPS-G initialisation using an ensemble of hybrid four-dimensional ensemble variational (En-4DEnVar) data assimilations

    Inverarity, GW and Tennant, WJ and Anton, L and Bowler, NE and Clayton, AM and Jardak, M and Lorenc, AC and Rawlins, F and Thompson, SA and Thurlow, MS and others. Met Office MOGREPS-G initialisation using an ensemble of hybrid four-dimensional ensemble variational (En-4DEnVar) data assimilations. Quarterly Journal of the Royal Meteorological Society. 202...

    work page doi:10.1002/qj.4431 2023

  12. [12]

    ASoP (v1

    Klingaman, Nicholas P and Martin, Gill M and Moise, Aurel F. ASoP (v1. 0): A set of methods for analyzing scales of precipitation in general circulation models. Geoscientific Model Development. 2017. doi:10.5194/gmd-10-57-2017

    work page doi:10.5194/gmd-10-57-2017 2017

  13. [13]

    2016 , doi=

    Kober, Kirstin and Craig, George C , journal=. 2016 , doi=

    2016

  14. [14]

    The performance of the CoMorph-A convection package in global simulations with the Met Office Unified Model

    Lock, AP and Whitall, M and Stirling, AJ and Williams, KD and Lavender, SL and Morcrette, C and Matsubayashi, K and Field, PR and Martin, G and Willett, M and others. The performance of the CoMorph-A convection package in global simulations with the Met Office Unified Model. Quarterly Journal of the Royal Meteorological Society. 2024. doi:10.1002/qj.4781

    work page doi:10.1002/qj.4781 2024

  15. [15]

    Representing convective organization in prediction models by a hybrid strategy

    Moncrieff, Mitchell W and Liu, Changhai. Representing convective organization in prediction models by a hybrid strategy. Journal of the Atmospheric Sciences. 2006. doi:10.1175/JAS3812.1

    work page doi:10.1175/jas3812.1 2006

  16. [16]

    Simulation, modeling, and dynamically based parameterization of organized tropical convection for global climate models

    Moncrieff, Mitchell W and Liu, Changhai and Bogenschutz, Peter. Simulation, modeling, and dynamically based parameterization of organized tropical convection for global climate models. Journal of the Atmospheric Sciences. 2017. doi:10.1175/JAS-D-16-0166.1

    work page doi:10.1175/jas-d-16-0166.1 2017

  17. [17]

    Hydrology and Earth System Sciences , volume=

    Evaluation of Asian summer precipitation in different configurations of a high-resolution general circulation model in a range of decision-relevant spatial scales , author=. Hydrology and Earth System Sciences , volume=. 2021 , publisher=

    2021

  18. [18]

    Environmental conditions affecting global mesoscale convective system occurrence

    Muetzelfeldt, Mark R and Plant, Robert S and Christensen, Hannah M and Zhang, Zhixiao and Woollings, Tim and Feng, Zhe and Li, Puxi. Environmental conditions affecting global mesoscale convective system occurrence. Journal of the Atmospheric Sciences. 2025. doi:10.1175/JAS-D-24-0058.1

    work page doi:10.1175/jas-d-24-0058.1 2025

  19. [19]

    Storm morphology and rainfall characteristics of TRMM precipitation features

    Nesbitt, Stephen W and Cifelli, Robert and Rutledge, Steven A. Storm morphology and rainfall characteristics of TRMM precipitation features. Monthly Weather Review. 2006. doi:10.1175/MWR3200.1

    work page doi:10.1175/mwr3200.1 2006

  20. [20]

    Stochastic parametrization and model uncertainty

    Palmer, Tim N and Buizza, Roberto and Doblas-Reyes, F and Jung, Thomas and Leutbecher, Martin and Shutts, Glenn J and Steinheimer, Martin and Weisheimer, Antje. Stochastic parametrization and model uncertainty. 2009. doi:10.21957/ps8gbwbdv

    work page doi:10.21957/ps8gbwbdv 2009

  21. [21]

    A stochastic parameterization for deep convection based on equilibrium statistics

    Plant, Robert S and Craig, George C. A stochastic parameterization for deep convection based on equilibrium statistics. Journal of the Atmospheric Sciences. 2008. doi:10.1175/2007JAS2263.1

    work page doi:10.1175/2007jas2263.1 2008

  22. [22]

    Assessing the spatial and temporal variation in the skill of precipitation forecasts from an NWP model

    Roberts, Nigel. Assessing the spatial and temporal variation in the skill of precipitation forecasts from an NWP model. Meteorological Applications. 2008. doi:10.1002/met.57

    work page doi:10.1002/met.57 2008

  23. [23]

    Improved stochastic physics schemes for global weather and climate models

    Sanchez, Claudio and Williams, Keith D and Collins, Matthew. Improved stochastic physics schemes for global weather and climate models. Quarterly Journal of the Royal Meteorological Society. 2016. doi:10.1002/qj.2640

    work page doi:10.1002/qj.2640 2016

  24. [24]

    Terra Incognita

    Schumacher, Courtney and Houze Jr, Robert A and Kraucunas, Ian. The tropical dynamical response to latent heating estimates derived from the TRMM precipitation radar. Journal of the Atmospheric Sciences. 2004. doi:10.1175/1520-0469(2004)061<1341:TTDRTL>2.0.CO;2

    work page doi:10.1175/1520-0469(2004)061 2004

  25. [25]

    2011 , doi=

    Tennant, Warren J and Shutts, Glenn J and Arribas, Alberto and Thompson, Simon A , journal=. 2011 , doi=

    2011

  26. [26]

    Advancing organized convection representation in the Unified Model: Implementing and enhancing multiscale coherent structure parameterization

    Zhang, Zhixiao and Christensen, Hannah M and Muetzelfeldt, Mark R and Woollings, Tim and Plant, Robert S and Stirling, Alison J and Whitall, Michael A and Moncrieff, Mitchell W and Chen, Chih-Chieh and Feng, Zhe. Advancing organized convection representation in the Unified Model: Implementing and enhancing multiscale coherent structure parameterization. J...

    work page doi:10.1029/2024ms004370 2025