Enhancing Deep Learning Soil Moisture Forecasting Models by Integrating Physics-based Models

Accurate soil moisture(SM)prediction is critical for understanding hydrological processes.Physics-based(PB)models exhibit large uncertainties in SM predictions arising from uncertain parameterizations and insufficient representation of land-surface processes.In addition to PB models,deep learning(DL)models have been widely used in SM predictions recently.However,few pure DL models have notably high success rates due to lacking physical information.Thus,we developed hybrid models to effectively integrate the outputs of PB models into DL models to improve SM predictions.To this end,we first developed a hybrid model based on the attention mechanism to take advantage of PB models at each forecast time scale(attention model).We further built an ensemble model that combined the advantages of different hybrid schemes(ensemble model).We utilized SM forecasts from the Global Forecast System to enhance the convolutional long short-term memory(ConvLSTM)model for 1–16 days of SM predictions.The performances of the proposed hybrid models were investigated and compared with two existing hybrid models.The results showed that the attention model could leverage benefits of PB models and achieved the best predictability of drought events among the different hybrid models.Moreover,the ensemble model performed best among all hybrid models at all forecast time scales and different soil conditions.It is highlighted that the ensemble model outperformed the pure DL model over 79.5%of in situ stations for 16-day predictions.These findings suggest that our proposed hybrid models can adequately exploit the benefits of PB model outputs to aid DL models in making SM predictions.

燃烧源氨排放对灰霾期间北京大气铵的重要贡献

Aerosol ammonium(NH_(4)^(+)),mainly produced from the reactions of ammonia(NH_(3))with acids in the atmosphere,has significant impacts on air pollution,radiative forcing,and human health.Understanding the source and formation mechanism of NH_(4)^(+)can provide scientific insights into air quality improvements.However,the sources of NH_(3)in urban areas are not well understood,and few studies focus on NH_(3)/NH_(4)^(+)at different heights within the atmospheric boundary layer,which hinders a comprehensive understanding of aerosol NH_(4)^(+).In this study,we perform both field observation and modeling studies(the Community Multiscale Air Quality,CMAQ)to investigate regional NH_(3)emission sources and vertically resolved NH_(4)^(+)formation mechanisms during the winter in Beijing.Both stable nitrogen isotope analyses and CMAQ model suggest that combustion-related NH_(3)emissions,including fossil fuel sources,NH_(3)slip,and biomass burning,are important sources of aerosol NH_(4)^(+)with more than 60%contribution occurring on heavily polluted days.In contrast,volatilization-related NH_(3)sources(livestock breeding,N-fertilizer application,and human waste)are dominant on clean days.Combustion-related NH_(3)is mostly local from Beijing,and biomass burning is likely an important NH_(3)source(~15%–20%)that was previously overlooked.More effective control strategies such as the two-product(e.g.,reducing both SO_(2)and NH_(3))control policy should be considered to improve air quality.

Vision-based measuring method for individual cow feed intake using depth images and a Siamese network

Feed intake is an important indicator to reflect the production performance and disease risk of dairy cows,which can also evaluate the utilization rate of pasture feed.To achieve an automatic and non-contact measurement of feed intake,this paper proposes a method for measuring the feed intake of cows based on computer vision technology with a Siamese network and depth images.An automated data acquisition system was first designed to collect depth images of feed piles and constructed a dataset with 24150 samples.A deep learning model based on the Siamese network was then constructed to implement non-contact measurement of feed intake for dairy cows by training with collected data.The experimental results show that the mean absolute error(MAE)and the root mean square error(RMSE)of this method are 0.100 kg and 0.128 kg in the range of 0-8.2 kg respectively,which outperformed existing works.This work provides a new idea and technology for the intelligent measuring of dairy cow feed intake.

