- Rating
- 4
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
The paper addresses the challenge of probabilistic nowcasting of three-dimensional (3D) radar data. Compared to two-dimensional (2D) data, generation is ensembles is often expensive, which presents a particular challenge given that nowcasting is time-critical. The existing works use different techniques to reduce the amount of computational resources required to produce the ensemble of forecasts. To this end, the state-of-the-art works use low-rank approximations, sampling voxels independently in a way that sacrifices between-voxel correlations. The authors pursue an alternative strategy, which they demonstrate not only to be computationally efficient but also, on some metrics, outperforming the state-of-the-art FlowCast. The authors implement the ideas of copula learning, separating the dependency structure from marginal distributions, using an optimal-transport technique for observed echoes. This corresponds to the known practice in motion-based video prediction, where the motion is modelled separately from the appearance.
Critical strengths
Soundness: the problem of making 3D nowcasting more efficient is an important conceptual problem in radar nowcasting. The authors propose a principled solution to this problem.
Clarity: the paper is clearly written (in general, but see the question below)
Quality: the evaluation includes a number of benchmarks, such as SEVIR, NEXRAD-3Y, TRRD. They reproduce baselines faithfully, including CasCast and performing adaptations of FlowCast. The adaptations look reproducible and fair, and the same can be said about the proposed method. They also go to an effort of reproducing the work on 2D SEVIR data.
Significance: it is significant to see the method that works faster and comparable in terms of scores most of the time, with significant improvements on some scores on some datasets.
Critical weaknesses
Clarity: see the question below. Also, while most of the work is written well, some caption need some improvements. This includes: "SEVIR comparison with deterministic and probabilistic models. " I know that SEVIR is a dataset, but the captions needs to be clearer.
Questions for authors
Clarity: while the idea of the copula modelling appears important to the proposed solution, it is only mentioned briefly in the Introduction and at the beginning of Section 3.5. I suggest there should be some reference in the appendix aimed for the clarity of existing solutions in copula models, introducing some form of review of current works in the area, and justifying the implementation in this paper (why is it exactly that you build on Salinas et al's implementation?).
6 comments
Q (2/3). Review of existing copula models
Reviewer. ... some form of review of current works in the area ...
Answer. Copula models separate marginal predictive distributions from dependence between forecast variables. Weather-ensemble post-processing reconstructs this dependence from ensemble ranks or historical data. Learned copula models estimate dependence with neural networks or Gaussian processes (App. A.4.1 and A.4.2).
Copulas as models of dependence. By Sklar's theorem, a joint distribution can be written in terms of its marginal distributions and a copula (Sklar, 1959; Nelsen, 2006). The copula describes dependence between the variables. A Gaussian copula uses the correlation matrix of a latent Gaussian vector to specify this dependence (App. A.4.1).
-
Post-processing of weather ensembles. These methods first calibrate the marginal predictive distributions of a numerical weather prediction ensemble and then restore dependence across variables. Ensemble copula coupling (Schefzik et al., 2013) uses the rank order of the raw ensemble. The Schaake shuffle (Clark et al., 2004) uses the rank order of past observations. The Gaussian copula approach (Möller et al., 2013) estimates a Gaussian copula from past data.
-
Learned copulas for multivariate time series. The low-rank Gaussian copula process (Salinas et al., 2019) predicts the diagonal and low-rank covariance components with a recurrent network. Each series is transformed to a Gaussian scale by applying its empirical distribution function followed by the inverse standard-normal CDF. TACTiS and TACTiS-2 (Drouin et al., 2022; Ashok et al., 2024) learn a nonparametric copula with an attention-based network. Vine copulas construct multivariate dependence from pair copulas. Lopez-Paz et al. (2013) learn the parameters of the pair copulas with Gaussian processes.
-
Generative nowcasting models. CasCast (Gong et al., 2024) refines a deterministic forecast with latent diffusion. FlowCast (Perrone Ribeiro & Faganeli Pucer, 2026) generates forecasts with conditional flow matching in a latent space. Both models represent dependence across lead times and grid cells through their generative networks. In our comparisons, CasCast forecasts VIL on SEVIR and FlowCast forecasts reflectivity on the 3D radar datasets. Each ensemble member starts from a Gaussian noise sample and requires multiple network evaluations.
References.
- Ashok et al. (2024). TACTiS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time Series. ICLR.
- Clark et al. (2004). The Schaake Shuffle: A Method for Reconstructing Space–Time Variability in Forecasted Precipitation and Temperature Fields. Journal of Hydrometeorology.
- Drouin et al. (2022). TACTiS: Transformer-Attentional Copulas for Time Series. ICML.
- Gong et al. (2024). CasCast: Skillful High-Resolution Precipitation Nowcasting via Cascaded Modelling. ICML.
- Lopez-Paz et al. (2013). Gaussian Process Vine Copulas for Multivariate Dependence. ICML.
- Möller et al. (2013). Multivariate Probabilistic Forecasting Using Ensemble Bayesian Model Averaging and Copulas. Quarterly Journal of the Royal Meteorological Society.
- Nelsen (2006). An Introduction to Copulas. Springer.
- Perrone Ribeiro & Faganeli Pucer (2026). FlowCast: Advancing Precipitation Nowcasting with Conditional Flow Matching. ICLR.
- Salinas et al. (2019). High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processes. NeurIPS.
