The sensitivity of mapping methods to reference data quality:training supervised image classifications with imperfect reference data

The accuracy of a map is dependent on the reference dataset used in its construction. Classification analyses used in thematic mapping can, for example, be sensitive to a range of sampling and data quality concerns. With particular focus on the latter, the effects of reference data quality on land cover classifications from airborne thematic mapper data are explored. Variations in sampling intensity and effort are highlighted in a dataset that is widely used in mapping and modelling studies; these may need accounting for in analyses. The quality of the labelling in the reference dataset was also a key variable influencing mapping accuracy. Accuracy varied with the amount and nature of mislabelled training cases with the nature of the effects varying between classifiers. The largest impacts on accuracy occurred when mislabelling involved confusion between similar classes. Accuracy was also typically negatively related to the magnitude of mislabelled cases and the support vector machine (SVM), which has been claimed to be relatively insensitive to training data error, was the most sensitive of the set of classifiers investigated, with overall classification accuracy declining by 8% (significant at 95% level of confidence) with the use of a training set containing 20% mislabelled cases.

The 'rotiferologist' effect and other global correlates of species richness in monogonont rotifers

Global biodiversity patterns are often driven by diff erent environmental variables at diff erent scales. However, it is still controversial whether there are general trends, whether similar processes are responsible for similar patterns, and/or whether confounding eff ects such as sampling bias can produce misleading results. Our aim is twofold: 1) assessing the global correlates of diversity in a group of microscopic animals little analysed so far, and 2) inferring the infl uence of sampling intensity on biodiversity analyses. As a case study, we choose rotifers, because of their high potential for dispersal across the globe. We assembled and analysed a new worldwide dataset of records of monogonont rotifers, a group of microscopic aquatic animals, from 1960 to 1992. Using spatially explicit models, we assessed whether the diversity patterns conformed to those commonly obtained for larger organisms, and whether they still held true after controlling for sampling intensity, variations in area, and spatial structure in the data. Our results are in part analogous to those commonly obtained for macroorganisms (habitat heterogeneity and precipitation emerge as the main global correlates), but show some divergence (potential absence of a latitudinal gradient and of a large-scale correlation with human population). Moreover, the eff ect of sampling eff ort is remarkable, accounting for 50% of the variability; this strong eff ect may mask other patterns such as latitudinal gradients. Our study points out that sampling bias should be carefully considered when drawing conclusions from large-scale analyses, and calls for further faunistic work on microorganisms in all regions of the world to better understand the generality of the processes driving global patterns in biodiversity.

X-ray absorption and phase contrast imaging to study the interplay between plant roots and soil structure

Plant performance is, at least partly, linked to the location of roots with respect to soil structure features and the micro-environment surrounding roots. Measurements of root distributions from intact samples, using optical microscopy and field tracings have been partially successful but are imprecise and labour-intensive. Theoretically, X-ray computed micro-tomography represents an ideal solution for non-invasive imaging of plant roots and soil structure. However, before it becomes fast enough and affordable or easily accessible, there is still a need for a diagnostic tool to investigate root/soil interplay. Here, a method for detection of undisturbed plant roots and their immediate physical environment is presented. X-ray absorption and phase contrast imaging are combined to produce projection images of soil sections from which root distributions and soil structure can be analyzed. The clarity of roots on the X-ray film is sufficient to allow manual tracing on an acetate sheet fixed over the film. In its current version, the method suffers limitations mainly related to (i) the degree of subjectivity associated with manual tracing and (ii) the difficulty of separating live and dead roots. The method represents a simple and relatively inexpensive way to detect and quantify roots from intact samples and has scope for further improvements. In this paper, the main steps of the method, sampling, image acquisition and image processing are documented. The potential use of the method in an agronomic perspective is illustrated using surface and sub-surface soil samples from a controlled wheat trial. Quantitative characterization of root attributes, e.g. radius, length density, branching intensity and the complex interplay between roots and soil structure, is presented and discussed.

On sampling collection procedure effectiveness for forest soil characterization

One of the most important measures to prevent wild forest fires is the use of prescribed and controlled burning actions as it reduce the fuel mass availability. The impact of these management activities on soil physical and chemical properties varies according to the type of both soil and vegetation. Decisions in forest management plans are often based on the results obtained from soil-monitoring campaigns. Those campaigns are often man-labor intensive and expensive. In this paper we have successfully used the multivariate statistical technique Robust Principal Analysis Compounds (ROBPCA) to investigate on the sampling procedure effectiveness for two different methodologies, in order to reflect on the possibility of simplifying and reduce the sampling collection process and its auxiliary laboratory analysis work towards a cost-effective and competent forest soil characterization.

