Abstract. Accurate wind profile measurements are important for applications ranging from aviation to numerical weather prediction. The spatial pattern of winds can be obtained with ground-based remote sensing instruments, such as weather radars and Doppler lidars. As the return signal in weather radars is mostly due to hydrometeors or insects, and in Doppler lidars due to aerosols, the instruments provide wind measurements in different weather conditions. However, the effect of various weather conditions on the measurement capabilities of these instruments has not been previously extensively quantified. Here we present results from a 7-month measurement campaign that took place in Vantaa, Finland, where a co-located Vaisala WRS400 X-band weather radar and WindCube 400S Doppler lidar were employed continuously to perform wind measurements. Both instruments measured plan position indicator (PPI) scans at 2.0∘ elevation from the horizontal. Direct comparison of radial Doppler velocities from both instruments showed good agreement with R2=0.96. We then examined the effect of horizontal visibility, cloud base height, and precipitation intensity on the measurement availability of each instrument. The Doppler lidar displayed good availability in clear air situations and the X-band radar in precipitation. Both instruments exhibited high availability in clear air conditions in summer when insects were present. The complementary performance in the measurement availability of the two instruments means that their combination substantially increases the spatial coverage of wind observations across a wide range of weather conditions.
Nowcasts (i.e., short-term forecasts from 5 min to 6 h) of heavy rainfall are important for applications such as flash flood predictions. However, current precipitation nowcasting methods based on the extrapolation of radar echoes have a limited ability to predict the growth and decay of rainfall. While deep learning applications have recently shown improvement compared to extrapolation-based methods, they still struggle to correctly nowcast small-scale high-intensity rainfall. To address this issue, we present a novel model called the Lagrangian convolutional neural network (L-CNN) that separates the growth and decay of rainfall from motion using the advection equation. In the model, differences between consecutive rain rate fields in Lagrangian coordinates are fed into a U-Net-based CNN, known as RainNet, that was trained with the root-mean-squared-error loss function. This results in a better representation of rainfall temporal evolution compared to the RainNet and the extrapolation-based LINDA model that were used as reference models. On Finnish weather radar data, the L-CNN underestimates rainfall less than RainNet, demonstrated by greater POD (29% at 30 min at 1 mm•h −1 threshold) and smaller bias (98% at 15 min). The increased ETS values over LINDA for leadtimes under 15 min, with maximum increases of 7% (5 mm•h −1 threshold) and 10% (10 mm•h −1 ), show that the L-CNN represents the growth and decay of heavy rainfall more accurately than LINDA. This implies that nowcasting of heavy rainfall is improved when growth and decay are predicted using a deep learning model.
Abstract. We investigate the boundary-layer (BL) height at Hyytiälä in southern Finland diagnosed from radiosonde observations, a microwave radiometer (MWR) and ERA5 reanalysis. Four different, pre-existing algorithms are used to diagnose the BL height from the radiosondes. The diagnosed BL height is sensitive to the method used. The level of agreement, and the sign of systematic bias between the four different methods, depends on the surface-layer stability. For very unstable situations, the median BL height diagnosed from the radiosondes varies from 600 to 1500 m depending on which method is applied. Good agreement between the BL height in ERA5 and diagnosed from the radiosondes using Richardson-number-based methods is found for almost all stability classes, suggesting that ERA5 has adequate vertical resolution near the surface to resolve the BL structure. However, ERA5 overestimates the BL height in very stable conditions, highlighting the ongoing challenge for numerical models to correctly resolve the stable BL. Furthermore, ERA5 BL height differs most from the radiosondes at 18:00 UTC, suggesting ERA5 does not resolve the evening transition correctly. BL height estimates from the MWR are also found to be reliable in unstable situations but often are inaccurate under stable conditions when, in comparison to ERA5 BL heights, they are much deeper. The errors in the MWR BL height estimates originate from the limitations of the manufacturer's algorithm for stable conditions and also the misidentification of the type of BL. A climatology of the annual and diurnal cycle of BL height, based on ERA5 data, and surface-layer stability, based on eddy covariance observations, was created. The shallowest (353 m) monthly median BL height occurs in February and the deepest (576 m) in June. In winter there is no diurnal cycle in BL height; unstable BLs are rare, yet so are very stable BLs. The shallowest BLs occur at night in spring and summer, and very stable conditions are most common at night in the warm season. Finally, using ERA5 gridded data, we determined that the BL height observed at Hyytiälä is representative of most land areas in southern and central Finland. However, the spatial variability of the BL height is largest during daytime in summer, reducing the area over which BL height observations from Hyytiälä would be representative.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.