Investigating the response of soil and vegetable crops to poultry and cow manure using ground and satellite data

Based on the massive production of cow and poultry manures, farmers in Saudi Arabia are moving towards the application of organic fertilizers in their farms. Therefore, the present work was conducted to study the response of soil and selected vegetable crops to poultry and cow manures, using ground data and Landsat-8 and Hyperion images. The studied vegetable crops are cabbage, cauliflower, broccoli, and lettuce. A total of 100 t ha^?1 organic manures were applied as a pre-planting treatment. A 12.5 ha field in Tawdeehiya Farms, 200 km southeast of Riyadh, was earmarked for this study. The field was divided into sectors cultivated with the above-mentioned vegetable crops. Soil characteristics, including the soil pH, the electric conductivity (EC), the nitrogen (N), the phosphorus (P) and the potassium (K), were examined before the application of manures and 25 days after the transplanting process. Observations on crops chlorophyll content, number of leaves, the diameter of merchantable products and yield were also investigated. Furthermore, the relationship between the crop performance and yield was investigated through the satellite images generated vegetation indices (VIs). This study revealed the better performance of poultry manure compared to cow manure in terms of development and production parameters of the experimental crops. Dynamics of the chlorophyll content across the crop growth period revealed that all the tested crops responded significantly (R² = 0.69; P = 0.001) to the poultry manure treatments. Among the tested crops, the chlorophyll content, curd or head sizes and crop yields were quite better in poultry manure applied plots. The investigation of crop yield was significant with poultry manure (R² = 0.64; P = 0.001) than cow manure (R² = 0.57; P = 0.001) using the OSAVI and mNDVI, respectively.     

Biological nitrogen fixation in resource-poor agriculture in South Africa

Farm lands of resource-poor communities in South Africa are depleted of nutrients due to continuous mono-cropping, limited use of fertilisers, and sometimes leaching caused by high rainfall. Despite the well-known advantages of biological nitrogen fixation (BNF) in cropping systems, less than 10% of the grain crops planted annually in these areas are legumes. Using a participatory research and development approach, resource-poor farmers were introduced to conservation agriculture (CA) practices, including BNF, that promoted zero (or reduced) tillage, increased retention of soil cover, as well as crop diversification. Because crop rotation and intercropping of legumes with cereals are known to contribute to soil fertility while enhancing food security, resource-poor farmers from various Provinces in South Africa were trained on the benefits of legume culture for eight years. As a result, these resource-poor farmers did not only get training in inoculation techniques, but were also supplied with inoculants for use on their farms. Data collected from Farmers Demonstration Trials at Belvedere, Dumbarton and Lusikisiki, showed that the grain and fodder yield of maize planted after legumes, and maize intercropped with legumes, were comparable to those of maize receiving high N fertilizer dose (i.e. 54 kg N at planting and 54 kg N as top-dressing). The same data further showed thatRhizobium inoculation, when combined with application of low levels of P and K, significantly increased crop yields within farmers’ trial plots. BNF therefore offers a great opportunity for resource-poor farmers in South Africa to increase their crop yields and thus improve the quality of their livelihoods through the adoption of affordable and sustainable biological technologies that enhance soil fertility.     

Effect of deep and shallow root systems on the dynamics of soil inorganic N during 3-year crop rotations

Unused inorganic nitrogen (N_inorg) left in agricultural soils will typically leach to deeper soil layers. If it moves below the root zone it will be lost from the system, but the depth of the root zone depends on the crop species grown. In this experiment we studied the effect of 3-year crop sequences, with different combinations of deep-rooted and shallow-rooted crops, on soil N_inorg dynamics to 2.5 m soil depth and the possibility of crop utilization of N leached to deep soil layers. We grew ten different crop sequences for 3 years. The crops and catch crops grown were selected to allow different sequences of deep-rooted and shallow-rooted crops. Very different rooting depths were obtained, from only 0.5 m (leek), to ∼1.0 m (ryegrass and barley), 1.5 m (red beet), 2.0 m (fodder radish and white cabbage) and more than 2.5 m by the chicory catch crop. The results showed a significant retention of N_inorg within the 2.5 m soil profile from one year to the next, but the retained N had leached to deeper parts of the profile during the winter season. Only little N_inorg was retained over two winter seasons. The retention in the deeper soil layers allowed N_inorg to be taken up by succeeding deep-rooted main crops or catch crops. The effects of crop rooting depth on N_inorg in the subsoil layers from 1.0 to 2.5 m were striking. White cabbage reduced N_inorg below 1.0 m with up to 113 kg N ha^-1 during its growth. Grown after catch crops, leek and red beet left on average 60 kg N ha⁻¹ less below 1.0 m than leek and red beet grown without a preceding catch crop. We conclude that it is possible to design crop rotations with improved nitrogen use efficiency by using the differences in crop rooting patterns; deep-rooted crops or catch crops can be used to recover N_inorg leached after previous crops, and catch crops can be grown before shallow-rooted crops to lift the deep N_inorg up to layers where these crops have their roots.     