Forest soil short term response to prescribed fire: an approach by Principal Components Analysis

In the current context of serious climate changes, where the increase of the frequency of some extreme events occurrence can enhance the rate of periods prone to high intensity forest fires, the National Forest Authority often implements, in several Portuguese forest areas, a regular set of measures in order to control the amount of fuel mass availability (PNDFCI, 2008). In the present work we’ll present a preliminary analysis concerning the assessment of the consequences given by the implementation of prescribed fire measures to control the amount of fuel mass in soil recovery, in particular in terms of its water retention capacity, its organic matter content, pH and content of iron. This work is included in a larger study (Meira-Castro, 2009(a); Meira-Castro, 2009(b)). According to the established praxis on the data collection, embodied in multidimensional matrices of n columns (variables in analysis) by p lines (sampled areas at different depths), and also considering the quantitative data nature present in this study, we’ve chosen a methodological approach that considers the multivariate statistical analysis, in particular, the Principal Component Analysis (PCA ) (Góis, 2004). The experiments were carried out in a soil cover over a natural site of Andaluzitic schist, in Gramelas, Caminha, NW Portugal, who was able to maintain itself intact from prescribed burnings from four years and was submit to prescribed fire in March 2008. The soils samples were collected from five different plots at six different time periods. The methodological option that was adopted have allowed us to identify the most relevant relational structures inside the n variables, the p samples and in two sets at the same time (Garcia-Pereira, 1990). Consequently, and in addition to the traditional outputs produced from the PCA, we have analyzed the influence of both sampling depths and geomorphological environments in the behavior of all variables involved.

Prevalence and intensity of soil-transmitted helminthiasis in the city of Portoviejo (Ecuador)

We studied the stool samples of 151 school children in a district of the city of Portoviejo (Ecuador) in order to determine the prevalence and intensity of soil-transmitted helminthiasis (STH) and their relationships with anthropometric indices. The samples were analyzed with the semiquantitative Kato-Katz technique and the intensity of infections was categorized as light, moderate or high according to the thresholds set by the World Health Organization. Prevalence of soil transmitted helmintiasis was 65% (92 out of 141 collected samples), Ascaris lumbricoides was the most common STH (63%) followed by Trichuris trichiura (10%) and hookworm (1.4%). Heavy intensity infections were found in 8.5% of the stool samples, with T. trichiura showing higher worm burdens than A. lumbricoides. Sixteen percent of the children were below the third percentile for weight (wasted), while 27% were below the third percentile for height (stunted). A significant relationship was found between the worm burden and the degree of stunting. This study suggests that the periodic administration of an antihelminthic drug should be targeted to preschool and school children to allow a normal growth spurt and prevent stunting.

Sampling and TDR probe insertion in the determination of the volumetric soil water content

Volumetric soil water content (theta) can be evaluated in the field by direct or indirect methods. Among the direct, the gravimetric method is regarded as highly reliable and thus often preferred. Its main disadvantages are that sampling and laboratory procedures are labor intensive, and that the method is destructive, which makes resampling of a same point impossible. Recently, the time domain reflectometry (TDR) technique has become a widely used indirect, non-destructive method to evaluate theta. In this study, evaluations of the apparent dielectric number of soils (epsilon) and samplings for the gravimetrical determination of the volumetric soil water content (thetaGrav) were carried out at four sites of a Xanthic Ferralsol in Manaus - Brazil. With the obtained epsilon values, theta was estimated using empirical equations (thetaTDR), and compared with thetaGrav derived from disturbed and undisturbed samples. The main objective of this study was the comparison of thetaTDR estimates of horizontally as well as vertically inserted probes with the thetaGrav values determined by disturbed and undisturbed samples. Results showed that thetaTDR estimates of vertically inserted probes and the average of horizontally measured layers were only slightly and insignificantly different. However, significant differences were found between the thetaTDR estimates of different equations and between disturbed and undisturbed samples in the thetaGrav determinations. The use of the theoretical Knight et al. model, which permits an evaluation of the soil volume assessed by TDR probes, is also discussed. It was concluded that the TDR technique, when properly calibrated, permits in situ, nondestructive measurements of q in Xanthic Ferralsols of similar accuracy as the gravimetric method.

