For example, by automatically comparing old and new image data, geospatial technology can detect new buildings, connected additions, swimming pools, parking lots, out buildings and more. Technological advances notwithstanding, property appraisals are about people. Property owners want more frequent and accurate valuations to take advantage of rising real estate markets or, as the market falls, for possible tax savings.
Appraisers are tasked with the staggering job of keeping up with changes to every property. For example, Harris County, Texas, which includes the city of Houston, has about 1 million properties, and approximately 20 percent of them have to be assessed each year. Homeowners are enthralled at seeing satellite imagery of their property via Google Earth. Now technology companies have been challenged to come up with different ways to marry aerial and satellite imagery, computers on the ground, and hand-drawn paper sketches and maps to offer solutions that aid the appraisal process, improve property record keeping and retrieval, and help derive valuation forecasts for budget building.
Aerial and satellite sensor capabilities developed during wars in Afghanistan and Iraq have been adapted for civilian needs, including those of property appraisers who seek geospatial help. Computer-aided mass appraisal CAMA systems, which use algorithms to match inputs of property descriptions and comparable values with often-volatile market trends, have grown in capability.
Such growth has fostered a demand for more ways to drive the process and direct the people who manage it. Operatively, following the creation of ground deformation maps, displacement TS for each measurement point from both ascending and descending orbits are systematically and automatically analysed to identify any change in the deformation pattern over the last few months.
Whenever a TS exhibits a non-linear behaviour, a breaking point Tb is identified and defined. The monitoring system presented in this paper has been tested, tuned and refined thanks to the combined effort of a number of geologists and signal processing engineers, as well as with input of both Regional and Civil Protection authorities, trying to fulfil their requirements. Different temporal windows and velocity thresholds were tested. Anomalies are analysed, update by update, with the support of thematic information i. Anomalies are interpreted, assigning them a driving force, i.
The latter includes both local soil consolidation, related to load imposition and interaction soil-structures, and wide-area land subsidence linked to groundwater table depletion. It should be pointed out, however, that SAR data — in the system described in this paper - do not directly generate any alert or early warning to decision makers.
Although supported by TS analysis tools, the data are analysed and interpreted by a group of geologists and engineers before generating bulletins highlighting any anomalous areas. To that end, two important parameters must be taken into consideration: i the spatial consistency: a single measurement point exhibiting an anomalous deformation pattern is not considered representative of actual deformation. Only a group of neighbour points i. It also worth recalling that TS are a zero-redundancy product, i. On average, anomalies are identified at each update.
A sudden increase in this value was recorded starting from Update This increase was linked to the appearance, within the provinces of Pistoia and Prato, of a large cluster of points affected by a significant change in displacement rates, most likely related to subsidence induced by groundwater depletion, which historically affects this part of Tuscany 14 , 15 , TS anomalies classified according to the driving force as at Update Anomalies related to slope instabilities are widespread in most of the mountain areas of the region.
Anomalies related to subsidence phenomena are identified in the alluvial plains, along with two uplifting areas within the province of Grosseto and Firenze. Anomalies linked to geothermal activities straddle the provinces of Pisa, Siena and Grosseto. Most of the anomalies related to slope instability are located around the flanks of the Monte Amiata volcanic cone, between the provinces of Siena and Grosseto and along the Apennines in the northern part of the Tuscany Region, as should be expected, since they are the most landslide-prone areas of the region.
Significant groups of anomalies are also located in the Chianti Hills and in the northern parts of the Apuan Alps. Anomalies driven by land subsidence are registered in the alluvial areas of the Tuscany region i. Two small clusters of anomalies related to uplift are present in the lower sector of the Ombrone River valley and in the plain area between Firenze and Prato.
All delivered data can be accessed via a web-service tool. The bulletin contains a classification of each Tuscan municipality according to the presence of persistent movement anomalies. Each municipality is classified according to the absence green or presence of new yellow or persistent orange anomalies of movement.
The colour red is assigned when a group of persistent anomalies, analysed together with other data sources land cover maps, rainfall data, hazard maps, high-resolution optical data, etc. The colour red means that further analyses are needed. Within the bulletin the localization of the areas which necessity of further analysis is included, along with a preliminary analysis of the anomalies of movement in term of spatial consistency and temporal persistency. Field investigations are then performed in these areas to determine the severity of the hazard, initiate management of the risk and decide, together with local authorities, the most appropriate actions to mitigate the threats.
Monitoring bulletins released every 12 days to the Tuscany Region authorities. The presence of a cluster of persistent anomalies affecting elements at risk determines a significant level of risk, with the necessity of field survey and further analysis.
