Satellite Image Time Series Analysis Research Articles

Satellite Image Time Series Analysis for Big Earth Observation Data

The development of analytical software for big Earth observation data faces several challenges. Designers need to balance between conflicting factors. Solutions that are efficient for specific hardware architectures can not be used in other environments. Packages that work on generic hardware and open standards will not have the same performance as dedicated solutions. Software that assumes that its users are computer programmers are flexible but may be difficult to learn for a wide audience. This paper describes sits, an open-source R package for satellite image time series analysis using machine learning. To allow experts to use satellite imagery to the fullest extent, sits adopts a time-first, space-later approach. It supports the complete cycle of data analysis for land classification. Its API provides a simple but powerful set of functions. The software works in different cloud computing environments. Satellite image time series are input to machine learning classifiers, and the results are post-processed using spatial smoothing. Since machine learning methods need accurate training data, sits includes methods for quality assessment of training samples. The software also provides methods for validation and accuracy measurement. The package thus comprises a production environment for big EO data analysis. We show that this approach produces high accuracy for land use and land cover maps through a case study in the Cerrado biome, one of the world’s fast moving agricultural frontiers for the year 2018.

Quality control and class noise reduction of satellite image time series

The extensive amount of Earth observation satellite images available brings opportunities and challenges for land mapping in global and regional scales. These large datasets have motivated the use of satellite image time series analysis coupled with machine learning techniques to produce land use and cover class maps. To be successful, these methods need good quality training samples, which are the most important factor for determining the accuracy of the results. For this reason, training samples need methods for quality control of class noise. In this paper, we propose a method to assess and improve the quality of satellite image time series training data. The method uses self-organizing maps (SOM) to produce clusters of time series and Bayesian inference to assess intra-cluster and inter-cluster similarity. Consistent samples of a class will be part of a neighborhood of clusters in the SOM map. Noisy samples will appear as outliers in the SOM. Using Bayesian inference in the SOM neighborhoods, we can infer which samples are noisy. To illustrate the methods, we present a case study in a large training set of land use and cover classes in the Cerrado biome, Brazil. The results prove that the method is efficient to reduce class noise and to assess the spatio-temporal variation of satellite image time series training samples.

Analyzing the dynamics of urbanization in Delhi National Capital Region in India using satellite image time-series analysis

Understanding urban land-use changes and accurately quantifying urban land transitions is essential to global land-change research. The present study aimed to capture non-linear land transitions within urban areas using an automated change detection technique in a satellite image time series. Traditional land-use and cover maps used to map and monitor urban areas assume land change is a linear process and that urbanization is the last stage of land transition. In reality, however, most land transitions are non-linear. The present study focused on Delhi National Capital Territory, in India, and its adjacent major cities. A popular time-series analysis method was applied on MODIS NDVI time-series (2000–2017) data to detect change within the impervious surface area of the region. Overall validation and analysis of the results showed that the method was able to capture the direction and timing of the changes very well within all levels of urban density (except very high-density areas with more than 98% built-up density). The majority of urban areas in the region experienced interrupted, abrupt, and gradual greening. The results show different examples of non-linear land transitions detected from satellite images. Until recently, these land transitions could only be observed via long-term field surveys and/or local knowledge. The results reveal that the land-change trajectories can be different based on the level of built-up density, size of the urban area, physical proximity, and accessibility to relatively bigger urban areas. Knowledge gained from this study can be useful in better understanding the micro-climatic patterns and environmental quality within a city.

