## Abstract

Through inverse modeling and adjustment techniques, the geodesists try to derive mathematical models from their measurements to get a better understanding of various processes in the system Earth. Sophisticated deterministic and stochastic models are developed to achieve the best possible reflection of reality and the remaining uncertainty.

The main focus of this article is on the further development of stochastic model representations, with the capability to switch from the usual assumption of homogeneous (time-stationary) to inhomogeneous (time-variable) stochastic models. To accomplish this we build up and extend a methodical framework to connect the filter and the covariance approach represented by time-variable autoregressive processes (AR) and time-variable (inhomogeneous) covariance models for least squares collocation.

We apply these time-variable covariance models to describe the temporal component of a spatio-temporal point stack of surface displacements derived from a DInSAR-SBAS analysis of the ERS1 and ERS2 missions from the Lower-Rhine Embayment in North Rhine-Westphalia. The construction of a time-variable spatio-temporal covariance model allows to use the least squares collocation approach to predict the vertical movements at any location and at any time. Furthermore, a report on the uncertainty of the prediction is provided.

You have full access to this open access chapter, Download chapter PDF

## Keywords

## 1 Introduction

Concepts like stationarity and homogeneity (invariance with respect to transformation) play a central role in the treatment of stochastic processes. According to the definition of (Brockwell and Davis 1991, Def. 1.3.1) stationary processes are defined on the one hand by the unchanged (marginal) distribution function, but according to definition (Brockwell and Davis 1991, Def. 1.3.2) stationarity can also be defined by covariance stationarity. This means that the covariance function is invariant with respect to a linear transformation (see also Moritz (1980, Sec. 12)). In practice, many phenomena modelled by stochastic processes do not satisfy this requirement and exhibit a time- or location-varying character.

The term time-variable is often used differently. Following Priestley (1989, Sec. 6.1), time-variable processes can be subdivided into models with a deterministic trend (e.g. polynomial or seasonal) or with an *“explosive”* AR models where the roots of the characteristic polynomial are not only inside but also outside the unit circle. Here we want to study another type of non-stationary processes, where the coefficients of a discrete AR process *𝓈*_{t}, *t* ∈ ℤ^{+} are variable in time.

However, the motion of the coefficients must be constrained to ensure a finite variance of the resulting process, *E* {S_{t} − *E*{S_{t}})^{2}} < *∞*. Korte et al. (2022) restricts the variability of the time-varying coefficients of an AR process by the restriction that the roots of the characteristic polynomials should be only within the unit circle. Since only linear motions of the roots (poles) are allowed, this requirement can be guaranteed also for higher order processes. For general (infinite) time-variable AR processes of first order, TVAR(1), however, the condition of convergence of the product sequence of the time-variable AR coefficients is a sufficient condition to guarantee finite variances/covariances (Knopp 1922, Sec. VII). For finite AR processes these restrictions simplify accordingly.

The article is organized as follows. In the Sect. 2, we examine the time-variable AR process of first order and put special focus on the inhomogeneity of the first and second central moment of the density function (expectation and covariances). In Sect. 3 we use this time-variable AR process to construct a time-variable spatio-temporal covariance model for a DInSAR-SBAS point stack of surface displacements from ERS1 and ERS2 data from the Lower-Rhine Embayment in North Rhine-Westphalia. A summary and outlook concludes the work.

## 2 Time-Variable Autoregressive Process of First Order (TVAR(1))

The focus of this section is to derive the first and second central moments of a time-variable autoregressive process of first order, TVAR(1), which is defined by

where {*α*_{t}}_{Δt} ∈ℝ form a sequence of time-variable coefficients under the condition that the product series lim*t*→*∞*∏ *j*=1*tαj*2 converges. {*ℯ*_{t}}_{Δt} represents an independent and identically distributed (i.i.d.) sequence of random variables with expectation *E* {*ℯ*_{t}} = 0 and a constant variance Σ {*ℯ*_{t}} = *σe*2. Δ*t* denotes the sampling rate.

**Process Definition and Moving Average Representation of a TVAR(1) Process**

To find an equivalent representation of the TVAR(1) process by a moving average process, we have to substitute the past signals

and obtain the general representation

where S_{0} denotes the signal at the initial point *t* = 0. This results in compact form to

(notice: ∏ *j*=*t*+1*t* *α*_{j} := 1). With this moving average representation of a TVAR(1) process it is now straightforward to compute the expectation and the covariances of the process.