Soil and crop management strategies to prevent iron deficiency in crops

Plants and humans cannot easily acquire iron from their nutrient sources although it is abundant in nature. Thus, iron deficiency is one of the major limiting factors affecting crop yields, food quality and human nutrition. Therefore, approaches need to be developed to increase Fe uptake by roots, transfer to edible plant portions and absorption by humans from plant food sources. Integrated strategies for soil and crop management are attractive not only for improving growing conditions for crops but also for exploiting a plant??s potential for Fe mobilization and utilization. Recent research progress in soil and crop management has provided the means to resolve complex plant Fe nutritional problems through manipulating the rhizosphere (e.g., rhizosphere fertilization and water regulation), and crop management (includes managing cropping systems and screening for Fe efficient species and varieties). Some simple and effective soil management practices, termed ??rhizosphere fertilization?? (such as root feeding and bag fertilization) have been developed and widely used by local farmers in China to improve the Fe nutrition of fruit plants. Production practices for rice cultivation are shifting from paddy-rice to aerobic rice to make more efficient use of irrigation water. This shift has brought about increases in Fe deficiency in rice, a new challenge depressing iron availability in rice and reducing Fe supplies to humans. Current crop management strategies addressing Fe deficiency include Fe foliar application, trunk injection, plant breeding for enriched Fe crop species and varieties, and selection of cropping systems. Managing cropping systems, such as intercropping strategies may have numerous advantages in terms of increasing Fe availability to plants. Studies of intercropping systems on peanut/maize, wheat/chickpea and guava/sorghum or -maize increased Fe content of crops and their seed, which suggests that a reasonable intercropping system of iron-efficient species could prevent or mitigate Fe deficiency in Fe-inefficient plants. This review provides a comprehensive comparison of the strategies that have been developed to address Fe deficiency and discusses the most recent advance in soil and crop management to improve the Fe nutrition of crops. These proofs of concept studies will serve as the basis for future Fe research and for integrated and optimized management strategies to alleviate Fe deficiency in farmers?? fields.     

Effect of crop and soil management practices on soil compatibility in maize and groundnut plots in a Paleustult in Southeastern Nigeria

Frequent occurrences of soil compaction damage resulting from high raindrop impact energy, and from human and animal trafficking during field operations pose a problem to farmers around the tropics. We studied the effect of some crop and soil management practices (manure, mulch, NPK applications, tillage and crop type) on some soil compactibility indices (dry bulk density, cone index, total soil porosity, gravimetric soil water content) in a Typic Paleustult in southeastern Nigeria. The study was carried out for three consecutive planting seasons using two tillage systems and four other soil management practices (poultry droppings + NPK, mulch + NPK, NPK alone and no amendment). These were laid out as split-plot in a RCB design replicated three times and using maize (Zea mays L.) and groundnut (Arachis hypogea) as test crops. Results indicate that the different soil management techniques adopted influenced dry bulk density, penetration resistance, total soil porosity and gravimetric soil water content at 44 and 66 days after planting (DAP) whereas only gravimetric soil water content was affected at 90 DAP. The dry bulk density of tilled maize and groundnut plots increased significantly (P<0.05) by between 2 and 14% relative to no-till plots at 44 and 66 DAP. In both maize and groundnut plots, dry bulk density decreased significantly (P<0.05) in plots amended with poultry droppings +NPK relative to the control plots by 3–10% at 44 and 66 DAP. Tilled maize and groundnut plots had 37–45% lower (P< 0.05) penetration resistance than their corresponding no-till plots at both 44 and 66 DAP. Penetration resistance measurements were lower by 16.5–25% in plots amended with poultry droppings + NPK relative to unamended plots at 44 and 66 DAP. Cumulative (1996, 1997, 1998) data indicate that gravimetric soil water content in maize and groundnut plots generally increased significantly (P<0.05) in no-till plots relative to tilled plots by 18–27% at both 44 and 66 DAP. Plots amended with poultry droppings + NPK had between 24 and 111% increase (P<0.05) in soil gravimetric soil water content at both 44 and 66 DAP. Results are indicative that all soil compactibility indices measured were not affected at 90 DAP except for soil gravimetric soil water content in 1996 and 1998. Results from this work demonstrate that some crop and soil management practices could be used to reduce soil compactibility problems thus increasing productivity of such soils.     

