scholarly journals A hierarchical framework for segmenting movement paths

2019 ◽  
Author(s):  
Wayne M. Getz
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Animal Movement ◽  
Spatial Scales ◽  
Process Models ◽  
Series Data ◽  
Temporal Scales ◽  
Environmental Covariates ◽  
Spatio Temporal ◽  
State And Local

AbstractComparative applications of animal movement path analyses are hampered by the lack of a comprehensive framework for linking structures and processes conceived at different spatio-temporal scales. Although many analyses begin by generating step-length (SL) and turning-angle (TA) distributions from relocation time-series data—some of which are linked to ecological, landscape, and environmental covariates—the frequency at which these data are collected may vary from sub-seconds to several hours, or longer. The kinds of questions that may be asked of these data, however, are very much scale-dependent. It thus behooves us to clarify how the scale at which SL and TA data are collected and relate to one another, as well as the kinds of ecological questions that can be asked. Difficulties arise because the information contained in SL and TA time series is not semantically aligned with the physiological, ecological, and sociological factors that influence the structure of movement paths. I address these difficulties by classifying movement types at salient temporal scales using two different kinds of vocabularies. The first is the language derived from behavioral and ecological concepts. The second is the language derived from mathematically formulated stochastic walks. The primary tools for analyzing these walks are fitting time-series and stochastic-process models to SL and TA statistics (means, variances, correlations, individual-state and local environmental covariates), while paying attention to movement patterns that emerge at various spatial scales. The purpose of this paper is to lay out a more coherent, hierarchical, scale-dependent, appropriate-complexity framework for conceptualizing path segments at different spatio-temporal scales and propose a method for extracting a simulation model, referred to as M3, from these data when at a relatively high frequencies (ideally minute-by-minute). Additionally, this framework is designed to bridge biological and mathematical movement ecology concepts; thereby stimulating the development of conceptually-rooted methods that facilitates the formulation of our M3 model for simulating theoretical and analyzing empirical data, which can then be used to test hypothesis regarding mechanisms driving animal movement and make predications of animal movement responses to management and global change.

Growth rings across the Tree of Life: demographic insights from biogenic time series data

2021 ◽  
pp. 77-96
Author(s):  
Margaret E. K. Evans ◽  
Bryan A. Black ◽  
Donald A. Falk ◽  
Courtney L. Giebink ◽  
Emily L. Schultz
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Spatial Scales ◽  
Ocean Dynamics ◽  
Individual Growth ◽  
Series Data ◽  
Rate Variation ◽  
Sampling Effort ◽  
Growth Rings ◽  
Demographic Processes

Biogenic time series data can be generated in a single sampling effort, offering an appealing alternative to the slow process of revisiting or recapturing individuals to measure demographic rates. Annual growth rings formed by trees and in the ear bones of fish (i.e. otoliths) are prime examples of such biogenic time series. They offer insight into not only the process of growth but also birth, death, movement, and evolution, sometimes at uniquely deep temporal and large spatial scales, well beyond 5–30 years of data collected in localised study areas. This chapter first reviews the fundamentals of how tree-ring and otolith time series data are developed and analysed (i.e. dendrochronology and sclerochronology), then surveys growth rings in other organisms, along with microstructural or microcompositional variation in growth rings, and other records of demographic processes. It highlights the answers to demographic questions revealed by these time series data, such as the influence of environmental (atmospheric or ocean) conditions, competition, and disturbances on demographic processes, and the genetic versus plastic basis of individual growth and traits that influence growth. Lastly, it considers how spatial networks of biogenic, annually resolved time series data can offer insights into the importance of macrosystem atmospheric and ocean dynamics on multispecies, trophic dynamics. The authors encourage demographers to integrate the complementary information contained in biogenic time series data into population models to better understand the drivers of vital rate variation and predict the impacts of global change.

scholarly journals Uncertainty Analysis of Object-Based Land-Cover Classification Using Sentinel-2 Time-Series Data

2020 ◽  
Vol 12 (22) ◽  
pp. 3798
Author(s):  
Lei Ma ◽  
Michael Schmitt ◽  
Xiaoxiang Zhu
Keyword(s):  
Time Series ◽  
Land Cover ◽  
Time Series Data ◽  
Land Cover Classification ◽  
Series Data ◽  
Spectral Bands ◽  
Object Based ◽  
Spatio Temporal ◽  
Optical Satellite Images ◽  
The Impact

Recently, time-series from optical satellite data have been frequently used in object-based land-cover classification. This poses a significant challenge to object-based image analysis (OBIA) owing to the presence of complex spatio-temporal information in the time-series data. This study evaluates object-based land-cover classification in the northern suburbs of Munich using time-series from optical Sentinel data. Using a random forest classifier as the backbone, experiments were designed to analyze the impact of the segmentation scale, features (including spectral and temporal features), categories, frequency, and acquisition timing of optical satellite images. Based on our analyses, the following findings are reported: (1) Optical Sentinel images acquired over four seasons can make a significant contribution to the classification of agricultural areas, even though this contribution varies between spectral bands for the same period. (2) The use of time-series data alleviates the issue of identifying the “optimal” segmentation scale. The finding of this study can provide a more comprehensive understanding of the effects of classification uncertainty on object-based dense multi-temporal image classification.

scholarly journals Spatio-temporal assessment of ecological disturbance and its intensity in the Mangrove forest using MODIS derived disturbance index