Soil compaction and eucalyptus growth in response to forwarder traffic intensity and load

During timber exploitation in forest stands harvesting machines pass repeatedly along the same track and can cause soil compaction, which leads to soil erosion and restricted tree root growth. The level of soil compaction depends on the number of passes and weight of the wood load. This paper aimed to evaluate soil compaction and eucalyptus growth as affected by the number of passes and wood load of a forwarder. The study was carried out in Santa Maria de Itabira county, Minas Gerais State - Brazil, on a seven-year-old eucalyptus stand planted on an Oxisol. The trees were felled by chainsaw and manually removed. Plots of 144 m² (four rows 12 m long in a 3 x 2 m spacing) were then marked off for the conduction of two trials. The first tested the traffic intensity of a forwarder which weighed 11,900 kg and carried 12 m³ wood (density of 480 kg m-3) and passed 2, 4, and 8 times along the same track. In the second trial, the forwarder carried loads of 4, 8, and 12 m³ of wood, and the machine was driven four times along the same track. In each plot, the passes affected four rows. Eucalyptus was planted in 30 x 30 x 30 cm holes on the compacted tracks. The soil in the area is clayey (470 clay and 440 g kg-1 sand content) and at depths of 0-5 cm and 5-10 cm, respectively, soil organic carbon was 406 and 272 g kg-1 and the moisture content during the trial 248 and 249 g kg-1. These layers were assessed for soil bulk density and water-stable aggregates. The infiltration rate was measured by a cylinder infiltrometer. After 441 days the measurements were repeated, with additional analyses of: soil organic carbon, total nitrogen, N-NH4+, N-NO3-, porosity, and penetration resistance. Tree height, stem diameter, and stem dry matter were measured. Forwarder traffic increased soil compaction, resistance to penetration and microporosity while it reduced the geometric mean diameter, total porosity, macroporosity and infiltration rate. Stem dry matter yield and tree height were not affected by soil compaction. Two passes of the forwarder were enough to cause the disturbances at the highest levels. The compaction effects were still persistent 441 days after forwarder traffic.

A simple device for sampling of volumetric forest soil cores

Summary

Variation in soil organic carbon and thickness of soil horizons within a boreal forest stand - effect of trees and implications for sampling

Summary

Relations between soil surface roughness, tortuosity, tillage treatments, rainfall intensity and soil and water losses from a red yellow latosol

The soil surface roughness increases water retention and infiltration, reduces the runoff volume and speed and influences soil losses by water erosion. Similarly to other parameters, soil roughness is affected by the tillage system and rainfall volume. Based on these assumptions, the main purpose of this study was to evaluate the effect of tillage treatments on soil surface roughness (RR) and tortuosity (T) and to investigate the relationship with soil and water losses in a series of simulated rainfall events. The field study was carried out at the experimental station of EMBRAPA Southeastern Cattle Research Center in São Carlos (Fazenda Canchim), in São Paulo State, Brazil. Experimental plots of 33 m² were treated with two tillage practices in three replications, consisting of: untilled (no-tillage) soil (NTS) and conventionally tilled (plowing plus double disking) soil (CTS). Three successive simulated rain tests were applied in 24 h intervals. The three tests consisted of a first rain of 30 mm/h, a second of 30 mm/h and a third rain of 70 mm/h. Immediately after tilling and each rain simulation test, the surface roughness was measured, using a laser profile meter. The tillage treatments induced significant changes in soil surface roughness and tortuosity, demonstrating the importance of the tillage system for the physical surface conditions, favoring water retention and infiltration in the soil. The increase in surface roughness by the tillage treatments was considerably greater than its reduction by rain action. The surface roughness and tortuosity had more influence on the soil volume lost by surface runoff than in the conventional treatment. Possibly, other variables influenced soil and water losses from the no-tillage treatments, e.g., soil type, declivity, slope length, among others not analyzed in this study.