**Expectation of a TVAR(1) Process**

The expectation

depends on the expectation of the initial state S_{0} and the stochastic behavior of the noise ℰ_{t}, which is by definition (1) of the AR process *E* {ℰ_{j}} = 0 for *j* = 1, …, *t*. The expectation of the initial state S_{0} is unknown, but we can deduce from (7) the conditional expectation of S_{t} given a known initial condition *E* {S_{0}} = *s*_{0}

In the following we restrict this general formulation of TVAR(1) processes by assuming that S_{0} has the same stochastic properties as a long convergent AR(1) process with constant coefficient |*α*| < 1,

Taking the properties of the i.i.d. sequence of the random variables ℰ_{t} into account, we can state that S_{i−1} and ℰ_{i} are uncorrelated and due to the convergence behaviour lim*t*→*∞* *α*^{t} = 0, the expectation and variance of S_{0} is asymptotically independent of the initial state of this process and given by

cf. e.g. Box and Jenkins (1970, pp. 57-58). Applying the expectation *E* {S_{0}} = 0 in (7) or (8) this immediately results in

for the TVAR(1) process under the assumption (9) for S_{0}. This choice of the initial state has of course an influence on the further deviations of the variances and covariances. In contrast to Wegman (1974) where the second moments are defined as conditional moments we integrate the stochastic properties of the initial state S_{0} under assumption (10) of a long convergent AR(1) process.

**Variance/Covariance of a TVAR(1) Process**

The covariance as joint second central moment is defined by

where *t* and *t* + *h* denote the time points. Substituting the moving average representation (5) and noting that the expectation value vanishes due to (11) we obtain

A reordering with respect to the expectation and the products provides

Taking into account the properties of the i.i.d. sequence of the random variables ℰ_{t}, we can state

where *δ*_{kℓ} denotes the Kronecker-Delta.

From (14) one obtains

This can be reformulated to

which gives the covariance sequence of a TVAR(1) process for the times *t* and *t* + *h*, for a positive lag *h*. Note that because of the symmetry properties of covariances Σ {S_{t}, S_{t+h}} = Σ {S_{t+h}, S_{t}} holds.

The covariance sequence (17) can be split into two parts. The first part involving only the future events connected with the index *n* and the second part involving only the events before time *t* including the time itself linked to the indices *j* and *k*. The expression in the brackets defines the variance (i.e. lag 0) at time *t*

The variance *γ*_{t}(0) is influenced by two quantities. The variance of the initial value S_{0} characterizes the warm up behavior of the process and the constant variance of the process *σ*ℰ2 establish in connection with the time-variable coefficients *α*_{j} the further variance behaviour of the process. It is thus clear that the variance of the TVAR(1) process is time-variant (see also Fig. 1).

Often it is convenient to use a simple recursion formula instead of (18). If we rewrite (18) and use *t* − 1 as maximal upper bound instead of *t* we get

which can be rewritten to

Here the term in the brackets represents *γ*_{t−1}(0). Therefore we end up with the simple recursion equation

for the variances at time *t*, with the initial state *γ*_{0}(0) = *σ*S_{0}2. The covariances with respect to a time lag *h* follows from (17) and (18)

The covariance matrix of the TVAR(1) process of finite length *t*_{max} can now be computed by arranging the covariances Σ {S_{t}, S_{t+h}} for *t* = 0, …, *t*_{max} and *h* = 0, …, *t*_{max} − *t* into the upper triangle of a matrix. The lower triangle part is completed symmetrically accordingly. Figure 2 gives an example for the covariance matrix of a TVAR(1) process.

**Filter Representation and Covariance Matrix**

It should be mentioned, that the same covariance matrix can be derived from the filter approach (cf. Schuh and Brockmann (2020)). The covariance matrix consists of of the filter part and the warm up part

The filter matrix *H* is defined by

and the covariance matrix for the filter part follows from

(cf. Schuh and Brockmann (2020, Sec. 5)). Because of the warm up phase of the filter approach this covariance matrix must be modified with respect to the influence of S_{0} by the matrix which elements are computed according to (17) by

The complete covariance matrix of a TVAR(1) process results from the sum of the two parts according to (23) and is identical to the calculation of the covariance matrix by (17) or (21) and (22) respectively (see also Fig. 1).

