Browsing by Subject "Spatial Patterns"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    An Introduction to Spatial Data Mining
    (University Consortium for Geographic Information Science, 2020) Golmohammadi, Jamal; Xie, Yiqun; Gupta, Jayant; Farhadloo, Majid; Li, Yan; Cai, Jiannan; Detor, Samantha; Roh, Abigail; Shekhar, Shashi
    The goal of spatial data mining is to discover potentially useful, interesting, and non-trivial patterns from spatial data-sets (e.g., GPS trajectory of smartphones). Spatial data mining is societally important having applications in public health, public safety, climate science, etc. For example, in epidemiology, spatial data mining helps to find areas with a high concentration of disease incidents to manage disease outbreaks. Computational methods are needed to discover spatial patterns since the volume and velocity of spatial data exceed the ability of human experts to analyze it. Spatial data has unique characteristics like spatial autocorrelation and spatial heterogeneity which violate the i.i.d (Independent and Identically Distributed) assumption of traditional statistic and data mining methods. Therefore, using traditional methods may miss patterns or may yield spurious patterns, which are costly in societal applications. Further, there are additional challenges such as MAUP (Modifiable Areal Unit Problem) as illustrated by a recent court case debating gerrymandering in elections. In this article, we discuss tools and computational methods of spatial data mining, focusing on the primary spatial pattern families: hotspot detection, colocation detection, spatial prediction, and spatial outlier detection. Hotspot detection methods use domain information to accurately model more active and high-density areas. Colocation detection methods find objects whose instances are in proximity to each other in a location. Spatial prediction approaches explicitly model the neighborhood relationship of locations to predict target variables from input features. Finally, spatial outlier detection methods find data that differ from their neighbors. Lastly, we describe future research and trends in spatial data mining.
  • Thumbnail Image
    Item
    Spatiotemporal Complexity of Fire in an Island-Lake Landscape, Border Lakes Region, Minnesota, USA
    (2020-08) Schneider, Elizabeth
    The Border Lakes Region of Minnesota is a unique location to evaluate historical patterns of fire events owing to complex dynamics between the landscape, climate, land use, and role of disturbance. The fragmented landscape and resulting variability in topography may impart important controls on where fire occurs and how fire behaves. My dissertation evaluates how spatial and temporal patterns of historical fire in red pine dominated forests are driven by climate, landscape characteristics, and human land use. My research is aimed at identifying the mechanisms responsible for variations in fire occurrence, such as those that lead to large (synchronous) fire events versus small (asynchronous) fire events. Specifically, I assessed (1) the spatial and temporal patterns of synchronous and asynchronous fire events, (2) the drivers associated with the occurrence of synchronous and asynchronous fires, and (3) evaluate how the tenets of Island Biogeography, area and isolation, help explain patterns in historical fire events in red pine forests of the Border Lakes Region of Minnesota. I have been able to demonstrate that climate, specifically periods of extended drought, are responsible for larger, synchronous fire events while smaller, asynchronous fire events were not related to the variability landscape characteristics and likely related to human land use. In addition, fires were frequent on both islands and mainland sites and the fire event dates between these sites are similar across the landscape. Significant temporal variability in fire events occurred on islands and mainland sites between 1780 and the late 1800s, with fire events accumulating more on islands prior to 1830 and mainland sites accumulating more fire events after 1860. I speculate that fires in the Border Lakes region accumulated more rapidly on islands between 1780 and 1830 due to intense use of the landscape by humans, corresponding to the fur trade era. This result has significant weight regarding management considerations where historically, research has suggested that Indigenous communities have contributed relatively little to the frequency of ignitions. My research argues for the greater integration of traditional practices in resource management, specifically regarding prescribed burning where Indigenous communities likely had a significant effect in red pine forests.
  • Thumbnail Image
    Item
    Temporal changes in spatial patterns of moose browse, causes and consequences.
    (2010-03) Hodgson, Angela Lynne
    Ecologists determine mechanisms by observing spatial and temporal patterns of abundance and distribution in natural systems. While there has been a long history of research on techniques for describing temporal patterns of abundance, and their causes and consequences, there is still a need for ecological research to focus on the causes and consequences of spatial patterns. Progress on this goal, though, has been hindered by the lack of long-term data on spatial patterns in natural ecosystems. I present findings from one of the first long-term studies of changes in spatial patterns of plants in response to herbivory, and discuss causes and consequences. My research was conducted in the southern boreal forest on Isle Royale, Michigan, and focused on temporal changes in the spatial pattern of woody browse species that are consumed by a large herbivore (moose). I concluded that browsed woody sapling biomass is aggregated within moose feeding stations and the degree of aggregation has changed over a 20-year period. The cause of this fluctuating pattern of aggregation is due to the competing influences of inverse density dependent browsing by moose, causing an increase in aggregation, and inverse density dependent growth, causing a decrease in aggregation. Annual changes in aggregation are determined by the relative contribution of consumption and growth to changes in spatial pattern in any given year. Next, I developed a simulation model to determine the consequence of aggregation of browse on the intake rate of large herbivores. I found that the spatial aggregation of browse within feeding stations can decrease the intake rate up to 30% for herbivores feeding on low density browse (<15 g/m2). Both density and the spatial arrangement of browse, therefore, is important to consider when determining the functional response of large herbivores. Finally, I used long-term data on consumption and spatial pattern of browse to test whether a mechanistic modified contingency model could predict observed moose diet selection. This model assumes that moose select their diet in order to maximize short-term intake rate. Model predictions were consistent with observed diet selection during both summer and winter in two study sites.