Whenever needed, some hot spots can become targets for detailed analyses with high-resolution radar sensors e. The main driver of our work is the design of new paradigms for monitoring our planet at regional and national scales by trying to better exploit Earth Observation EO satellites. While meteorological satellites are already considered an operational tool providing a stream of information that allows operators to continuously update weather models, EO images in particular radar have been rarely used in operational monitoring projects at the regional scale for risk mitigation.
In fact, so far, satellite radar data have been mostly used after an event 27 i. Some landslides have been instrumented, some others slowed through remedial works, but it is not feasible to decrease all the threats or protect all the affected areas. Considering that we can neither monitor nor prevent them all, a different strategy for risk mitigation must be conceived. We suggest using Sentinel-1 radar images more and more operationally, similarly to weather data. The project described in this paper is somewhat unique, since it implements an advanced monitoring service at the regional scale, where satellite radar data feed a decision support system for hydrogeological risk mitigation strategies.
Leveraging the enhanced imaging capabilities of Sentinel-1, we provide regional authorities continuous information on where, when and how fast the ground is moving. However, prioritization and mitigation of these hazards can be done, starting with issues deemed to be most urgent. We also suggest that satellite radar data, systematically acquired over large areas with short revisit times, could be used not only as a tool for mapping unstable areas but also for monitoring, which is one of the pillars of any Early Warning Systems EWSs So far, EWSs based on displacement monitoring are usually implemented at the local scale and for single landslides e.
On the contrary, EWSs at the regional scale are typically developed for rainfall-induced landslides and are based on indirect indicators of slope instabilities, including rainfall intensity and duration, cumulative storm or event rainfall, combined with antecedent soil moisture conditions e. To this end, Sentinel-1 satellites can be considered a breakthrough, paving the way for the use of InSAR measurements for slope instability monitoring at regional or even wider scales and in places where installation of ground-based devices would be unfeasible.
Regional-scale EWSs based on continuous processing of Sentinel-1 data are now possible. Designed to acquire for decades to come, the Sentinel-1 satellites are ideally suited to capture deformation data over a sufficiently long period, providing a stream of measurements for the continuous updating of ground deformation patterns on a wide scale. The regularity of acquisition guarantees the creation, immediately after every new SAR images, of displacement time series that can be used to systematically feed landslide failure forecast models e.
THIS IS SOME TITLE
This service marks the transition from historical satellite analyses to near-real-time monitoring schemes based on systematic SAR imagery processing and analysis of deformation time series, coupled with automatic tools for data mining and screening of large datasets with millions of measurement points covering thousands of square kilometres. In any monitoring system supporting civil protection activities, one of the most important requirements - from the point of view of the end user - is the timely delivery of up-to-date, reliable, high quality, products. Processing time must be as short as possible.
As already pointed out, in the monitoring system activated by the Regional Government in Tuscany, as soon as a new Sentinel-1 A or B image is acquired over the region, data are immediately downloaded and processed, time series are updated, and possible anomalies are automatically detected and highlighted, for further data interpretation and integration. The whole SqueeSAR processing chain has been carefully reviewed and optimized, trying to reduce significantly the delay between data acquisition and data delivery.
It should be remarked anomalies of movement identified with the continuous processing of SAR data do not directly generate any alert or early warning to decision makers. The presence of clusters of anomalies and their temporal persistence are the most important parameters to assign, reliably, a link to a driving force.
- Satellite Remote Sensing: Sensors, Applications and Techniques | SpringerLink?
- Handbook of Anatomical Models for Radiation Dosimetry (Series in Medical Physics and Biomedical Engineering)?
- You are here;
- Site-Specific Management Guidelines.
- All Themes | ESA Business Applications.
We demonstrate that advances in satellite radar sensors, processing algorithms, data mining tools and cloud computing allow the design of new monitoring systems, providing an advanced tool for risk mitigation strategies across wide scales. This monitoring system is designed to capture changes in the deformation pattern occurring at regional scale, such as precursor movements that are usually recorded before landslide failures or sinkholes.
With this purpose, the revisiting time should be as low as possible to effectively track the evolution of the deformation 36 without aliasing affects 37 and to forecast the incoming failure The persistent view of the land surface is the key parameter to observe and understand processes related to ground movements.
Sentinel-1A and Sentinel-1B mark the start of a new era in the Earth Observation system worldwide: they represent the flagships of six families of the Sentinel satellite missions of the revolutionary Copernicus programme and are set to provide timely and accurate data and to support long-term operational programmes for decades to come.