- Schefzik et al. (2013). Uncertainty Quantification in Complex Simulation Models Using Ensemble Copula Coupling. Statistical Science.
- Sklar (1959). Fonctions de répartition à n dimensions et leurs marges. Publications de l'Institut de Statistique de l'Université de Paris.
Q (3/3). How is the copula formulation used in PSVT?
Reviewer. ... justifying the implementation in this paper (why is it exactly that you build on Salinas et al's implementation?).
Answer. PSVT computes the shared volumetric factors and the local Gaussian reference mean and standard deviation once per radar history. Each member then reuses these network outputs at a cost of for forecast coordinates and factors. The path-wise predictions provide these factors without additional trainable parameters. The local predictive distributions are fitted with the closed-form CRPS term in Eq. (9). Evaluating this CRPS term requires no ensemble sampling.
PSVT as a copula model. At coordinates where the shared factor vector is nonzero the weighted deviation vector is normalized as (Eq. 5 and App. A.3). The uncensored Gaussian reference variable has mean and standard deviation :
In the Gaussian reference, one coefficient vector is shared across spatial coordinates and lead times. For coordinates and with nonzero shared factor vectors, the covariance of and is
This shared covariance component has rank at most and follows the low-rank structure of Salinas et al. (2019). Censoring this Gaussian construction to dBZ gives the local predictive distribution of Eq. (3) at these coordinates.
PSVT uses 16 paired and whitened member coefficient vectors on NEXRAD-3Y. At coordinates where , the uncensored finite ensemble exactly matches the mean and covariance of the Gaussian reference. These empirical moments use equal member weights . The censored members approximate the local predictive distributions of Eq. (3) (Eq. 7 and App. A.3).
-
Linear cost of member generation. Combining shared factors over forecast coordinates costs operations per member. PSVT uses on NEXRAD-3Y and reuses the factors computed by the network. The number of ensemble members can be chosen separately from the number of paths . The Gaussian coefficient vectors can span the coefficient dimensions for even . Pairing and whitening these vectors gives the moment conditions of Eq. (7).
-
Factors from predicted echo evolution. PSVT derives the shared factors from differences among the path-wise predictions for the current radar history. These predictions describe echo motion and changes in echo intensity. We replace the path-derived factors with random normalized factors to evaluate their contribution to dependence modelling. The local parameters and remain unchanged (Table 3 and Fig. 6). Voxel CRPS changes by 0.02% in this ablation. Regional reflectivity CRPS increases by 21.5%. The CRPS scores for the temporal change () and vertical difference (Vert. ) in regional mean reflectivity also increase. FSS at 40 dBZ decreases by 25.0%.
-
Marginal fitting without ensemble sampling. The closed-form CRPS term in Eq. (9) fits the Gaussian reference mean and standard deviation without sampling ensemble members (Sec. 3.6). At coordinates where the factor vector has unit norm. The paired and whitened coefficients then give the uncensored members a standard deviation of (Eq. 7). The empirical variance uses divisor . As an ablation, we replace the unit-length normalization of the weighted deviation vector with one global scale. Member spread then also depends on the magnitude of the path deviations. Global normalization increases CRPS by 33.9% (Table 3). The separation of marginal distributions and dependence is also used in weather-ensemble post-processing (App. A.4.2).
Reference.
- Salinas et al. (2019). High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processes. NeurIPS.
Q (1/3). Why build on Salinas et al. (2019)?
We appreciate your recognition of PSVT's principled design and reproducible evaluation.
Answer. We adopt the low-rank Gaussian covariance structure of Salinas et al. (2019) because it allows each ensemble member to reuse a small set of shared factors ak(q). Combining K factors over n forecast coordinates costs O(nK) operations per member (Eq. 6). PSVT computes the factors once and generates 16 members in 3.3 s. FlowCast generates 16 members in 98.6 s (Table 1).
Low-rank covariance structure. Salinas et al. (2019) represent latent Gaussian dependence with covariance Σ=D+VV⊤ for n variables. The diagonal entries of the nonnegative diagonal matrix D give the variances of the independent Gaussian variation. The matrix V∈Rn×r contains r shared factors with r≪n. Sampling combines these factors with the independent variation at a cost of O(nr) operations (App. A.4.1).
A forecast on NEXRAD-3Y contains 1.9×108 values across 20 lead times, 36 height levels and a 512×512 grid. PSVT computes K=3 shared factors ak(q) and the Gaussian reference mean μ(q) and standard deviation σ(q) once per radar history. Each member reuses these network outputs and is censored to [0,70] dBZ (Eq. 6).
PSVT factors, marginals and member construction. We subtract the weighted mean of the path-wise predictions from each prediction Pτ(k). We then weight and normalize these deviations to obtain the shared factors ak(q) (Sec. 3.4). The local predictive distribution is a Gaussian censored to [0,70] dBZ (Eq. 3). PSVT draws M/2 Gaussian coefficient vectors c and adds their negatives. We then whiten the paired set to obtain the member coefficient vectors c(m). With equal member weights 1/M, these vectors have zero empirical mean and identity empirical covariance (Eq. 7). The closed-form CRPS term in Eq. (9) fits the local predictive distributions without sampling ensemble members (Sec. 3.6 and App. A.4.3).
Transport in PSVT is advection of radar reflectivity. PSVT samples the latest observed volume XTin along the K predicted 3D displacement fields Δτ(k) (Eq. 1).
Reference.