This paper presents a viewpoint from computer vision to the radar echo extrapolation task in the precipitation nowcasting domain. Inspired by the success of some convolutional recurrent neural network models in this domain, including convolutional LSTM, convolutional GRU and trajectory GRU, we designed a new sequence-to-sequence neural network structure to leverage these models in a realistic data context. In this design, we decreased the numbers of channels in high abstract recurrent layers rather than increasing them. We formulated the task as a problem of encoding five radar images and predicting 10 steps ahead at the pixel level, and found that using only the common mean squared error can misguide the training and mislead the testing. Especially, the image quality of last predictions usually degraded rapidly. As a solution, we employed some visual image quality assessment techniques including Structural Similarity (SSIM) and multi-scale SSIM to train our models. Experimental results show that our structure was more tolerant to increasing uncertainty in the data, and the use of image quality metrics can significantly reduce the blurry image issue. Moreover, we found that using SSIM was very effective and a combination of SSIM with mean squared error and mean absolute error yielded the best prediction quality.

https://www.mdpi.com/2073-4433/10/5/244/pdf

Computer Vision in Precipitation Nowcasting: Applying Image Quality Assessment Metrics for Training Deep Neural Networks

Nonintrusive image-based methods have the potential to advance hydrological streamflow observations by providing spatially distributed data at high temporal resolution. Due to their simplicity, correlation-based approaches have until recent been preferred to alternative image-based approaches, such as optical flow, for camera-based surface flow velocity estimate. In this work, we introduce a novel optical flow scheme, optical tracking velocimetry (OTV), that entails automated feature detection, tracking through the differential sparse Lucas-Kanade algorithm, and then a posteriori filtering to retain only realistic trajectories that pertain to the transit of actual objects in the field of view. The method requires minimal input on the flow direction and camera orientation. Tested on two image data sets collected in diverse natural conditions, the approach proved suitable for rapid and accurate surface flow velocity estimations. Five different feature detectors were compared and the features from accelerated segment test (FAST) resulted in the best balance between the number of features identified and successfully tracked as well as computational efficiency. OTV was relatively insensitive to reduced image resolution but was impacted by acquisition frequencies lower than 7–8 Hz. Compared to traditional correlation-based techniques, OTV was less affected by noise and surface seeding. In addition, the scheme is foreseen to be applicable to real-time gauge-cam implementations.

https://www.mdpi.com/2072-4292/10/12/2010/pdf

Optical tracking velocimetry (OTV): leveraging optical flow and trajectory-based filtering for surface streamflow observations.

Broad-UNet: Multi-scale feature learning for nowcasting tasks.

The convective rainfall rate from cloud physical properties (CRPh) algorithm for Meteosat second-generation satellites is a day-only precipitation algorithm developed at the Spanish Meteorological Agency (AEMET) for EUMETSAT’ Satellite Application Facility in support of nowcasting and very short-range forecasting (NWC SAF). It is therefore mainly intended to provide input for monitoring and near-real-time forecasts for a few hours. This letter critically discusses the theoretical basis of the algorithm with special emphasis on the empirical values and assumptions in the microphysics of precipitation, and compares the qualitative performances of the CRPh with its antecessor, the convective rainfall rate algorithm (CRR), using an object-based method applied to a case-study. The analyses show that AEMET’s CRPh is physically consistent and outperforms the CRR. The applicability of the algorithm for nowcasting and the challenges of improving the product to an all-day algorithm are also presented.

The Convective Rainfall Rate from Cloud Physical Properties Algorithm for Meteosat Second-Generation Satellites: Microphysical Basis and Intercomparisons using an Object-Based Method

In order to solve the existing problems of easy spatiotemporal information loss and low forecast accuracy in traditional radar echo nowcasting, this paper proposes an encoding-forecasting model (3DCNN-BCLSTM) combining 3DCNN and bi-directional convolutional long short-term memory. The model first constructs dimensions of input data and gets 3D tensor data with spatiotemporal features, extracts local short-term spatiotemporal features of radar echoes through 3D convolution networks, then utilizes constructed bi-directional convolutional LSTM to learn global long-term spatiotemporal feature dependencies, and finally realizes the forecast of echo image changes by forecasting network. This structure can capture the spatiotemporal correlation of radar echoes in continuous motion fully and realize more accurate forecast of moving trend of short-term radar echoes within a region. The samples of radar echo images recorded by Shenzhen and Hong Kong meteorological stations are used for experiments, the results show that the critical success index (CSI) of this proposed model for eight predicted echoes reaches 0.578 when the echo threshold is 10 dBZ, the false alarm ratio (FAR) is 20% lower than convolutional LSTM network (ConvLSTM), and the mean square error (MSE) is 16% lower than the real-time optical flow by variational method (ROVER), which outperforms the current state-of-the-art radar echo nowcasting methods.