The continuous monitoring of ground deformation at regional or national scales is now possible using Sentinel-1, coupling the short revisiting time, the wide-scale mapping capability, the regularity of acquisitions and the free data access with new automatic tools for data mining and screening of large datasets with millions of measurement points covering thousands of square kilometres.
This operational service is presented and discussed through the case study of the Tuscany Region Central Italy , known to be a landslide-prone area with also subsiding areas mapped throughout the region. To accomplish the transition from historical satellite analyses to near-real-time monitoring schemes based on systematic SAR imagery processing, acquisition plans of Sentinel-1 images have been set for both ascending and descending geometry.
As soon as a new Sentinel-1 image is available, the satellite data are automatically downloaded and processed using the SqueeSAR algorithm via parallel processing and cloud computing. Information on persistent anomalies affecting elements at risk is routinely delivered to regional authorities.
Opinion | What it would take for India to become a proper space power
Field investigations are then performed in these areas to determine the severity of the hazard and the level of risk. The methodology adopted in this monitoring service follows a step-wise approach which encompasses four different phases Fig. Finally, a monitoring bulletin is released every 12 days to the Regional authorities, with the necessary information of the areas where field survey and more detailed analysis are needed.
More details of the adopted methodology are presented in the following sections. Flow chart of the methodology adopted in this monitoring service. As it becomes available, the new Sentinel-1 images is immediately downloaded and processed with the archive stack. TS are analysed with automatic tools for data mining and screening to identify anomalies. To ensure a timely delivery, anomalies are interpreted by a group of radar-interpreter within few hours from their identification.
Datasets used in this service include Sentinel-1 C-band images central frequency 5. In fact, the combination of the rotation of the Earth and the motion of the satellite along an almost-polar orbit makes it possible to acquire data all over the planet from two different acquisition geometries: one with the satellite moving from south to north ascending geometry and one with the sensor moving from north to south descending.
Ground deformation maps for the Tuscany Region have been generated through the continuous SqueeSAR processing of Sentinel-1 acquisitions.
The main idea behind PSInSAR is to identify point-like targets corresponding to single pixels or groups of a few pixels exhibiting good phase coherence over the entire observation period via proper statistical analyses. These radar-bright and radar-phase stable points, which exist within any radar scene of the available stack, are usually referred to as Permanent Scatterers PS. Having a stable radar signature, PSs are slightly affected by decorrelation phenomena and the level of backscattered signal is much higher than the inherent noise of the sensor i.
Amplitude data of backscattered signal are analysed computing the so-called amplitude stability index ratio, between the average amplitude of return relative to each pixel and its standard deviation. PS are identified to separate a modelled deformation rate, atmospheric screen, and elevation error components of the phase changes in the radar signal between different satellite passes.
PS usually correspond to rock outcrops, roads, buildings and all manmade objects widely available over a city but are less common in non-urban areas. DS correspond to groups of pixels and are typically associated with homogeneous, low-reflectivity areas, such as scattered outcrops, bare soil, debris-covered zones and non-cultivated lands. Given a multitemporal stack of SAR scenes properly re-sampled on the same grid, the basic idea of SqueeSAR is to identify sets of pixels sharing the same kind of radar return, i. Using the amplitude rather than complex values of the stack of SAR images, the cumulative distribution functions of the amplitudes for the pixels points under consideration are obtained.
KS is a nonparametric test, i. The most important parameter is the maximum absolute difference between two cumulative distribution functions. The distance between the distributions determines if the two points are statistically drawn from the same distribution. Specifically, DS are identified according to the following steps: 1 selection and analysis of each single pixel of the image; 2 creation of a window centred around the pixel; 3 comparison of adjacent pixels with the KS test; 4 further processing and analysis of statistically homogeneous pixels, while pixels with different distribution functions are discarded; 5 the DS identified within statistically homogeneous areas are processed using the traditional PSInSAR algorithm for the estimation of the deformation maps and the construction of displacement time series of each measurement point MP.
Common to conventional geodetic networks, all data are differential measurements with respect to a reference point that is assumed to be motionless. Apart from constraints set by radar parameters and MP distribution, all reference point candidates were carefully selected based on: 1 geological considerations e.
The actual values differ from point to point, depending on the characteristic of the scatterer and the distance from the reference point. Reset Brightness 0. Reset Contrast 0. Reset Saturation 0. Reset Sharpness 0. Reset Exposure 0. Reset Hue 0. Reset Gama 0. Applying filters. Search Inside This report can be searched.