Identifying Spatiotemporal Patterns in Land Use and Cover Samples from Satellite Image Time Series

The use of satellite image time series analysis and machine learning methods brings new opportunities and challenges for land use and cover changes (LUCC) mapping over large areas. One of these challenges is the need for samples that properly represent the high variability of land used and cover classes over large areas to train supervised machine learning methods and to produce accurate LUCC maps. This paper addresses this challenge and presents a method to identify spatiotemporal patterns in land use and cover samples to infer subclasses through the phenological and spectral information provided by satellite image time series. The proposed method uses self-organizing maps (SOMs) to reduce the data dimensionality creating primary clusters. From these primary clusters, it uses hierarchical clustering to create subclusters that recognize intra-class variability intrinsic to different regions and periods, mainly in large areas and multiple years. To show how the method works, we use MODIS image time series associated to samples of cropland and pasture classes over the Cerrado biome in Brazil. The results prove that the proposed method is suitable for identifying spatiotemporal patterns in land use and cover samples that can be used to infer subclasses, mainly for crop-types.

Radar Satellite Image Time Series Analysis for High-Resolution Mapping of Man-Made Forest Change in Chongming Eco-Island

Forests are important oxygen sources and carbon sinks. They provide a series of ecosystem services that are crucial to eco-island sustainability, such as the protection of soils, conservation of biodiversity, and development of the eco-tourism, etc. Interestingly, Chongming eco-island is located in the borderlands between fresh- and saltwater environments, where the Yangtze River meets the East China Sea. Most forests in Chongming island are therefore man-made and very vulnerable to the ecological environment mixing of freshwater streams and rivers with salty ocean tides, and are affected by climate and human activity. Estimating and monitoring forest change within an estuary is required for the sustainable management of forest resources and forest-based eco-tourism. Most optical satellites are unsuitable for continuous forest mapping due to cloud cover and their relatively low spatial and temporal resolution. Here, using Sentinel-1 satellite carrying an imaging C-band synthetic aperture radar, radar vegetation index was derived to detect the forest dynamics on the island. Furthermore we quantified forest area change in three well known and the most strictly protected and representative areas, namely Dongping National Forest Park, Dongtan National Wetland Park, and Xisha National Wetland Park, in the Chongming eco-island over the past five years at 10-metre resolution. We recorded the early and mid summers when the forest canopies grew to the peak in the study areas. The planted forest in Dongping National Forest Park grew an area of 7.38 hectares from 2015 to 2019, and disappeared from an area of almost 2.59 hectares in 2018. The man-made forest of Xisha National Wetland Park increased at an area of almost 20.19 hectares over the past five years. The forest in Dongtan National Wetland Park increased to an area of almost 2.12 hectares in the period of 2015–2017 and 2018–2019. However, from 2017 to 2018, the area of planted forests in Dongtan National Wetland Park decreased by 1.35 hectares. This study shows man-made forest change can be measured and that evidence can be collected to show how the forest is altered by human activities, and informs forest management decision-making for Chongming eco-island.

Land Cover Change Impacts on Land Surface Temperature in Jakarta and Its Satellite Cities

Indonesia’s capital city, Jakarta, is surrounded by several satellite cities and has grown spatially into a larger region called Jabodetabek. The development in Jabodetabek has resulted in changes in the landscape. The increase of impervious surface and decrease of green space has led to the formation of an urban heat island (UHI). UHI could be detected through land surface temperature (LST) monitoring. Therefore, the objectives of this study were to analyze the land cover changes in Jakarta and its satellite cities and to analyze the impact of these changes on LST using GIS-based analysis. We analyzed Jakarta and its three satellite cities, Depok, Tangerang, and Bekasi, through a satellite image time-series analysis in 1989/1990 and 2015/2018. It is found that, among the four cities we analyzed, Tangerang has the highest built-up area expansion proportion, followed by Bekasi with a slight difference, then Depok and Jakarta. As for the LST, the city that has the most significant difference between mean LST in the initial and recent year is Bekasi with 12.66°C, then followed by Tangerang (11.05°C), Jakarta (8.34°C), and Depok (6.43°C). These orders’ inconsistency might be caused by higher proportion of built-up area combined with higher loss of vegetated area. This shows that the land cover change played a big role in the temperature increase.