Strong Spatiotemporal Radar Echo Nowcasting Combining 3DCNN and Bi-Directional Convolutional LSTM

Short-term high-resolution quantitative precipitation forecasting (QPF) is very important for flash-flood warning, navigation safety, and other hydrological applications. This paper proposes a subpixel-based QPF algorithm using a pyramid Lucas–Kanade optical flow technique (SPLK) for short-time rainfall forecast. The SPLK tracks the storm on the subpixel level by using the optical flow technique and then extrapolates the precipitation using a linear method through redistribution and interpolation. The SPLK compares with object-based and pixel-based nowcasting algorithms using eight thunderstorm events to assess its performance. The results suggest that the SPLK can perform better nowcasting of precipitation than the object-based and pixel-based algorithms with higher adequacy in tracking and predicting severe storms in 0–2 h lead-time forecasting.

/pdf/subpixel-based-precipitation-nowcasting-with-the-pyramid-4vl46k8yjq.pdf

Subpixel-Based Precipitation Nowcasting with the Pyramid Lucas–Kanade Optical Flow Technique

Abstract Incident data, a form of big data frequently used in urban studies, are characterized by point features with high spatial and temporal resolution and categorical values. In contrast to panel data, such spatial data pooled over time reflect multi-directional spatial effects but only unidirectional temporal effects, which are challenging to analyze. This paper presents an innovative approach to address this challenge – a geographically and temporally weighted co-location quotient which includes global and local computation, a method to calculate a spatiotemporal weight matrix and a significance test using Monte Carlo simulation. This new approach is used to identify spatio-temporal crime patterns across Greater Manchester in 2016 from open source recorded crime data. The results show that this approach is suitable for the analysis and visualization of spatio-temporal dependence and heterogeneity in categorical spatial data pooled over time. It is particularly useful for detecting symmetrical spatio-temporal co-location patterns and mapping local clusters. The method also addresses the unbalanced temporal scale problem caused by unidirectional temporal data representation and explores potential impacts. The empirical evidence of the spatiotemporal crime patterns might usefully be deployed to inform the development of criminological theory by helping to disentangle the relationships between crime and the urban environment.

Geographically and temporally weighted co-location quotient: an analysis of spatiotemporal crime patterns in greater Manchester

Short-term high-resolution Quantitative Precipitation Nowcasting (SQPN), which refers to the forecasting of future precipitation within a very short time (I. e. 0–2h), is useful for flash-flood warning, navigation safety, and other hydrological and meteorological concerns.

Sub-pixel precipitation nowcasting over Guangdong Province using optical flow algorithm

A New Hybrid Cuckoo Quantum-Behavior Particle Swarm Optimization Algorithm and its Application in Muskingum Model

The Precipitation nowcasting can provide high-resolution forecasts of rainfall and hydrometeors in 2 hours and play an important role in risk management for flash flood and debris flow events, but it is a very challenge work. This study proposed a new method which combine of XGBoost method and the PPLK model for precipitation nowcasting (XGB-PPLK). The proposed method was assessed in four different types of storms, and the experimental results show that the XGB-PPLK can improve the probability of detection (POD), as while as reducing the normalized mean square error (NMSE), and maintain the basically equivalent false-alarm ratio (FAR).

/pdf/combination-of-xgboost-and-pplk-method-for-improving-the-2ghwn60e.pdf

Xiong-Fa Mai

Papers

Subpixel-Based Precipitation Nowcasting with the Pyramid Lucas–Kanade Optical Flow Technique

Geographically and temporally weighted co-location quotient: an analysis of spatiotemporal crime patterns in greater Manchester

Sub-pixel precipitation nowcasting over Guangdong Province using optical flow algorithm

A New Hybrid Cuckoo Quantum-Behavior Particle Swarm Optimization Algorithm and its Application in Muskingum Model

Combination of XGBoost and PPLK method for improving the precipitation nowcasting