## 3 Time-Variable Collocation of a DInSAR Point Stack

We apply these inhomogeneous covariances to model the temporal component of a spatio-temporal point stack derived from a DInSAR-SBAS analysis. The test region is the Lower-Rhine Embayment in North Rhine-Westphalia, Germany, with the still active open-cast mines Garzweiler, Hambach and Inden and the already closed coal mines Sophia-Jacoba in the mining region Erkelenz and Emil Mayrisch in the mining region Aachen. The Remote Sensing Software Graz (RSG) is used to analyze the data from the ERS1 and ERS2 mission. This results in a spatio-temporal point stack of surface displacements with respect to the initial frame 1992, May 5th up to 2000, Dec. 12th (cf. Esch et al. (2019)).

The construction of a time-variable spatio-temporal covariance model allows to use the least squares collocation approach to estimate the surface displacements at any place and at any time and provide a report on the uncertainty of this estimation.

When evaluating the deformations, the estimation error of the prediction should be minimized according to the Wiener-Kolmogorov-Principle. For this purpose, we consider the measured deformations as a special realization of a random process. Since the distribution function of this random process is unknown and no assumptions are to be made about it, we choose a linear approach via the principle of the **B**est **L**inear **P**redictor (BLP) (Teunissen 2007). Due to the pre-processing of the DInSAR data it can be assumed that the expected value of the signal (deformations) becomes zero over the entire area. This implicitly transforms the best linear predictor into the **B**est **L**inear **U**nbiased **P**redictor (BLUP) (cf. e.g. Schuh (2016, Sec. 3.2) or Teunissen (2007, Corollary I(i))). The BLUP corresponds to the Least Squares Collocation approach (cf. e.g. Moritz (1980, Sec. 11) or Schuh (2016)) and the predictor is defined by

where denotes the covariance matrix between the observed locations , whereas represents the covariance matrix between the observed locations and the location which are supposed to be predicted. reflects the noise characteristics. Here represents the observed displacements of the point stack. In this example 144.302 scatterers are identified in 64 time frames. The data points are clustered in urban regions. To achieve a homogeneous data distribution as well in urban regions as in rural regions the whole area is divided in [9*x*7] tiles and in each tiles the same number of points are randomly selected.

The huge computational effort to solve

for which the dimension follows from the number of measurements can be significantly reduced in case the covariances can be separated into a spatial and temporal domain and by the use of finite covariance functions (Schuh 1989).

**Spatial Covariance Model**

To make the covariances in space independent from the time we only consider the observed displacements of the same time difference to compute the spatial empirical covariance function. For each chosen time difference the empirical covariance function is computed and provides a sample of the stochastic behavior. All samples are documented in Fig. 3 (left). By plotting the confidence region for the estimates it can be stated that the spatial behavior is homogenous with respect to the time. These samples of empirical covariance functions are approximated by a finite covariance function which is constructed by the autocorrelation of truncated polynomial base functions (cf. Schubert and Schuh (2022)). The positive definite finite analytical covariance function can be seen in Fig. 3 (right). Due to the finite support of this positive semidefinite function the covariance matrix is sparse.

**Temporal Covariance Model**

The data characteristics in the time domain are characterized by the epochs of the available SAR recordings. Especially for the images of the ERS1 and ERS2 satellites the recorded data are irregularly distributed in time. From the variance plot in Fig. 4 (a) the time dependence of this signal is obvious. We approximate these variances by an equidistant TVAR(1) process with a sampling that is twice as high as the time difference of the ERS1 and ERS2 recordings. The time variation of the coefficients is modeled by a polynomial of degree three.

The variances of the TVAR(1) model can also be seen in Fig. 4 (a). The covariance matrix for all equidistant time points of the TVAR(1) model follows (22) and can be downsampled to the measurement epochs. The identification of the measurement dates is done by the nearest neighbors. Thus, we obtain the temporal covariance matrix at the identified measurement dates from the TVAR(1) model which is shown in Fig. 4 (b).

It should be mentioned, that the temporal covariance can be computed only for discrete times, but of arbitrarily small time intervals.