Gene Features Selection for Three-Class Disease Classification via Multiple Orthogonal Partial Least Square Discriminant Analysis and S-Plot Using Microarray Data

Motivation

DNA microarray analysis is characterized by obtaining a large number of gene variables from a small number of observations. Cluster analysis is widely used to analyze DNA microarray data to make classification and diagnosis of disease. Because there are so many irrelevant and insignificant genes in a dataset, a feature selection approach must be employed in data analysis. The performance of cluster analysis of this high-throughput data depends on whether the feature selection approach chooses the most relevant genes associated with disease classes.

Results

Here we proposed a new method using multiple Orthogonal Partial Least Squares-Discriminant Analysis (mOPLS-DA) models and S-plots to select the most relevant genes to conduct three-class disease classification and prediction. We tested our method using Golub’s leukemia microarray data. For three classes with subtypes, we proposed hierarchical orthogonal partial least squares-discriminant analysis (OPLS-DA) models and S-plots to select features for two main classes and their subtypes. For three classes in parallel, we employed three OPLS-DA models and S-plots to choose marker genes for each class. The power of feature selection to classify and predict three-class disease was evaluated using cluster analysis. Further, the general performance of our method was tested using four public datasets and compared with those of four other feature selection methods. The results revealed that our method effectively selected the most relevant features for disease classification and prediction, and its performance was better than that of the other methods.     

From forest to farm: systematic review of cultivar feeding by chimpanzees--management implications for wildlife in anthropogenic landscapes

Crop-raiding is a major source of conflict between people and wildlife globally, impacting local livelihoods and impeding conservation. Conflict mitigation strategies that target problematic wildlife behaviours such as crop-raiding are notoriously difficult to develop for large-bodied, cognitively complex species. Many crop-raiders are generalist feeders. In more ecologically specialised species crop-type selection is not random and evidence-based management requires a good understanding of species' ecology and crop feeding habits. Comprehensive species-wide studies of crop consumption by endangered wildlife are lacking but are important for managing human-wildlife conflict. We conducted a comprehensive literature search of crop feeding records by wild chimpanzees (Pan troglodytes), a ripe-fruit specialist. We assessed quantitatively patterns of crop selection in relation to species-specific feeding behaviour, agricultural exposure, and crop availability. Crop consumption by chimpanzees is widespread in tropical Africa. Chimpanzees were recorded to eat a considerable range of cultivars (51 plant parts from 36 species). Crop part selection reflected a species-typical preference for fruit. Crops widely distributed in chimpanzee range countries were eaten at more sites than sparsely distributed crops. We identified 'high' and 'low' conflict crops according to their attractiveness to chimpanzees, taking account of their importance as cash crops and/or staple foods to people. Most (86%) high conflict crops were fruits, compared to 13% of low conflict crops. Some widely farmed cash or staple crops were seldom or never eaten by chimpanzees. Information about which crops are most frequently consumed and which are ignored has enormous potential for aiding on-the-ground stakeholders (i.e. farmers, wildlife managers, and conservation and agricultural extension practitioners) develop sustainable wildlife management schemes for ecologically specialised and protected species in anthropogenic habitats. However, the economic and subsistence needs of local people, and the crop-raiding behaviour of sympatric wildlife, must be considered when assessing suitability of particular crops for conflict prevention and mitigation.     