A 60-year chronology of glacier surges in the central Karakoram from the analysis of satellite image time-series

The glaciers in the Karakoram have recently been studied intensely using satellite data. The related studies investigated a large variety of characteristics (e.g. short and long-term variability in flow velocities, elevation changes and extent) and revealed a behaviour that differs largely from most other regions of the world (e.g. advancing or stable termini and balanced to positive mass budgets over the past decades). In part, this is also due to the high concentration of surge-type glaciers that are particular attractive for remote sensing studies owing to their large short-term changes. Other studies provided a comprehensive overview on the timing of surges in the Karakoram over recent decades. However, limited information is available for the details of the individual surges and what happened before 1989, partly resulting in a wrong understanding of the dynamics. To provide a more extended base of glacier surging in the central Karakoram, this study utilises the abundant declassified satellite images from the Keyhole Mission (Corona and Hexagon) covering the period 1961 to 1980 as well as a historic map from 1937 to decipher the surge history of 27 glaciers since the 1960s and before. The focus is on the detailed description of the pictorial evidence and the conclusions that can be drawn from it. A major outcome of this analysis is that most of the glaciers that have surged recently, have also surged in the 1950 to 1960s as surge cycles are often in the 40 to 60 year range. Combined with further evidence of previous surges, it seems that the so-called ‘Karakoram Anomaly’ is around since at least 200 years. It is also shown that two glaciers that are listed as surging are only expanding, that flow blocking of tributaries plays an important role for several surges, that surging glaciers are mostly free of debris, and that surges show a great variability in regard to surge durations (from 2 to 50+ years) and surge cycles (from 20 to 80+ years). Despite this variability, the pattern of post-surge down-wasting and disintegration is always similar, allowing to derive the timing of surges beyond the available pictorial evidence and also to distinguish a surge from a usual advance which is otherwise difficult.

DtwSat: Time-Weighted Dynamic Time Warping for Satellite Image Time Series Analysis in R

The opening of large archives of satellite data such as LANDSAT, MODIS and the SENTINELs has given researchers unprecedented access to data, allowing them to better quantify and understand local and global land change. The need to analyze such large data sets has led to the development of automated and semi-automated methods for satellite image time series analysis. However, few of the proposed methods for remote sensing time series analysis are available as open source software. In this paper we present the R package dtwSat. This package provides an implementation of the time-weighted dynamic time warping method for land cover mapping using sequence of multi-band satellite images. Methods based on dynamic time warping are flexible to handle irregular sampling and out-of-phase time series, and they have achieved significant results in time series analysis. Package dtwSat is available from the Comprehensive R Archive Network (CRAN) and contributes to making methods for satellite time series analysis available to a larger audience. The package supports the full cycle of land cover classification using image time series, ranging from selecting temporal patterns to visualizing and assessing the results.

Object-oriented satellite image time series analysis using a graph-based representation

Nowadays, remote sensing technologies produce huge amounts of satellite images that can be helpful to monitor geographical areas over time. A satellite image time series (SITS) usually contains spatio-temporal phenomena that are complex and difficult to understand. Conceiving new data mining tools for SITS analysis is challenging since we need to simultaneously manage the spatial and the temporal dimensions at the same time. In this work, we propose a new clustering framework specifically designed for SITS data. Our method firstly detects spatio-temporal entities, then it characterizes their evolutions by mean of a graph-based representation, and finally it produces clusters of spatio-temporal entities sharing similar temporal behaviors. Unlike previous approaches, which mainly work at pixel-level, our framework exploits a purely object-based representation to perform the clustering task. Object-based analysis involves a segmentation step where segments (objects) are extracted from an image and constitute the element of analysis. We experimentally validate our method on two real world SITS datasets by comparing it with standard techniques employed in remote sensing analysis. We also use a qualitative analysis to highlight the interpretability of the results obtained.

