No abstract

Details of the kinematics of free flight are very important to understanding insect flight mechanics. Important data for aerodynamic analysis and modeling include flight trajectory, body attitude and wing kinematics for individuals flying over a diverse array of behaviors, such as hovering, climbing and turning.A number of recent studies have focused on the kinematics of hovering and forward flight, using a variety of techniques. Azuma and Watanabe (1988) changed the velocity of the wind tunnel in their measurements. Dudley and Ellington (1990) calculated angles of attack in the free forward flight of bumblebees. Willmott and Ellington (1997) employed a variable-speed wind tunnel associated with the optomotor response to investigate wing and body kinematics during free forward flight of a hawkmoth over a range of speeds from hovering to 5 m s -1 . Wakeling and Ellington (1997) filmed the free flights of dragonflies and damselflies flying over the pond in the greenhouse at the University of Cambridge. The individuals were not restrained by either tethers or wind tunnels, but were free to vary the velocity and acceleration and could perform any flight action. In their analyses of forward flight, the stroke plane was constructed based on the assumption of bilateral wing symmetry, and variations in roll, yaw and pitch angles of the body through each flapping cycle were neglected. To date no detailed information on wing orientation or shape during free flight has been acquired.All kinds of flight behaviors are important for studying the aerodynamics and the control of flight. In turning maneuvers, the wings move asymmetrically, and the change in attitude is obvious even during one flapping cycle. We have also found that dragonflies exhibit substantial chordwise deformation and changes in camber during free flight, which might be important for aerodynamic models of flight performance.To study turning maneuvers involving obvious changing of the insect attitude, the description of wing kinematics should be based on a local body-centered coordinate system, together with the body attitude and flight trajectory. We have developed a method utilizing a Projected, Comb-Fringe technique combined with the Landmarks procedure (PCFL), in which a comb-fringe pattern with high intensity and sharpness was projected onto the transparent wing of a dragonfly in free flight. Images of the wings with distorted fringes were then recorded by a high-speed camera. Based on the distorted fringe pattern and the natural landmarks on the dragonfly wings, we reconstructed wing shape and established the body-centered coordinate system. This method allowed us to derive kinematic parameters without assumptions of rigid chords or kinematic symmetry, except for the assumption of rigid leading edges. The instantaneous attitude of the body was also measured simultaneously. We measured dragonflies in two flight behaviors: forward flight and turning maneuvers, and compared the kinematics results obtained for each of them. A robust technique for determinin...

No abstract

Precipitation downscaling is widely employed for enhancing the resolution and accuracy of precipitation products from general circulation models (GCMs). In this study, we propose a novel statistical downscaling method to foster GCMs’ precipitation prediction resolution and accuracy for the monsoon region. We develop a deep neural network composed of a convolution and Long Short Term Memory (LSTM) recurrent module to estimate precipitation based on well-resolved atmospheric dynamical fields. The proposed model is compared against the GCM precipitation product and classical downscaling methods in the Xiangjiang River Basin in South China. Results show considerable improvement compared to the European Centre for Medium-Range Weather Forecasts (ECMWF)-Interim reanalysis precipitation. Also, the model outperforms benchmark downscaling approaches, including (1) quantile mapping, (2) the support vector machine, and (3) the convolutional neural network. To test the robustness of the model and its applicability in practical forecasting, we apply the trained network for precipitation prediction forced by retrospective forecasts from the ECMWF model. Compared to the ECMWF precipitation forecast, our model makes better use of the resolved dynamical field for more accurate precipitation prediction at lead times from 1 day up to 2 weeks. This superiority decreases with the forecast lead time, as the GCM’s skill in predicting atmospheric dynamics is diminished by the chaotic effect. Finally, we build a distributed hydrological model and force it with different sources of precipitation inputs. Hydrological simulation forced with the neural network precipitation estimation shows significant advantage over simulation forced with the original ERA-Interim precipitation (with NSE value increases from 0.06 to 0.64), and the performance is only slightly worse than the observed precipitation forced simulation (NSE = 0.82). This further proves the value of the proposed downscaling method, and suggests its potential for hydrological forecasts.