冬季作物对稻田土壤微生物量碳、氮和微生物熵的短期影响

研究不同的冬季作物马铃薯、黑麦草、紫云英、油菜在"冬季作物-双季稻"轮作种植制度下短期内对稻田土壤微生物碳、氮和微生物熵的影响,在湖南省土壤肥料研究所的实验网室内设置了小区试验.试验结果表明:几种冬季作物均提高了稻田土壤微生物碳、氮含量,黑麦草明显提高了土壤微生物量碳和微生物熵,紫云英明显提高了土壤微生物量氮.冬季作物对土壤微生物量碳和土壤微生物量氮的季节性影响变化趋势基本一致,紫云英、马铃薯处理的土壤微生物量C、N含量均在水稻生育期间8月中旬达到最大值.     

Some measures of reducing leaching loss of nitrates beyond potential rooting zone

Summary Distribution of nitrates in soil profiles under eight different crop rotations was studied after the harvest of various constituent crops. Nitrate distribution patterns at different dates reveal that maximum leaching loss of nitrates occurs from rotations consisting of heavily fertilized shallow rooted crops like potato. Wheat and maize, in a rotation reduce nitrate leaching to deeper soil layers because of their deep and extensive rooting systems. In rainy season, when maximum movement of nitrates occurs in the profile, raising of deep rooted crops like maize leaves for leaching only a small amount of nitrates as compared to crops like groundnut and soybean with shallow rooting systems and low N-requirement. Summer season crops like moong, cowpeas and maize (fodder), do not alter nitrate distribution patterns, unless a large amount of nitrates is present in the profile prior to their sowing.Post-graduate student and Professor of Soils, respectivelyPost-graduate student and Professor of Soils, respectively     

Assessing long-term impacts of cover crops on soil organic carbon in the central US Midwestern agroecosystems

Cover crops have been reported as one of the most effective practices to increase soil organic carbon (SOC) for agroecosystems. Impacts of cover crops on SOC change vary depending on soil properties, climate, and management practices, but it remains unclear how these control factors affect SOC benefits from cover crops, as well as which management practices can maximize SOC benefits. To address these questions, we used an advanced process-based agroecosystem model, ecosys, to assess the impacts of winter cover cropping on SOC accumulation under different environmental and management conditions. We aimed to answer the following questions: (1) To what extent do cover crops benefit SOC accumulation, and how do SOC benefits from cover crops vary with different factors (i.e., initial soil properties, cover crop types, climate during the cover crop growth period, and cover crop planting and terminating time)? (2) How can we enhance SOC benefits from cover crops under different cover crop management options? Specifically, we first calibrated and validated the ecosys model at two long-term field experiment sites with SOC measurements in Illinois. We then applied the ecosys model to six cover crop field experiment sites spanning across Illinois to assess the impacts of different factors on SOC accumulation. Our modeling results revealed the following findings: (1) Growing cover crops can bring SOC benefits by 0.33 ± 0.06 MgC ha⁻¹ year⁻¹ in six cover crop field experiment sites across Illinois, and the SOC benefits are species specific to legume and non-legume cover crops. (2) Initial SOC stocks and clay contents had overall small influences on SOC benefits from cover crops. During the cover crop growth period (i.e., winter and spring in the US Midwest), high temperature increased SOC benefits from cover crops, while the impacts from larger precipitation on SOC benefits varied field by field. (3) The SOC benefits from cover crops can be maximized by optimizing cover crop management practices (e.g., selecting cover crop types and controlling cover crop growth period) for the US Midwestern maize–soybean rotation system. Finally, we discussed the economic and policy implications of adopting cover crops in the US Midwest, including that current economic incentives to grow cover crops may not be sufficient to cover costs. This study systematically assessed cover crop impacts for SOC change in the US Midwest context, while also demonstrating that the ecosys model, with rigorous validation using field experiment data, can be an effective tool to guide the adaptive management of cover crops and quantify SOC benefits from cover crops. The study thus provides practical tools and insights for practitioners and policy-makers to design cover crop related government agricultural policies and incentive programs for farmers and agri-food related industries.     

