Search

The atmospheric low-level vortex is an important weather system that causes rainstorm and flood in southern China. The vortex induced rainstorm is sudden and local, and its occurence time, intensity and falling area have always been difficult for meteorological forecasting. In order to improve the accuracy of rainstorm prediction in Zhejiang Province and improve the localization application of the prediction model, based on daily precipitation data of Zhejiang Province from 1979 to 2020 and reanalysis dataset from the National Centers for Environmental Prediction Climate Forecast System, climatic characteristics of the heavy rain in Zhejiang Province induced by low-level vortex is studied, the vortex induced rainstorm in Zhejiang mostly occurs in the rainy season and has obvious interannual variation characteristics. Then, the 850 hPa vertical velocity, water vapor flux divergence, velocity divergence, low-level jet, shear line and 700 hPa vertical velocity are used as ingredient factors, and the potential falling area forecast for heavy rain induced by vortex is constructed by using logistic regression algorithm. Finally, the forecast products of the European Medium-Range Weather Prediction Center are used to forecast the rainstorm falling area based on this model, and the actual data of the same period are used for systematic test and weather process test. Results show that the overall performance of the model is good, and the rainstorm prediction ability of 24 h and 12 h ahead is improved, which can increase from 12% accuracy and 20% to around 14% and 40% respectively, and the omission rate drops from more than 75% to around 60%, but there is a certain degree of empty report. The forecast model of Zhejiang Province based on logistic regression method has a good indicator for the prediction of such rainstorm areas, and is of great significance for meteorological disaster prevention and mitigation.

To understand and improve the performance and quality of Global Precipitation Measurement (GPM) satellite-derived precipitation products during the typhoon process, with gauge observations from ground meteorological stations, this study used the successive and optimal interpolation correction methods to correct the GPM precipitation products. The performance of GPM products (Version 6) before and after the correction at different time scales are evaluated with the products during the two landfalls of Typhon"In-Fa". Results show that the original GPM precipitation products fail to describe the strong rainfall center over the high terrain area, which significantly underestimates the precipitation by approximately 45%. This is mainly due to the underestimation of the large values in the northeast region of Zhejiang province. The analysis of precipitation classification inspection suggests that the uncorrected GPM satellite-derived products have good capability for light rain, but poor for other rainfall levels. After the successive and optimal interpolation correction, significant improvement can be found in the cumulative, daily, and hourly rainfall products. The products with successive correction are better, while the product with optimal interpolation slightly overestimates the actual precipitation. The corrected GPM products can not only describe the heavy precipitation center related to the terrain but also reflect heavy rainstorms and heavy precipitation with hourly rainfall intensity greater than 23 mm·h-1.

Powdery mildew significantly impacts the yield of natural rubber by being one of the predominant diseases that affect rubber trees. Accurate, non-destructive recognition of powdery mildew in the early stage is essential for the cultivation management of rubber trees. The objective of this study is to establish a technique for the early detection of powdery mildew in rubber trees by combining spectral and physicochemical parameter features. At three field experiment sites and in the laboratory, a spectroradiometer and a hand-held optical leaf-clip meter were utilized, respectively, to measure the hyperspectral reflectance data (350–2500 nm) and physicochemical parameter data of both healthy and early-stage powdery-mildew-infected leaves. Initially, vegetation indices were extracted from hyperspectral reflectance data, and wavelet energy coefficients were obtained through continuous wavelet transform (CWT). Subsequently, significant vegetation indices (VIs) were selected using the ReliefF algorithm, and the optimal wavelengths (OWs) were chosen via competitive adaptive reweighted sampling. Principal component analysis was used for the dimensionality reduction of significant wavelet energy coefficients, resulting in wavelet features (WFs). To evaluate the detection capability of the aforementioned features, the three spectral features extracted above, along with their combinations with physicochemical parameter features (PFs) (VIs + PFs, OWs + PFs, WFs + PFs), were used to construct six classes of features. In turn, these features were input into support vector machine (SVM), random forest (RF), and logistic regression (LR), respectively, to build early detection models for powdery mildew in rubber trees. The results revealed that models based on WFs perform well, markedly outperforming those constructed using VIs and OWs as inputs. Moreover, models incorporating combined features surpass those relying on single features, with an overall accuracy (OA) improvement of over 1.9% and an increase in F1-Score of over 0.012. The model that combines WFs and PFs shows superior performance over all the other models, achieving OAs of 94.3%, 90.6%, and 93.4%, and F1-Scores of 0.952, 0.917, and 0.941 on SVM, RF, and LR, respectively. Compared to using WFs alone, the OAs improved by 1.9%, 2.8%, and 1.9%, and the F1-Scores increased by 0.017, 0.017, and 0.016, respectively. This study showcases the viability of early detection of powdery mildew in rubber trees.

Powdery mildew is one of the most significant rubber tree diseases, with a substantial impact on the yield of natural rubber. This study aims to establish a detection approach that coupled continuous wavelet transform (CWT) and machine learning for the accurate assessment of powdery mildew severity in rubber trees. In this study, hyperspectral reflectance data (350–2500 nm) of healthy and powdery mildew-infected leaves were measured with a spectroradiometer in a laboratory. Subsequently, three types of wavelet features (WFs) were extracted using CWT. They were as follows: WFs dimensionally reduced by the principal component analysis (PCA) of significant wavelet energy coefficients (PCA-WFs); WFs extracted from the top 1% of the determination coefficient between wavelet energy coefficients and the powdery mildew disease class (1%R2-WFs); and all WFs at a single decomposition scale (SS-WFs). To assess the detection capability of the WFs, the three types of WFs were input into the random forest (RF), support vector machine (SVM), and back propagation neural network (BPNN), respectively. As a control, 13 optimal traditional spectral features (SFs) were extracted and combined with the same classification methods. The results revealed that the WF-based models all performed well and outperformed those based on SFs. The models constructed based on PCA-WFs had a higher accuracy and more stable performance than other models. The model combined PCA-WFs with RF exhibited the optimal performance among all models, with an overall accuracy (OA) of 92.0% and a kappa coefficient of 0.90. This study demonstrates the feasibility of combining CWT with machine learning in rubber tree powdery mildew detection.