2014 ◽  
Vol XL-8 ◽  
pp. 555-559 ◽  
Author(s):  
D. Dutta ◽  
P. K. Das ◽  
S. Paul ◽  
J. R. Sharma ◽  
V. K. Dadhwal
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Mangrove Ecosystem ◽  
Series Data ◽  
Landsat 8 ◽  
Mangrove Forests ◽  
Ecological Disturbance ◽  
Disturbance Index ◽  
Disturbed Areas ◽  
Spatio Temporal

The mangrove ecosystem of Sundarbans region plays an important ecological and socio-economical role in both India and Bangladesh. The ecological disturbance in the coastal mangrove forests are mainly attributed to the periodic cyclones caused by deep depression formed over the Bay of Bengal. In the present study, three of the major cyclones in the Sundarbans region were analyzed to establish the cause-and-effect relationship between cyclones and the resultant ecological disturbance. The Moderate Resolution Imaging Spectroradiometer (MODIS) time-series data was used to generate MODIS global disturbance index (MGDI) and its potential was explored to assess the instantaneous ecological disturbance caused by cyclones with varying landfall intensities and at different stages of mangrove phenology. The time-series MGDI was converted into the percentage change in MGDI using its multi-year mean for each pixel, and its response towards several cyclonic events was studied. The affected areas were identified by analyzing the Landsat-8 satellite data before and after the cyclone and the MGDI values of the affected areas were utilized to develop the threshold for delineation of the disturbed pixels. The selected threshold was applied on the time-series MGDI images to delineate the disturbed areas for each year individually to identify the frequently disturbed areas. The classified intensity map could able to detect the chronically affected areas, which can serve as a valuable input towards modelling the biomigration of the invasive species and efficient forest management.

scholarly journals Missing Value Imputation of Time-Series Air-Quality Data via Deep Neural Networks

2021 ◽  
Vol 18 (22) ◽  
pp. 12213
Author(s):  
Taesung Kim ◽  
Jinhee Kim ◽  
Wonho Yang ◽  
Hunjoo Lee ◽  
Jaegul Choo
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Air Quality ◽  
Time Series Data ◽  
Quality Data ◽  
Series Data ◽  
Missing Value ◽  
Missing Value Imputation ◽  
Spatio Temporal ◽  
Air Quality Data

To prevent severe air pollution, it is important to analyze time-series air quality data, but this is often challenging as the time-series data is usually partially missing, especially when it is collected from multiple locations simultaneously. To solve this problem, various deep-learning-based missing value imputation models have been proposed. However, often they are barely interpretable, which makes it difficult to analyze the imputed data. Thus, we propose a novel deep learning-based imputation model that achieves high interpretability as well as shows great performance in missing value imputation for spatio-temporal data. We verify the effectiveness of our method through quantitative and qualitative results on a publicly available air-quality dataset.

scholarly journals Gaussian processes for time-series modelling

2013 ◽  
Vol 371 (1984) ◽  
pp. 20110550 ◽  
Author(s):  
S. Roberts ◽  
M. Osborne ◽  
M. Ebden ◽  
S. Reece ◽  
N. Gibson ◽  
...  
Keyword(s):  
Time Series ◽  
Gaussian Processes ◽  
Domain Knowledge ◽  
Time Series Data ◽  
Process Models ◽  
Series Data ◽  
Parametric Modelling ◽  
Case Examples ◽  
Time Series Modelling ◽  
Time Series Data Analysis

In this paper, we offer a gentle introduction to Gaussian processes for time-series data analysis. The conceptual framework of Bayesian modelling for time-series data is discussed and the foundations of Bayesian non-parametric modelling presented for Gaussian processes . We discuss how domain knowledge influences design of the Gaussian process models and provide case examples to highlight the approaches.

scholarly journals Crop Yield Prediction Using Multitemporal UAV Data and Spatio-Temporal Deep Learning Models

2020 ◽  
Vol 12 (23) ◽  
pp. 4000
Author(s):  
Petteri Nevavuori ◽  
Nathaniel Narra ◽  
Petri Linna ◽  
Tarmo Lipping
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Crop Yield ◽  
Time Series Data ◽  
Weather Data ◽  
Series Data ◽  
Percentage Error ◽  
Time Series Modelling ◽  
Spatio Temporal ◽  
3D Cnn

Unmanned aerial vehicle (UAV) based remote sensing is gaining momentum worldwide in a variety of agricultural and environmental monitoring and modelling applications. At the same time, the increasing availability of yield monitoring devices in harvesters enables input-target mapping of in-season RGB and crop yield data in a resolution otherwise unattainable by openly availabe satellite sensor systems. Using time series UAV RGB and weather data collected from nine crop fields in Pori, Finland, we evaluated the feasibility of spatio-temporal deep learning architectures in crop yield time series modelling and prediction with RGB time series data. Using Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM) networks as spatial and temporal base architectures, we developed and trained CNN-LSTM, convolutional LSTM and 3D-CNN architectures with full 15 week image frame sequences from the whole growing season of 2018. The best performing architecture, the 3D-CNN, was then evaluated with several shorter frame sequence configurations from the beginning of the season. With 3D-CNN, we were able to achieve 218.9 kg/ha mean absolute error (MAE) and 5.51% mean absolute percentage error (MAPE) performance with full length sequences. The best shorter length sequence performance with the same model was 292.8 kg/ha MAE and 7.17% MAPE with four weekly frames from the beginning of the season.

Sign in / Sign up

Export Citation Format

Share Document