N2O emissions from a cultivated mollisol: optimal time of day for sampling and the role of soil temperature

The correct use of closed field chambers to determine N2O emissions requires defining the time of day that best represents the daily mean N2O flux. A short-term field experiment was carried out on a Mollisol soil, on which annual crops were grown under no-till management in the Pampa Ondulada of Argentina. The N2O emission rates were measured every 3 h for three consecutive days. Fluxes ranged from 62.58 to 145.99 ∝g N-N2O m-2 h-1 (average of five field chambers) and were negatively related (R² = 0.34, p < 0.01) to topsoil temperature (14 - 20 ºC). N2O emission rates measured between 9:00 and 12:00 am presented a high relationship to daily mean N2O flux (R² = 0.87, p < 0.01), showing that, in the study region, sampling in the mornings is preferable for GHG.

Towards a universal sampling protocol for soil biotas in the humid tropics

This paper reviews the methods for the inventory of below-ground biotas in the humid tropics, to document the (hypothesized) loss of soil biodiversity associated with deforestation and agricultural intensification at forest margins. The biotas were grouped into eight categories, each of which corresponded to a major functional group considered important or essential to soil function. An accurate inventory of soil organisms can assist in ecosystem management and help sustain agricultural production. The advantages and disadvantages of transect-based and grid-based sampling methods are discussed, illustrated by published protocols ranging from the original "TSBF transect", through versions developed for the alternatives to Slash-and-Burn Project (ASB) to the final schemes (with variants) adopted by the Conservation and Sustainable Management of Below-ground Biodiversity Project (CSM-BGBD). Consideration is given to the place and importance of replication in below-ground biological sampling and it is argued that the new sampling protocols are inclusive, i.e. designed to sample all eight biotic groups in the same field exercise; spatially scaled, i.e. provide biodiversity data at site, locality, landscape and regional levels, and link the data to land use and land cover; and statistically robust, as shown by a partial randomization of plot locations for sampling.

Prediction of soil orders with high spatial resolution: response of different classifiers to sampling density

The objective of this work was to evaluate sampling density on the prediction accuracy of soil orders, with high spatial resolution, in a viticultural zone of Serra Gaúcha, Southern Brazil. A digital elevation model (DEM), a cartographic base, a conventional soil map, and the Idrisi software were used. Seven predictor variables were calculated and read along with soil classes in randomly distributed points, with sampling densities of 0.5, 1, 1.5, 2, and 4 points per hectare. Data were used to train a decision tree (Gini) and three artificial neural networks: adaptive resonance theory, fuzzy ARTMap; self‑organizing map, SOM; and multi‑layer perceptron, MLP. Estimated maps were compared with the conventional soil map to calculate omission and commission errors, overall accuracy, and quantity and allocation disagreement. The decision tree was less sensitive to sampling density and had the highest accuracy and consistence. The SOM was the less sensitive and most consistent network. The MLP had a critical minimum and showed high inconsistency, whereas fuzzy ARTMap was more sensitive and less accurate. Results indicate that sampling densities used in conventional soil surveys can serve as a reference to predict soil orders in Serra Gaúcha.

