Search

This paper proposes a novel data preprocessing method that converts numeric data into representative graphs (polygons) expressing all of the relationships between data variables in a systematic way based on Hamiltonian cycles. The advantage of the proposed method is that it has an embedded feature extraction capability in which each generated polygon depicts a class-specific representation in the data, thereby supporting accurate "end-to-end learning" in industrial fault classification applications. Moreover, the generated polygons can play a significant role in the interpretation of trained deep learning fault classifiers. The performance of the proposed method was demonstrated using a benchmark dataset in the process industry. It was also tested successfully to classify challenging faults in major equipment in a thermomechanical pulp mill located in Canada. The results of the proposed method show better performance than other comparable fault classifiers. [ABSTRACT FROM AUTHOR]

Changing environmental and geomorphological conditions are resulting in vegetation change in ice-wedge polygons in Arctic tundra. However, we do not yet know how microscale vegetation patterns relate to individual environmental and geomorphological parameters. This work aims at examining these relations in polygonal terrain. We analysed composition and cover of vascular plant taxa and surface height, active layer depth, soil temperature, carbon and nitrogen content, pH and electrical conductivity in four polygon mires located on the Yukon coast. We found that vascular plant species composition and cover correlates best with relative surface height. Ridges of low-centred polygons and raised centres of high-centred polygons support the growth of mesic and wetland species (e.g., Betula glandulosa, Salix pulchra, S. reticulata, Rubus chamaemorus, various ericaceous dwarf shrubs, Eriophorum vaginatum, Poa arctica). Wetland and aquatic plant species (e.g., E. angustifolium, Carex aquatilis, C. chordorrhiza, Pedicularis sudetica) grow in low-lying centres of polygons and in troughs between polygons. We also found a relationship between vascular plant species composition and substrate characteristics such as pH, electrical conductivity and total organic carbon, although the individual influence of these parameters could not be determined because of their correlation with relative surface height. Our findings stress the regulatory role of microtopography and substrate in vegetation dynamics of polygonal terrain. Ongoing warming in this region will lead to changes to polygonal terrain through permafrost degradation and subsequent conversion of low-centred into high-centred polygons. Our results indicate that shrubs, particularly Betula glandulosa and heath species, have the potential to expand most. [ABSTRACT FROM AUTHOR]

We recently completed the accuracy assessment of a Landsat-derived landcover polygon layer covering the entire province of Alberta (660,000 km2), Canada, for which we gathered reference information for nearly 5000 randomly selected polygons ranging from two hectares to thousands of hectares in size. This gave us the unique opportunity to quantify, for the first time, how the probability of correctly classifying a landcover object varies with its size. Irrespective of whether they are represented as polygons or as sets of connected pixels with the same label, the classification accuracy of landcover objects decreases as their size decreases, steadily for large and medium sizes, and more dramatically when they are within two orders of magnitude of the pixel size of the input image. We show that this size-dependency is bound to occur whenever the size distribution of landcover objects follows an inverse power law. Our results are consistent with previous studies on related issues, confirm the need to account for size when assessing the accuracy of object-based landcover maps, and cast doubts on the validity of (1) recently proposed object-based accuracy estimators, and (2) landscape pattern analyses where the minimum patch size is close to the pixel size. [ABSTRACT FROM AUTHOR]

The article assesses the risk of phosphorus (P) desorption from Canadian agricultural land by the source component of the indicator of water contamination by P (IROWC-P). The IROWC-P source component (P_source) allegedly characterized the mobilization potential of soluble P and integrated four models of P desorption by water for dominant agricultural soil series of Canada on the soil landscape of Canada polygon scale. A standardized method for deriving P_source component is described and evaluated over 5-year intervals from 1981-2006. Results reveal trends of soil P enrichment for most Canadian provinces over the 25-year period but also an increased percentage of farmland classified above the water extractable soil P environmental threshold of 4mg P kg.

Long-term vegetation succession and permafrost dynamics in subarctic peat plateaus of west-central Canada have been studied through detailed plant macrofossil analysis and extensive AMS radiocarbon dating of two peat profiles. Peatland inception at these sites occurred around 5800-5100 yr BP (6600-5900 cal. BP) as a result of paludification of upland forests. At the northern peat plateau site, located in the continuous permafrost zone, palaeobotanical evidence suggests that permafrost was already present under the forested upland prior to peatland development. Paludification was initiated by permafrost collapse, but re-aggradation of permafrost occurred soon after peatland inception. At the southern site, located in the discontinuous permafrost zone, the aggradation of permafrost occurred soon after peatland inception. In the peat plateaus, permafrost conditions have remained very stable until present. Sphagnum fuscum-dominated stages have alternated with more xerophytic communities characterized by ericaceous shrubs. Local peat fires have occurred, but most of these did not cause degradation of the permafrost. Starting from 2800-1100 yr BP (2900-1000 cal. BP) consistently dry surface conditions have prevailed, possibly related to continued frost heave or nearby polygon crack formation. [ABSTRACT FROM AUTHOR]