Soil mercury accumulation and transference to different crop grains

Mercury contamination in agro-ecosystems is one of the most important environmental issues. Farmland soil mercury accumulation and transference to crops in Changshu City, eastern China, were investigated to identify mercury migration capacity from soil to crops. The mean content of mercury for soil samples slightly increased year after year. The mercury accumulation capacity of rice grown (bioaccumulation factor (BAF) 0.028) in submerged soils under reductive conditions was stronger than that of wheat (BAF 0.0073) in dried soils under oxidative conditions. There were clear relationships between soil mercury with organic matter (OM), cation exchange capacity (CEC), and CaCO₃ of soil samples, while apparent negative relationships between Hg in rice grain with OM, CEC, and CaCO₃ of soil existed. No clear association for Hg between crops and soil was found, indicating that mercury in crop grains is mostly affected by other factors besides soil mercury. Also, soil properties and farming patterns affected mercury transference from soil to crop grains and mercury enrichment capacity in crop grains. The results suggested that appropriate selection of crop species and water management are two major possible ways to reduce total mercury accumulation in crop grains grown in mercury-contaminated regions.     

Optimized 1D-CNN model for medicinal Psyllium Husk crop mapping with temporal optical satellite data

Deep Learning models are preferred for complex image analysis-based solutions to application-oriented problems. However, the architecture of such models largely influences the results which includes several hyperparameters that need to be tuned. This study aims at developing an optimized 1D-CNN model for medicinal Psyllium Husk crop mapping using open source temporal optical Sentinel-2A/2B satellite data. In this study, a sequential 1D-CNN model architecture was developed by optimizing hyperparameters which includes convolution layers, number of neurons, activation function, and batch size. Psyllium Husk crop fields were mapped in the Jalore district of Rajasthan using Sentinel 2A/ 2B (10 m) optical data. For spectral dimensionality reduction of the data, Modified Soil Adjusted Vegetation Index (MSAVI2) was used to maintain the data dimensionality since temporal data was utilized. The dataset was subsequently refined to include the target crop's specific phenological stages that distinguish it from other closely resembling species. The information corresponding to these specific crop stages was fed to the 1D-CNN model to carry out the classification. A range of training sample sizes were explored to determine the optimal number of training data points. As the output from the model, fractional images are obtained consisting of values proportional to the probability of a pixel lying in the target class. Accuracy assessment was carried out using fuzzy error matrix (FERM) by generating fractional output images from temporal optical PlanetScope data (3m) which was used as a reference. The best overall accuracy among the test cases came out to be 89.85% using conventional MSAVI2 with 1000 training samples.     

The role of weed invasion in controlling sand dune reactivation in abandoned fields in semi-arid Inner Mongolia, China

The growth of crops and invading weeds were examined at six sites under different cultivation conditions in semi-arid Inner Mongolia, where sand dune reactivation has been controlled by weed invasion. The soil moisture content was not sufficient for crop growth in over-cultivated farmland. The shortage of moisture in the soil suppresses the growth of both crops and weeds. Accordingly, farmers carry out weeding to maintain better soil moisture conditions for crop growth. Over-cultivated farmland is often abandoned when the crop yield decreases to an economically unprofitable level. If weed vegetation before abandonment remains, it becomes a core of vegetation expansion and recovers the entire soil surface. While weedy vegetation is lacking or has been grazed by livestock, sand dune reactivation occurs.     

Autecology of Bradyrhizobium japonicum in soybean-rice rotations

The effect of rice culture on changes in the number of a strain of soybean root-nodule bacteria, (Bradyrhizobium japonicum CB1809), already established in the soil by growing inoculated soybean crops, was investigated in transitional red-brown earth soils at two sites in south-western New South Wales. At the first site, 5.5 years elapsed between the harvest of the last of four successive crops of soybean and the sowing of the next. In this period three crops of rice and one crop of triticale were sown and in the intervals between these crops, and after the crop of triticale, the land was fallowed. Before sowing the first rice crop, the number of Bradyrhizobium japonicum was 1.32×10⁵ g^–1 soil. The respective numbers of bradyrhizobia after the first, second and third rice crops were 4.52 ×10⁴, 1.26×10⁴ and 6.40×10² g^–1 soil. In the following two years the population remained constant. Thus sufficient bradyrhizobia survived in soil to nodulate and allow N₂-fixation by the succeeding soybean crop. At the second site, numbers of bradyrhizobia declined during a rice crop, but the decline was less than when the soil was fallowed (400-fold cf. 2200-fold). Multiplication of bradyrhizobia was rapid in the rhizosphere of soybean seedlings sown without inoculation in the rice bays. At 16 days after sowing, their numbers were not significantly different (p<0.05) from those in plots where rice had not been sown. Nodulation of soybeans was greatest in plots where rice had not been grown, but yield and grain nitrogen were not significantly different (p<0.05). Our results indicate that flooding soil has a deleterious effect on the survival of bradyrhizobia but, under the conditions of the experiments, sufficient B. japonicum strain CB 1809 survived to provide good nodulation after three crops of rice covering a total period of 5.5 years between crops of soybean.     