Effects of tillage intensity on soil C, N, and P pools with different stability

Es ist bekannt, dass die Umsatzdynamik der organischen Substanz von der Bodenbearbeitungsintensität abhängt. Bis jetzt sind nur wenige Daten zum Einfluss der Bearbeitungsintensität und des Zwischenfruchtanbaus auf C-, N-, und P-Dynamik im Ober- (0-5 cm Tiefe) und Unterboden (5-25 cm Tiefe) von Lössböden verfügbar. Hauptziele dieser Arbeit waren die (i) Quantifizierung des Einflusses von verschiedenen langzeitig durchgeführten Bearbeitungssystemen auf labile, intermediäre, und passive C- und N-Pools; (ii) Quantifizierung des Einflusses dieser Systeme auf P-Fraktionen mit unterschiedlicher Verfügbarkeit für die Pflanzenaufnahme; (iii) Quantifizierung des Einflusses des Zwischenfruchtanbaus in Verbindung mit einer unterschiedlichen Einarbeitungstiefe der der Zwischenfrüchte auf mineralisierbares C und N. Die Ergebnisse des 1. und 2. Teilexperiments basieren auf Untersuchungen von 4 Langzeitfeldexperimenten (LFE) in Ost- und Süddeutschland, die zwischen 1990 und 1997 durch das Institut für Zuckerrübenforschung angelegt wurden. Jedes LFE umfasst 3 Bearbeitungssysteme: konventionelle Bearbeitung (CT), reduzierte Bearbeitung (RT) und Direktsaat (NT). Die Ergebnisse des 3. Teilexperiments basieren auf einem Inkubationsexperiment. Entsprechend den Hauptfragestellungen wurden folgende Untersuchungsergebnisse beschrieben: (i) Im Oberboden von NT wurden höhere labile C-Vorräte gefunden (C: 1.76 t ha-1, N: 166 kg ha-1), verglichen mit CT (C: 0.44 t ha-1, N: 52 kg ha-1). Im Gegensatz dazu waren die labile- C-Vorräte höher im Unterboden von CT mit 2.68 t ha-1 verglichen zu NT mit 2 t ha-1 und RT mit 1.87 t ha-1. Die intermediären C-Vorräte betrugen 73-85% der gesamten organischen C-Vorräte, intermediäre N-Vorräte betrugen 70-95% des Gesamt-N im Ober- und Unterboden und waren vielfach größer als die labilen und passiven C- und N-Vorräte. Nur im Oberboden konnte ein Effekt der Bearbeitungsintensität auf die intermediären N-Pools mit höheren Vorräten unter NT als CT festgestellt werden. Die passiven C- und N-Pools waren eng mit den mineralischen Bodeneigenschaften verbunden und unabhängig vom Bearbeitungssystem. Insgesamt hat sich gezeigt, dass 14 bis 22 Jahre durchgängige Direktsaatverfahren nur im Oberboden zu höheren labilen C- und N-Vorräten führen, verglichen zu konventionellen Systemen. Dies lässt eine tiefenabhängige Stärke der Dynamik der organischen Bodensubstanz vermuten. (ii) Die Konzentration des Gesamt-P (Pt) im Oberboden war höher in NT (792 mg kg-1) und ~15% höher als die Pt-Konzentration in CT (691 mg kg 1). Die Abnahme der Pt-Konzentration mit zunehmender Bodentiefe war höher in NT als in CT. Dies gilt auch für die einzelnen P-Fraktionen, ausgenommen der stabilsten P-Fraktion (residual-P). Generell hatte das Bearbeitungssystem nur einen kleinen Einfluss auf die P-Konzentration mit höheren Pt-Konzentrationen in Böden unter NT als CT. Dies resultiert vermutlich aus der flacheren Einarbeitung der Pflanzenreste als in CT. (iii) Im Zwischenfruchtexperiment war der Biomassezuwachs von Senf am höchsten und nimmt in der Reihenfolge ab (oberirdischer Ertrag in t / ha): Senf (7.0 t ha-1) > Phacelia (5.7 t ha-1) > Ölrettich (4.4 t ha-1). Damit war potentiell mineralisierbares C und N am höchsten in Böden mit Senfbewuchs. Kumulative CO2- und N2O-Emissionen während der Inkubation unterschieden sich nicht signifikant zwischen den Zwischenfruchtvarianten und waren unabhängig von der Verteilung der Pflanzenreste im Boden. Die kumulativen ausgewaschenen mineralisierten N (Nmin)-Vorräte waren in den brachliegenden Böden am höchsten. Die Nmin-Vorräte waren 51-72% niedriger in den Varianten mit Zwischenfrucht und Einarbeitung verglichen zur Brache. In den Varianten ohne Einarbeitung waren die Nmin-Vorräte 36-55% niedriger verglichen zur Brache. Dies weißt auf einen deutlichen Beitrag von Zwischenfrüchten zur Reduzierung von Nitrat-Auswaschung zwischen Winter und Frühjahr hin. Insgesamt führte reduzierte Bearbeitung zu einer Sequestrierung von C und N im Boden und der Zwischenfruchtanbau führte zu reduzierten N-Verlusten. Die P-Verfügbarkeit war höher unter Direktsaat verglichen zur konventionellen Bearbeitung. Diese Ergebnisse resultieren aus den höheren Konzentrationen der OS in den reduzierten, als in den konventionellen Systemen. Die Ergebnisse zeigen deutlich das Potential von reduzierter Bearbeitung zur Sequestrierung von intermediärem C und N zur Reduzierung von klimarelevanten Treibhausgasen. Gleichzeitig steigen die Konzentrationen an pflanzenverfügaren P-Gehalten. Zwischenfrüchte führen auch zu einem Anstieg der C- und N-Vorräte im Boden, offensichtlich unabhängig von der Zwischenfruchtart.