Mechanism of action,benefits,and research gap in fermented soybean meal utilization as a high-quality protein source for livestock and poultry

Animal nutritionists have incessantly worked towards providing livestock with high-quality plant protein feed resources.Soybean meal(SBM)has been an essential and predominantly adopted vegetable protein source in livestock feeding for a long time;however,several SBM antinutrients could potentially impair the animal's performance and growth,limiting its use.Several processing methods have been employed to remove SBM antinutrients,including fermentation with fungal or bacterial microorganisms.According to the literature,fermentation,a traditional food processing method,could improve SBM's nutritional and functional properties,making it more suitable and beneficial to livestock.The current interest in health-promoting functional feed,which can enhance the growth of animals,improve their immune system,and promote physiological benefits more than conventional feed,coupled with the ban on the use of antimicrobial growth promoters,has caused a renewed interest in the use of fermented SBM(FSBM)in livestock diets.This review details the mechanism of SBM fermentation and its impacts on animal health and discusses the recent trend in the application and emerging advantages to livestock while shedding light on the research gap that needs to be critically addressed in future studies.FSBM appears to be a multifunctional high-quality plant protein source for animals.Besides removing soybean antinutrients,beneficial bioactive peptides and digestive enzymes are produced during fermentation,providing probiotics,antioxidants,and immunomodulatory effects.Critical aspects regarding FSBM feeding to animals remain uncharted,such as the duration of fermentation,the influence of feeding on digestive tissue development,choice of microbial strain,and possible environmental impact.

Effects of applying lactic acid bacteria to the fermentation on a mixture of corn steep liquor and air-dried rice straw

This study was to determine the fermentation quality of a mixture of corn steep liquor(CSL)(178 g/kg wet basis) and air-dried rice straw(356 g/kg wet basis) after being treated with inoculants of different types of lactic acid bacteria(LAB). The treatments included the addition of no LAB additive(control),which was deionized water; homo-fermentative LAB alone(^(ho)LAB), which was Lactobacillus plantarum alone), and a mixture of homo-fermentative and hetero-fermentative LAB(^(he+ho)LAB), which were L. plantarum, Lactobacillus casei, and Lactobacillus buchneri. The results showed that the inoculation of the mixture of CSL and air-dried rice straw with ^(he+ho)LAB significantly increased the concentration of acetic acid and lactic acid compared with the control(P < 0.05). The addition of ^(he+ho)LAB effectively inhibited the growth of yeast in the silage. The concentration of total lactic acid bacteria in the ^(he+ho)LAB-treated silage was significant higher than those obtained in other groups(P < 0.05). The duration of the aerobic stability of the silages increased from 56 h to >372 h. The control group was the first to spoil, whereas the silage treated with ^(he+ho)LAB remained stable throughout the 372 h period of monitoring. The results demonstrated that the ^(he+ho)LAB could effectively improve the fermentation quality and aerobic stability of the silage.

Effect of different levels of corn steep liquor addition on fermentation characteristics and aerobic stability of fresh rice straw silage

The object of this study was to determine the proper mixing ratio of fresh rice straw to corn steep liquor(CSL) to obtain a high protein content silage feed. The following experimental silages were generated: the control(C1), composed of fresh rice straw without CSL additive, mixed with CSL in the ratios of 4:1(C4),3:1(C3) and 2:1(C2). Lactic acid bacteria(LAB) inoculant was applied at the rate of 50 mL/kg(fresh basis)of forage to achieve a final application rate of 1 x 10~6 cfu/g of fresh matter(FM). Duplicate silos for each treatment were opened after 0,3, 7,10,20,30,45 and 60 d for microbiological and chemical analysis. The results showed that the addition of CSL significantly increased crude protein(CP) contents, and decreased neutral detergent fiber(NDF) and acid detergent fiber(ADF) contents of treatments after 60 d of ensiling(P < 0.05). The lactic acid contents in C4 and C3 were significantly higher than that in C1(P <0.05). In summary, mixing fresh rice straw with CSL at addition levels of 4:1(C4) and 3:1(C3) can improve the fermentation quality and nutrient composition of fresh rice straw silage. However, a large proportion of CSL(C3) had a negative impact on the aerobic stability of fresh rice straw.