Vegetation Regrowth Following Wildfires in the Santa Cruz Mountains of Northern California Monitored Using Landsat Satellite Image Analysis

The Santa Cruz Mountain range in northern California is a coastal landscape with a history of extensive forest logging and frequent large wildfires that have recently destroyed numerous residential structures at the wildland interface. Results from Landsat satellite image time-series analysis since 1984 of the study area within the Los Gatos Creek and Corralitos Creek watersheds showed that none of the severe drought periods since the 1980s have notably inhibited rapid tree and shrub regrowth rates on steep hill slopes burned recently by the 1985 Lexington Fire and the 2008 Summit Fire. In high burn severity areas of both fires, post-fire vegetation types showed a marked increase in shrub cover, mainly at the expense of evergreen tree cover. Most of these low (·ha-1 of standing woody biomass on the burned areas. This study is the first of its kind to utilize a full 30-year record of Landsat vegetation index data to monitor tree and shrub regrowth after stand-replacing wildfires in California.

Thirty years of vegetation change in the coastal Santa Cruz Mountains of Northern California detected using landsat satellite image analysis

The Santa Cruz Mountains is a coastal landscape with a history of extensive forest logging, and a future with projected climate warming that may alter vegetation cover and surface water runoff in new ways. Results from Landsat satellite image time-series analysis since 1983 of this study area showed gradual, statistically significant increases in the normalized difference vegetation index (NDVI) in more than 90 % of the (predominantly second-growth) evergreen forest locations sampled. The cumulative distribution of NDVI values in 2013 was significantly different and higher overall from the cumulative distribution of NDVI values in 1983. The extreme drought year of 2013 (and other previous years of low precipitation) did not affect average NDVI growth rates in most drainage basins of the study area, with the exception of four relatively small basins that had less than 30 % forested land cover. Notably different patterns of NDVI change were detected in areas burned by wildfires in recent years. Within the perimeters of the 2008 Summit Fire and the 2009 Lockheed Fire, NDVI showed notable declines from pre-fire levels to those calculated in 2013 Landsat imagery. In contrast to these recent fires, the burned area from the 1985 Lexington Fire showed the highest rate of NDVI increase (over 27 years of regrowth) of any relatively large contiguous area within the Santa Cruz Mountains.

Monitoring vegetation change and dynamics on U.S. Army training lands using satellite image time series analysis

Given the significant land holdings of the U.S. Department of Defense, and the importance of those lands to support a variety of inherently damaging activities, application of sound natural resource conservation principles and proactive monitoring practices are necessary to manage military training lands in a sustainable manner. This study explores a method for, and the utility of, analyzing vegetation condition and trends as sustainability indicators for use by military commanders and land managers, at both the national and local levels, in identifying when and where vegetation-related environmental impacts might exist. The BFAST time series decomposition method was applied to a ten-year MODIS NDVI time series dataset for the Fort Riley military installation and Konza Prairie Biological Station (KPBS) in northeastern Kansas. Imagery selected for time-series analysis were 16-day MODIS NDVI (MOD13Q1 Collection 5) composites capable of characterizing vegetation change induced by human activities and climate variability. Three indicators related to gradual interannual or abrupt intraannual vegetation change for each pixel were calculated from the trend component resulting from the BFAST decomposition. Assessment of gradual interannual NDVI trends showed the majority of Fort Riley experienced browning between 2001 and 2010. This result is supported by validation using high spatial resolution imagery. The observed versus expected frequency of linear trends detected at Fort Riley and KPBS were significantly different and suggest a causal link between military training activities and/or land management practices. While both sites were similar with regards to overall disturbance frequency and the relative spatial extents of monotonic or interrupted trends, vegetation trajectories after disturbance were significantly different. This suggests that the type and magnitude of disturbances characteristic of each location result in distinct post-disturbance vegetation responses. Using a remotely-sensed vegetation index time series with BFAST and the indicators outlined here provides a consistent and relatively rapid assessment of military training lands with applicability outside of grassland biomes. Characterizing overall trends and disturbance responses of vegetation can promote sustainable use of military lands and assist land managers in targeting specific areas for various rehabilitation activities.