**Separable Spatio-Temporal Collocation Approach**

The above investigations have shown that the spatio-temporal covariance function can be separated into a time-variable temporal *γ*_{t}(*t*, *t* + *h*) and a homogeneous spatial component *γ*_{sp}( Δ*x*),

Since only permanent back scatterers, which are detected in all recordings, are included in the SBAS solution, the temporal distances are, however, the same for all scatterers. This allows for a compact representation of the covariance matrices by the Kronecker product

with *k* ∈{*p*, *o*} and (28) can thus be represented by

where characterizes the noise component. If the noise is designed appropriately, the calculations of the estimator can be split into a temporal and spatial component according to the rules of array algebra (cf. e.g. Blaha (1977); Rauhala (1974))

Here the observations are arranged in the matrix , each column represents the displacement of all scatterers for a specific epoch, i.e.

where *nosp* denotes the number of observed scatterers and *not* the number of recordings (time frames). The same rearrangement is done for the predicted values

with *npsp* the number of points to predict and *npt* the number of time frames to be predicted. According to the rules of array algebra the noise can be designed in two different ways without destroying the Kronecker structure, either

But in both cases the interpretation of the noise behaviour is not straightforward. A much more obvious choice for the noise would be

As shown in Schuh et al. (2022) an eigenvalue decomposition of into or into again gives a separable form for the prediction,

The great advantage of the collocation approach is that besides the predicted values, the accuracy of the prediction can also be determined by variance propagation (Moritz 1980, Sec. 17). Also these calculations can be separated into a temporal and spatial component (cf. Schuh et al. (2022)).

**Results of the Rigorous Collocation of a DInSAR-Stack**

Our test region is, as mentioned above the Lower-Rhine Embayment in North Rhine-Westphalia, Germany. For ERS1 and ERS2, the DInSAR-SBAS analysis results in a spatio-temporal point stack with 144.302 permanent scatterers in 64 time frames. The covariances are separated in a time-variant temporal component and a homogeneous space component. Since the data are available strictly at the respective recording times, a Kronecker representation of the covariance matrices is possible, which allows to split the calculations into a temporal and a spatial one. Thus, the numerical complexity of the task can be reduced significantly and it becomes possible to compute this very extensive collocation task on a workstation or notebook within about one to two hours.

With the collocation methods, surface deformations can be predicted for any location at any discrete time point. The tailored collocation approach elaborated here thus provides a continuous prediction in space for previously freely defined discrete time points. Beside the predicted values, their uncertainty is also quantified. In Fig. 5 the effects caused by groundwater management from the active opencast mines Garzweiler, Hambach and Inden are clearly recognizable by subsidence. Whereas in the already closed coal mines, an uplift is taking place. The accuracy (standard deviation) of the prediction is in a range 5–15 [mm] and it is immediately apparent that this accuracy is very heterogeneous. The bright points correspond to the measured permanent scatterers, while measurements in the vicinity are missing for the brown areas or they correspond to an extrapolation outside the image scene.

The orange line from the northwest to the southeast shown in Fig. 5 (up) marks a profile. Figure 6 (up) displays the behaviour of the displacement in time along this profile. Figure 6 (down) shows the displacement for the time span 1992.4 to 2000.9[yr] and their predicted accuracies. These are just a few examples to illustrate the many possibilities of the collocation approach. In Schuh et al. (2022) the patterns of movement in time are provided as an animation.

To study the benefit of the time-variable covariance model it will be now of interest to show the difference between modeling with a time-variable and a static temporal covariance function. As static covariance function, a Gaussian function with a standard deviation of 5[mm] and a half-value width of approx. 4[yr] years is fitted to the mean empirical covariances calculated from all ’training points’. To stabilize the temporal covariance matrix, an additional i.i.d. noise of with a standard deviation of 2[mm] has to be introduced. Using the residuals between predicted and measured deformations on 912 randomly selected test locations, the mean, standard deviation, RMS and maximum deviations for each time epoch is empirically determined and compared to the predicted formal standard deviations. Figure 7 summarizes the differences between static and time-varying modeling. While only minor differences can be observed in the predicted values, the predicted formal variances show a significantly different behavior. While in the static case the variances are constant over time, the time-variable modeling shows a steady increase in the variances in line with the empirical values. In contrast to static modeling, the time-variable formulation thus results in better consistent behavior between the model and the data and opens up further possibilities to fit the model even better to the data.

## 4 Summary and Outlook

In addition to the many advantages of the collocation method as a data-adaptive method, it is repeatedly stated that there is a lack of flexibility in the modeling of the covariances and that the models cannot be implemented due to the enormous computational effort. In this work, we demonstrate that these limitations can be overcome by appropriate methodological approaches. The advantages of the collocation method can be even used in case of time-varying behavior and extensive measurement points. Since the collocation method can be used to estimate function values and their uncertainties for arbitrary locations and times, this method is also very well suited for the fusion of SAR data with other data, e.g. epoch-wise levelling campaigns, which will be investigated in the future.