Assessment of dairy cow feed intake based on BP neural network with polynomial decay learning rate

To overcome the shortcomings of traditional dairy cow feed intake assessment model andBP neural network, this paper proposes a method of optimizing BP neural network usingpolynomial decay learning rate, taking the cow’s body weight, lying duration, lying times,walking steps, foraging duration and concentrate-roughage ratio as input variables andtaking the actual feed intake is the output variable to establish a dairy cow feed intakeassessment model, and the model is trained and verified by experimental data collectedon site. For the sake of comparative study, feed intake is simultaneously assessed by SVRmodel, KNN logistic regression model, traditional BP neural network model, and multilayerBP neural network model. The results show that the established BP model using the polynomial decay learning rate has the highest assessment accuracy, the MSPE, RMSE, MAE,MAPE and R2 are 0.043 kg2/d and 0.208 kg/d, 0.173 kg/d, 1.37% and 0.94 respectively. Compared with SVR model and KNN mode, the RMSE value reduced by 43.9% and 26.5%, it isalso found that the model designed in this paper has many advantages in comparison withthe BP model and multilayer BP model in terms of precision and generalization. Therefore,this method is ready to be applied for accurately evaluating the dairy cow feed intake, andit can provide theoretical guidance and technical support for the precise-feeding and canalso be of high significance in the improvement of dairy precise-breeding.

Automatic recognition of ingestive-related behaviors of dairy cows based on triaxial acceleration

Ingestive-related behaviors including feeding and ruminating are important indexes to measure the health and welfare of dairy cows.The purpose of this study is to develop a method based on triaxial acceleration to automatically recognize feeding and ruminating of dairy cows.During the experiment,five diary cows raised in a barn were used as experimental subjects.A triaxial acceleration sensor was used as the device to collect jawmovement data of dairy cows,and the behaviors of dairy cows were classified into three categories:feeding,ruminating and other behavior.The features of time-domain and frequency-domain were extracted from the raw acceleration data.Three machine learning algorithms including k-nearest neighbor,support vector machine and probabilistic neural network were used for the classification and the results based on four different data segment lengths were compared.The results show that the three algorithms can be used for recognition of feeding and ruminating with high accuracy.Under the condition that the sampling frequency of the sensor is 5 Hz,the combination of data segment length of 256 and k-nearest neighbor algorithm is the best scheme for recognition of feeding and ruminating in this study.The precision and recall of recognition for feeding were 92.8%and 95.6%respectively,and those of recognition for ruminating were 93.7%and 94.3%respectively.The specificity and AUC of recognition for feeding were 96.1%and 0.959 respectively,and those of recognition for ruminating were 97.5%and 0.959 respectively.This will provide an effective method for real-time monitoring of ingestive-related behaviors of dairy cows and lay a foundation for prediction of dairy cows’health status and welfare to further achieve the purpose of disease prediction and adjusting feeding and management methods.

Predicting the excretion of feces,urine and nitrogen using support vector regression:A case study with Holstein dry cows

Predicting the excretion of feces,urine and nitrogen(N)from dairy cows is an effective way to prevent and control the environmental pollution caused by scaled farming.The traditional prediction methods such as pollutant generation coefficient(PGC)and mathematical model based on linear regression(LR)may be limited by prediction range and regression function assumption,and sometimes may deviate from the actual condition.In order to solve these problems,the support vector regression(SVR)was applied for predicting the cows'feces,urine and N excretions,taking Holstein dry cows as a case study.SVR is a typical non-parametric machine learning model that does not require any specific assumptions about the regression function in advance and only by learning the training sample data,and also it can fit the function closest to the actual in most cases.To evaluate prediction accuracy effectively,the SVR technique was compared with the LR and radial basis function artificial neural network(RBF-ANN)methods,using the required sample data obtained from actual feeding experiments.The prediction results indicate that the proposed technique is superior to the other two conventional(especially LR)methods in predicting the main indicators of feces,urine,and N excretions of Holstein dry cows.