Comparison between artificial neural networks and maximum likelihood classification in digital soil mapping

Soil surveys are the main source of spatial information on soils and have a range of different applications, mainly in agriculture. The continuity of this activity has however been severely compromised, mainly due to a lack of governmental funding. The purpose of this study was to evaluate the feasibility of two different classifiers (artificial neural networks and a maximum likelihood algorithm) in the prediction of soil classes in the northwest of the state of Rio de Janeiro. Terrain attributes such as elevation, slope, aspect, plan curvature and compound topographic index (CTI) and indices of clay minerals, iron oxide and Normalized Difference Vegetation Index (NDVI), derived from Landsat 7 ETM+ sensor imagery, were used as discriminating variables. The two classifiers were trained and validated for each soil class using 300 and 150 samples respectively, representing the characteristics of these classes in terms of the discriminating variables. According to the statistical tests, the accuracy of the classifier based on artificial neural networks (ANNs) was greater than of the classic Maximum Likelihood Classifier (MLC). Comparing the results with 126 points of reference showed that the resulting ANN map (73.81 %) was superior to the MLC map (57.94 %). The main errors when using the two classifiers were caused by: a) the geological heterogeneity of the area coupled with problems related to the geological map; b) the depth of lithic contact and/or rock exposure, and c) problems with the environmental correlation model used due to the polygenetic nature of the soils. This study confirms that the use of terrain attributes together with remote sensing data by an ANN approach can be a tool to facilitate soil mapping in Brazil, primarily due to the availability of low-cost remote sensing data and the ease by which terrain attributes can be obtained.

Artificial neural networks applied for soil class prediction in mountainous landscape of the Serra do Mar¹

Soil information is needed for managing the agricultural environment. The aim of this study was to apply artificial neural networks (ANNs) for the prediction of soil classes using orbital remote sensing products, terrain attributes derived from a digital elevation model and local geology information as data sources. This approach to digital soil mapping was evaluated in an area with a high degree of lithologic diversity in the Serra do Mar. The neural network simulator used in this study was JavaNNS and the backpropagation learning algorithm. For soil class prediction, different combinations of the selected discriminant variables were tested: elevation, declivity, aspect, curvature, curvature plan, curvature profile, topographic index, solar radiation, LS topographic factor, local geology information, and clay mineral indices, iron oxides and the normalized difference vegetation index (NDVI) derived from an image of a Landsat-7 Enhanced Thematic Mapper Plus (ETM+) sensor. With the tested sets, best results were obtained when all discriminant variables were associated with geological information (overall accuracy 93.2 - 95.6 %, Kappa index 0.924 - 0.951, for set 13). Excluding the variable profile curvature (set 12), overall accuracy ranged from 93.9 to 95.4 % and the Kappa index from 0.932 to 0.948. The maps based on the neural network classifier were consistent and similar to conventional soil maps drawn for the study area, although with more spatial details. The results show the potential of ANNs for soil class prediction in mountainous areas with lithological diversity.

Early Changes in Soil Metabolic Diversity and Bacterial Community Structure in Sugarcane under Two Harvest Management Systems

Preharvest burning is widely used in Brazil for sugarcane cropping. However, due to environmental restrictions, harvest without burning is becoming the predominant option. Consequently, changes in the microbial community are expected from crop residue accumulation on the soil surface, as well as alterations in soil metabolic diversity as of the first harvest. Because biological properties respond quickly and can be used to monitor environmental changes, we evaluated soil metabolic diversity and bacterial community structure after the first harvest under sugarcane management without burning compared to management with preharvest burning. Soil samples were collected under three sugarcane varieties (SP813250, SP801842 and RB72454) and two harvest management systems (without and with preharvest burning). Microbial biomass C (MBC), carbon (C) substrate utilization profiles, bacterial community structure (based on profiles of 16S rRNA gene amplicons), and soil chemical properties were determined. MBC was not different among the treatments. C-substrate utilization and metabolic diversity were lower in soil without burning, except for the evenness index of C-substrate utilization. Soil samples under the variety SP801842 showed the greatest changes in substrate utilization and metabolic diversity, but showed no differences in bacterial community structure, regardless of the harvest management system. In conclusion, combined analysis of soil chemical and microbiological data can detect early changes in microbial metabolic capacity and diversity, with lower values in management without burning. However, after the first harvest, there were no changes in the soil bacterial community structure detected by PCR-DGGE under the sugarcane variety SP801842. Therefore, the metabolic profile is a more sensitive indicator of early changes in the soil microbial community caused by the harvest management system.

Metabolic attributes, yield and stability of milk in Jersey cows fed diets containing sodium citrate and sodium bicarbonate

The objective of this work was to evaluate the inclusion of sodium citrate and sodium bicarbonate in the diet of lactating Jersey cows, and its effects on the metabolic attributes, productivity and stability of milk. We evaluated urinary pH, levels of glucose and urea in blood, body weight, body condition score, milk yield, milk stability (ethanol test), and milk physicochemical properties of 17 cows fed diets containing sodium citrate (100 g per cow per day), sodium bicarbonate (40 g per cow per day) or no additives. Assessments were made at the 28th and 44th days. Supply of sodium citrate or bicarbonate has no influence on the metabolic attributes, productivity, body weight, and body condition score of the cows, neither on the composition and stability of milk.

Metabolic responses of guava trees irrigated withdifferent N and K levels in São Francisco Valley

The guava (Psidium guajava L.) cv. Paluma has been cultivated in São Francisco Valley, Northeastern of Brazil, for in natura consumption and processing purposes. In spite of its importance, there are few scientific knowledge regarding guava physiology, nutrition, irrigation and fertigation. The objective of this work was to evaluate the effect of weather conditions and different concentrations of N and K applied by fertigation in foliar contents of reducing sugars, total soluble sugars, starch, sucrose, amino acids, and proteins. The field experiment was carried out at Bebedouro Experimental Field and the biochemical evaluations at the Laboratory of Seed and Plant Physiology, both located at Embrapa Semi-Árido, Petrolina-PE. The doses of 200 g N and 100 g K2O; 400 g N and 200 g K2O; 600 g N and 300 g K2O; and 800 g N and 400 g K2O per plant were applied in an experiment field. The experimental design was totally randomized blocks, with four treatments and five blocks. The weather conditions influenced the plant photosynthesis, which affects the plants metabolism. Guava presented specific responses to N and K fertigation for each parameter evaluated. The weather conditions during the evaluation period influenced guava responses to N and K fertigation.