## References

Blaha G (1977) Least squares prediction and filtering in any dimension using the principles of array algebra. Bull Géodésique 51:265–286. https://doi.org/10.1007/BF02525104

Box G, Jenkins G (1970) Time series analysis forcasting and control. Holden-Day, San Francisco

Brockwell PJ, Davis RA (1991) Time series: theory and methods, 2nd edn. Springer, Berlin. iSBN 1-4419-0319-8

Esch C, Köhler J, Gutjahr K, et al (2019) 25 Jahre Bodenbewegung in der Niederrheinischen Bucht - Ein kombinierter Ansatz aus D-InSAR und amtlichen Leitnivellements. Zeitschrift für Geodäsie Geoinform Landmanag 144(3):173–186. https://doi.org/10.12902/zfv-0257-2019

Knopp K (1922) Theorie und Anwendung der unendlichen Reihen. Verlag Julius Springer, Berlin

Korte J, Schubert T, Brockmann JM, et al (2022) On the estimation of time varying AR processes. In: X Hotine-Marussi-Symposium, IAG Symposia. No. (this issue) in lecture notes in earth science. Springer, Cham

Moritz H (1980) Advanced physical geodesy. Wichmann, Karlsruhe

Priestley M (1989) Non-linear and non-stationary time series analysis. Academic Press, Amsterdam

Rauhala U (1974) Array algebra with applications in photogrammetry and geodesy. Thesis, Stockholm: Royal Inst. of Techn.; Fotogrammetriska meddelanden, No. 6

Schubert T, Schuh WD (2022) A flexible family of compactly-supported covariance functions based on cutoff polynomials. In: X Hotine-Marussi-Symposium, IAG Symposia. No. (this issue) in lecture notes in earth science. Springer, Cham

Schuh WD (1989) Kollokation - zu rechenaufwendig? ZAMM, Z angew Math Mech 69 4:T73–T75. http://onlinelibrary.wiley.com/doi/10.1002/zamm.19890690403/pdf

Schuh WD (2016) Signalverarbeitung in der Physikalischen Geodäsie. In: Freeden W, Rummel R (eds) Handbuch der Geodäsie. Springer Reference Naturwissenschaften, vol Erdmessung und Satellitengeodäsie. Springer, Berlin, Heidelberg, pp 73–121. https://doi.org/10.1007/978-3-662-47100-5_15

Schuh WD, Brockmann J (2020) Numerical treatment of covariance stationary processes in least squares collocation. In: Freeden W, Rummel R (eds) Mathematische Geodäsie/Mathematical Geodesy. Springer Reference Naturwissenschaften. Springer, Berlin, Heidelberg, pp 933–966. https://doi.org/10.1007/978-3-662-46900-2_95-1

Schuh WD, Korte J, Schubert T, et al (2022) Modeling of inhomogeneous spatio-temporal signals by least squares collocation. In: X Hotine-Marussi Symposium, Milano, oral presentation, 16.6.2022. https://doi.org/10.48565/bonndoc-39

Teunissen PJG (2007) Best prediction in linear models with mixed integer/real unknowns: theory and application. J Geodesy 81(12):759–780. https://doi.org/10.1007/s00190-007-0140-6

Wegman EJ (1974) Some results on non stationary first order autoregression. Technometrics 16(2):321–322. https://doi.org/10.1080/00401706.1974.10489190

## Acknowledgements

This research is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Grant No. 435703911 (SCHU 2305/7-1 *Nonstationary stochastic processes in least squares collocation–NonStopLSC*). The authors acknowledge the European Space Agency (ESA-Project ID 17055 *Integrated Modelling of SAR Interferometry and Leveling*) for provision of the ERS 1/2 data. The DInSAR-SBAS evaluation was done by the Remote Sensing Software Graz (RSG). The authors gratefully acknowledge the granted access to the Bonna cluster hosted by the University of Bonn.

## Author information

### Authors and Affiliations

### Corresponding author

## Rights and permissions

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## Copyright information

© 2023 The Author(s)

## About this chapter

### Cite this chapter

Schuh, WD., Johannes, K., Till, S., Jan Martin, B. (2023). Modeling of Inhomogeneous Spatio-Temporal Signals by Least Squares Collocation. In: International Association of Geodesy Symposia. Springer, Berlin, Heidelberg. https://doi.org/10.1007/1345_2023_202

### Download citation

DOI: https://doi.org/10.1007/1345_2023_202

Published:

Publisher Name: Springer, Berlin, Heidelberg