Types, structure and potential for axial water flow in the deepest roots of field-grown cereals

Deep root systems that extend into moist soil can significantly increase plant productivity. Here, the components of soil-grown root systems of wheat, barley and triticale are characterized, and types and water conducting potential of deep roots in the field are assessed. Root system components were characterized in plants grown in soil in PVC tubes, based on their origin and number and the arrangement of xylem tracheary elements (XTE) viewed using fluorescence microscopy. A new nomenclature is proposed. Deep roots were harvested in the field, and root types of the current crop and remnant roots from previous crops were identified by fluorescence and cryo-scanning electron microscopy. Four types of axile (framework) and five types of branch root were distinguished in the three cereals. Six per cent of deep roots were axile roots that originated from the base of the embryo; 94% were branch roots, of which 48% had only two XTE (10 microm diameter), and thus potentially low axial flow. Only 30% of roots in the cores were from the current crop, the remainder being remnants. Selection for more deep-penetrating axile roots and increased vascular capacity of deep branches is of potential benefit. Conventional root-length density measurements should be interpreted and applied cautiously.     

Effects of stubble and N fertilization management on N availability and uptake under successive rice (Oryza sativa L.) crops

Experiments were conducted in fields which had a history of nil to four rice (Oryza sativa L.) crops during the previous four summers. Incorporating stubble after each harvest reduced soil nitrate-N content between crops, but increased soil N mineralization potential. During the fourth successive crop, plots where stubble had been incorporated after the previous three harvests had an average 21% more soil NH₄N and 22% more N uptake than plots where stubble had been burnt.Soil fertility fell rapidly with increasing numbers of crops, and the unfertilized fifth crop accumulated approximately half the N (60 kg N ha^-1) found in the unfertilized first crop (116 kg). Fertilizer N alleviated the effects of annual cropping; the application of 210 kg N ha^-1 to the fifth crop (uptake of 156 kg N ha^-1) resulted in similar N uptake to the first crop fertilized with 50 kg N ha^-1 (154 kg N ha^-1).Applying N at sowing had no significant effect on soil NH₄-N concentration after permanent flood (PF), while N application at PF resulted in increased NH₄-N concentration and N uptake until panicle initiation (PI). N applied at PI increased soil NH₄-N concentration at least until the microspore stage.Management factors such as stubble incorporation and increasing N application rate, maintained N supply and enabled successive rice crops to accumulate similar quantities of N at maturity.     

Stratification of density in dry deciduous forest using satellite remote sensing digital data—An approach based on spectral indices

Forest density expressing the stocking status constitutes the major stand physiognomic parameter of Indian forest. Density and age are often taken as surrogate to structural and compositional changes that occur with the forest succession. Satellite remote sensing spectral response is reported to provide information on structure and composition of forest stands. The various vegetation indices are also correlated with forest canopy closure. The paper presents a three way crown density model utilizing the vegetation indices viz., advanced vegetation index, bare soil index and canopy shadow index for classification of forest crown density. The crop and water classes which could not be delineated by the model were finally masked from normalized difference vegetation index and TM band 7 respectively. The rule based approach has been implemented for land use and forest density classification. The broad land cover classification accuracy has been found to be 91.5%. In the higher forest density classes the classification accuracy ranged between 93 and 95%, whereas in the lower density classes it was found to be between 82 and 85%.     