A group of control (CTL) plus ten sensitivity numerical experiments have been conducted to investigate the sensitivity of planetary boundary layer (PBL) schemes in simulating the asymmetric precipitation distribution of typhoon Lekima (2019) during landfall. The simulated track and intensity are quite sensitive to the choice of the PBL scheme. In CTL that applies the Mellor–Yamada Nakanishi and Niino (MYNN) PBL and the surface layer scheme, the observed eyewall replacement 6 h prior to landfall and the asymmetric precipitation during landfall have been simulated well. However, in the PBL1 experiment that applies the Yonsei University (YSU) PBL scheme and the Revised Mesoscale Model version 5 (MM5) Monin–Obukhov surface layer scheme, no double eyewall is simulated. PBL1 and the other sensitivity experiments also simulate more axisymmetric precipitation distribution. PBL1 simulates intensification just before landfall, sustains intensity longer after landfall, but then dissipates quite rapidly. Such differences from CTL are due to larger enthalpy flux, higher PBL height (which is almost unchanged in CTL), and eddy diffusivity extending more into the free atmosphere in PBL1. These factors lead to outward expansion of the radius of maximum wind, larger radial inflow, larger axisymmetric tangential wind in the boundary layer, and larger updrafts in the eyewall. After landfall, larger momentum flux and larger friction velocity in PBL1 enable the more rapid dissipation. The intensification before landfall in PBL1 makes the axisymmetric component stronger. Asymmetry developed in the outer eyewall, and PBL1 was less successful in simulating the eyewall replacement that affects the degree of rainfall asymmetry. These results indicate that the model PBL schemes largely influence the simulated tropical cyclone (TC) intensity and structure including asymmetric rainfall distribution during landfall.

Post-harvest wet coffee processing is a commonly applied method to transform coffee cherries into green coffee beans through depulping or demucilaging, fermentation, washing, soaking, drying, and dehulling. Multiple processing parameters can be modified and thus influence the coffee quality (green coffee beans and cup quality). The present study aimed to explore the impacts of these parameters, including processing type (depulping or demucilaging), fermentation duration, and application of soaking, on the microbial community dynamics, metabolite compositions of processing waters (fermentation and soaking) and coffee beans, and resulting cup quality through a multiphasic approach. A large-scale wet coffee processing experiment was conducted with Coffea arabica var. Catimor in Yunnan (China) in duplicate. The fermentation steps presented a dynamic interaction between constant nutrient release (mainly from the cherry mucilage) into the surrounding water and active microbial activities led by lactic acid bacteria, especially Leuconostoc and Lactococcus. The microbial communities were affected by both the processing type and fermentation duration. At the same time, the endogenous coffee bean metabolism remained active at different stages along the processing, as could be seen through changes in the concentrations of carbohydrates, organic acids, and free amino acids. Among all the processing variants tested, the fermentation duration had the greatest impact on the green coffee bean compositions and the cup quality. A long fermentation duration resulted in a fruitier and more acidic cup. As an ecological alternative for the depulped processing, the demucilaged processing produced a beverage quality comparable to the depulped one. The application of soaking, however, tempered the positive fermentation effects and standardized the green coffee bean quality, regardless of the preceding processing practices applied. Lastly, the impact strength of each processing parameter would also depend on the coffee variety used and the local geographical conditions. All these findings provide a considerable margin of opportunities for future coffee research.

A high-resolution atmospheric model of the Weather Research and Forecast (WRF) is used to investigate the climatic effects of mesoscale oceanic eddies (OEs) in the North Pacific (NPac) in spring and the respective effects of OEs in the northern NPac associated with the Kuroshio Extension (KE) and of OEs in the southern NPac related to the subtropical countercurrent. Results show that mesoscale OEs in the NPac can strengthen the upper-level ridge (trough) in the central (eastern) subtropical NPac, together with markedly weakened (strengthened) westerly winds to its south. The mesoscale OEs in the whole NPac act to weaken the upper-level storm track and strengthen lower-level storm activities in the NPac. However, atmospheric responses to the northern and southern NPac OEs are more prominent. The northern NPac OEs can induce tropospheric barotropic responses with a tripole geopotential height (GPH) anomaly pattern to the north of 30° N, while the OEs in both the northern and southern NPac can enhance the upper-level ridge (trough) in the central (eastern) subtropical NPac. Additionally, the northern NPac OEs can shrink the lower-level subtropical high and weaken the easterly trade winds at the low latitudes, while the southern NPac OEs result in a southward shift of the lower-level subtropical high and an eastward shift of the upper-level westerly jet stream. The southern and northern NPac OEs have similar effects on the storm track, leading to an enhanced lower-level storm track over the KE via moistening the atmospheric boundary layer; and they can also exert significant remote influences on lower- and upper-level storm activities over the Northeast Pacific off the west coast of North America. When the intensities of OEs are doubled in the model, the spatial distribution of atmospheric responses is robust, with a larger and more significant magnitude. Additionally, although OEs are part of the mesoscale oceanic processes, the springtime OEs play an opposite role in mesoscale sea-surface temperature anomalies. These findings point to the potential of improving the forecasts of extratropical springtime storm systems and the projections of their responses to future climate change, by improving the representation of ocean eddy-atmosphere interaction in forecast and climate models.