Abstract. GPS radio occultation (RO) has been recognised as an alternative atmospheric upper air observation technique due to its distinct features and technological merits. The CHAllenging Minisatellite Payload (CHAMP) RO satellite and FORMOSAT-3/COSMIC (Constellation Observing System for Meteorology, Ionosphere and Climate) RO constellation together have provided about ten years of high quality global coverage RO atmospheric profiles. This technique is best used for meteorological studies in the difficult-toaccess areas such as deserts and oceans. To better understand and use RO data, effective quality assessment using independent radiosonde data and its associated collocation criteria used in tempo-spatial domain are important. This study compares GPS RO retrieved temperature profiles from both CHAMP (between May 2001 and October 2008) and FORMOSAT-3/COSMIC (between July 2006 and December 2009) with radiosonde data from 38 Australian radiosonde stations. The overall results show a good agreement between the two data sets. Different collocation criteria within 3 h and 300 km between the profile pairs have been applied and the impact of these different collocation criteria on the evaluation results is found statistically insignificantly. The CHAMP and FORMOSAT-3/COSMIC temperature profiles have been evaluated at 16 different pressure levels and the differences between GPS RO and radiosonde at different levels of the atmosphere have been studied. The result shows that the mean temperature difference between radiosonde and CHAMP is 0.39 • C (with a standard deviation of 1.20 • C) and the one Correspondence to: E. Fu (f.fu@bom.gov.au) between radiosonde and FORMOSAT-3/COSMIC is 0.37 • C (with a standard deviation of 1.24 • C). Different collocation criteria have been applied and insignificant differences were identified amongst the results.

The leaf area index (LAI) is of great significance for crop growth monitoring. Recently, unmanned aerial systems (UASs) have experienced rapid development and can provide critical data support for crop LAI monitoring. This study investigates the effects of combining spectral and texture features extracted from UAS multispectral imagery on maize LAI estimation. Multispectral images and in situ maize LAI were collected from test sites in Tongshan, Xuzhou, Jiangsu Province, China. The spectral and texture features of UAS multispectral remote sensing images are extracted using the vegetation indices (VIs) and the gray-level co-occurrence matrix (GLCM), respectively. Normalized texture indices (NDTIs), ratio texture indices (RTIs), and difference texture indices (DTIs) are calculated using two GLCM-based textures to express the influence of two different texture features on LAI monitoring at the same time. The remote sensing features are prescreened through correlation analysis. Different data dimensionality reduction or feature selection methods, including stepwise selection (ST), principal component analysis (PCA), and ST combined with PCA (ST_PCA), are coupled with support vector regression (SVR), random forest (RF), and multiple linear regression (MLR) to build the maize LAI estimation models. The results reveal that ST_PCA coupled with SVR has better performance, in terms of the VIs + DTIs (R2 = 0.876, RMSE = 0.239) and VIs + NDTIs (R2 = 0.877, RMSE = 0.236). This study introduces the potential of different texture indices for maize LAI monitoring and demonstrates the promising solution of using ST_PCA to realize the combining of spectral and texture features for improving the estimation accuracy of maize LAI.

No abstract

The convolutional neural network (CNN) method has been widely used in the classification of hyperspectral images (HSIs). However, the efficiency and accuracy of the HSI classification are inevitably degraded when small samples are available. This study proposes a multidimensional CNN model named MDAN, which is constructed with an attention mechanism, to achieve an ideal classification performance of CNN within the framework of few-shot learning. In this model, a three-dimensional (3D) convolutional layer is carried out for obtaining spatial–spectral features from the 3D volumetric data of HSI. Subsequently, the two-dimensional (2D) and one-dimensional (1D) convolutional layers further learn spatial and spectral features efficiently at an abstract level. Based on the most widely used convolutional block attention module (CBAM), this study investigates a convolutional block self-attention module (CBSM) to improve accuracy by changing the connection ways of attention blocks. The CBSM model is used with the 2D convolutional layer for better performance of HSI classification purposes. The MDAN model is applied for classification applications using HSI, and its performance is evaluated by comparing the results with the support vector machine (SVM), 2D CNN, 3D CNN, 3D–2D–1D CNN, and CBAM. The findings of this study indicate that classification results from the MADN model show overall classification accuracies of 97.34%, 96.43%, and 92.23% for Salinas, WHU-Hi-HanChuan, and Pavia University datasets, respectively, when only 1% HSI data were used for training. The training and testing times of the MDAN model are close to those of the 3D–2D–1D CNN, which has the highest efficiency among all comparative CNN models. The attention model CBSM is introduced into MDAN, which achieves an overall accuracy of about 1% higher than that of the CBAM model. The performance of the two proposed methods is superior to the other models in terms of both efficiency and accuracy. The results show that the combination of multidimensional CNNs and attention mechanisms has the best ability for small-sample problems in HSI classification.