Experimental determination and modelling of the soil water extraction capacities of crops of maize,sunflower, soya bean,sorghum and wheat

The estimation of soil water reserves is essential for irrigation management. The usual way of calculating these reserves, held between the soil moisture content at field capacity and the classical limit of –1.5 MPa considered as the lower limit of available water, over the rooting depth of the crop, does not correspond with the real behaviour of crops as regards their ability to extract soil water and should be only considered as the apparent available water (AAW). Measurements of moisture profiles made using a neutron probe soil moisture meter from 1970 until 1991 on unirrigated crops at the INRA Agronomy Station at Toulouse-Auzeville, France, on a deep silty clay soil with a high water holding capacity have enabled us to define the water extraction capacities of maize ( Zea mays L.), sunflower (Helianthus annuus L.), sorghum (Sorghum bicolor L. Moench), soya bean (Glycine max L. Merr.), and winter wheat (Triticum aestivum L.). The results show, not only that all the crops can extract soil water from beyond –1.5 MPa in the surface layers to varying degrees and depths, depending on the crop, but also that deeper down, AAW is not fully used, as the moisture profile gradually returns to field capacity. Of the five crops studied, maize extracts the most water from the top 0.5 m, removing 150% of AAW. This amount falls rapidly lower down, reaching nil at 1.6 m. Conversely sunflower extracts less near the surface, but uses all AAW up to 1.2 m, and still extracts 85% of AAW at 1.6 m. Sorghum is somewhat comparable to sunflower, but with a lower use over the entire profile. Soya bean exhibits strong extraction to 1.0 m, and then much less at depth. As to wheat, its extraction capability is quite high near the surface, and then falls steadily with depth where it is still 30% of AAW at 1.6 m. Soil moisture measurements realised on a bare soil during several successive years were used to fix the maximum soil evaporation and to suggest the contribution of crops in soil water depletion from uppermost layers.The water extraction capacities have been modelled and introduced into the model EPICphase, a modified version of the model EPIC, adapted for irrigation management. Four parameters have been introduced to simulate: (1) the rooting pattern of the crop (parameter ), (2) the degree of involvement of deep layers (parameter p), (3) the fraction of AAW beyond which crop transpiration is affected (parameter t) and (4) the intensity of extraction beyond the limit of –1.5 MPa as a function of soil depth (parameter d). Calibrated on the basis of the driest year since 1970 for each crop, the model was then validated under unirrigated conditions, and then tested on irrigated maize plots. Under unirrigated conditions, the simulations correctly reproduced the water extraction by the five crops, both in an extremely dry year and in a wet year. The observed differences between simulations and observations were found mostly at about 0.1 m depth, and were due to lack of precision of moisture measurements with the neutron probe. From 0.2 to 0.6 m the simulations have a tendency to overestimate the extraction. These differences are explained by water fluxes which are especially high in these layers because of the processes of evaporation from the soil and plant transpiration, which are difficult to simulate with precision. Below 0.6 m, a more stable zone where water movements are of minor importance, the simulations are very precise. For irrigated maize, the results show a very good fit between simulation and measurement, indicating that these water extraction capacity figures could be used for irrigation management provided that the rules for exploitation of the water reserves are well established.     

Automated classification of urinary stones based on microcomputed tomography images using convolutional neural network

PurposeThe classification of urinary stones is important prior to treatment because the treatments depend on three types of urinary stones, i.e., calcium, uric acid, and mixture stones. We have developed an automatic approach for the classification of urinary stones into the three types based on microcomputed tomography (micro-CT) images using a convolutional neural network (CNN).Materials and methodsThirty urinary stones from different patients were scanned in vitro using micro-CT (pixel size: 14.96 μm; slice thickness: 15 μm); a total of 2,430 images (micro-CT slices) were produced. The slices (227 × 227 pixels) were classified into the three categories based on their energy dispersive X-ray (EDX) spectra obtained via scanning electron microscopy (SEM). The images of urinary stones from each category were divided into three parts; 66%, 17%, and 17% of the dataset were assigned to the training, validation, and test datasets, respectively. The CNN model with 15 layers was assessed based on validation accuracy for the optimization of hyperparameters such as batch size, learning rate, and number of epochs with different optimizers. Then, the model with the optimized hyperparameters was evaluated for the test dataset to obtain classification accuracy and error.ResultsThe validation accuracy of the developed approach with CNN with optimized hyperparameters was 0.9852. The trained CNN model achieved a test accuracy of 0.9959 with a classification error of 1.2%.ConclusionsThe proposed automated CNN-based approach could successfully classify urinary stones into three types, namely calcium, uric acid, and mixture stones, using micro-CT images.