**2.1 Measurements**

The results of the actual measurements of laser systems for the remote-sensing of the atmosphere are used to verify the proposed approach. Currently, the laser systems for remote-sensing use high-power pulsed lasers, and the backscattering signal is written with a certain sampling step corresponding to the required spatial resolution. Moreover, the growth of the length of the track-sensing leads to a disproportionate growth of the power source and the dynamic range of the incoming signal. It also causes the multiple scattering effects which can be difficult to take into account.

The new approach is based upon the use of a low-power radiation source (for example, a source of white light) within the specified parameters of the gating. The dark pulse of the continuous light source has a duration equal to usual laser pulse lidar (about 10-8 c). The time interval between the dark pulses is close to the time of the radiation propagation in an area where we can neglect the multiple scattering. The digitisation of the remote-sensing signal can be performed with standard digital systems (a constant gate) as well as with systems based on the proposed approach (an increasing gate).

I propose to restore the average characteristics of the medium to long sections of a length close to the length of the track measurements. This will significantly increase the accuracy of the reconstruction of the properties of the real heterogeneous medium. Signal processing assumes the creation of the registration system with an increasing time-step gate of the incoming signal (the one-dimensional case).

The comparative calculation of the required radiation power was held for a given signal/noise ratio for different average atmosphere extinction coefficients σ = 10-2,..., 1 km-1 (the old and new systems).

For high transparency (σ = 10-2 km-1), the maximum length of the zone of measurement is chosen equal to the length of the layer of a dense atmosphere, significantly affecting the scattering signal (Lmax = 30 km). To muddy the atmosphere, this distance is set by the condition that the optical depth does not exceed τ = 2σL = 10. This allows us to consider the scattering of the signal with an accuracy of 0.05% and to neglect the signal over large distances. Single scattering occurs with the condition τ = 2σL <3 (Kovalev, V. A., 1973; Ablavskij, L. M. and Kruglov, P. A., 1974). The calculations were made on the assumption of single scattering. The data obtained is used only so as to illustrate the detected trends (τ> 3).

All estimates are carried out based on an expression derived from the lidar equation for systems I and II (Polkanov, Y. A. and Ashkinadze, D. A., 1988):

$$\mathbf{n}\_{\mathrm{i}} = \mathrm{AW}\_{\mathrm{i}} \left( \mathrm{e}^{\circ 2a \mathrm{i}i} \left( \mathrm{1} \cdot \mathrm{e}^{\circ 2ab} \right) \right) / \mathrm{L}^2 \tag{1}$$

$$\mathbf{m}\_{\rm i1,i2} = \mathbf{A} \mathbf{W}\_2 \sum\_{\rm i \vdash l\_\star}^{\rm l\_{1,2}/l\_\star} \left( \mathbf{e}^{-2\sigma \rm li} \left( \mathbf{1} \, \mathbf{1} \, -\, \mathbf{e}^{-2\sigma \rm lc} \right) \right) / \left( \mathbf{l}\_\mathbf{i} + \mathbf{l}\_\star / \, \mathbf{2} \right)^2 \tag{2}$$

Where ni, ni1,i2 - obtained discrete values from the scattering signal (number of photon counts); A – a coefficient which brings together the supporting equipment characteristics; W1,2 – the power of the laser radiation; L – the distance from the centre section of the route, by which the signal is recorded; li - the distance from the system to this site; ls – the length of the section; lT the length of the shadow zone of the lidar where the signal is not recorded (600 m). We assume for the system that II L2 > L1, (ni2 - ni1) = ni. An advanced assessment of the relative measurement error of the signal (δi, δix for System I and System II) was conducted on the basis the expressions (Polkanov, Y. A. et al., 1985; Polkanov, Y. A. et al., 2004):

$$\delta\_{\rm i} = \mathfrak{t}\_{\beta}((\mathfrak{n}\_{\rm i} - \mathfrak{n}\_{\rm n})^{1/2}) / \ n\_{\rm i} \tag{3}$$

$$\delta \mathbf{S}\_{\rm ix} = \mathbf{t}\_{\beta} ((\mathbf{n}\_{\rm ix} - \mathbf{2} \mathbf{B}\_{\rm x} \mathbf{n}\_{\rm n})^{1/2}) / \ \mathbf{n}\_{\rm ix} \tag{4}$$

Where tβ – the coefficient equal to the probability of the matching error computed to its actual value (if tβ = 2, the probability is equal to 0.95). The necessity of this evaluation is due to the appearance depending δix (t) for system II (signal/noise = const). This is due to the progressive rise in the value of the time intervals recording the scattering signal (with the digitisation step – ts). The level of background illumination takes into account the introduction of the coefficient B = f (t) in (4). The measurement error for individuals counts the signal and background-level measurement errors, becoming comparable for large intervals of TS. They are significantly higher than the level of internal noise (in. ns.) receiving system (in this case, n in.ns ~ 0.1, ts = 0,4 ms). Moreover, the summed value of the signal increases to a certain point in time, reaching a maximum level of accumulated signal (Kovalev, V. A., 1973; Ablavskij, L. M. and Kruglov, P. A., 1974). However, the level of background illumination increases linearly with time. The calculations used the results of the actual measurement system I (ni, nb, σ). The coefficient A in (1) is also evaluated and used in subsequent calculations for the system II (2).

#### **2.2 Processing**

218 Remote Sensing of Planet Earth

structure is directly dependent upon the thermodynamic stability of the environment. Changes within the structure of the inhomogeneities are more mobile and are preceded by changes in the thermodynamic state of the environment as a whole. We take this as an axiom. As such, the structure of the inhomogeneities is central to the prediction of processes within the environment. This becomes especially important during the development process of a catastrophic scenario. Their nonlinear nature makes standard methods for the analysis of irregularities ineffective because of the number of initial assumptions, which often only apply to the environment in the classical sense. Therefore, I propose a structural-statistical

The results of the actual measurements of laser systems for the remote-sensing of the atmosphere are used to verify the proposed approach. Currently, the laser systems for remote-sensing use high-power pulsed lasers, and the backscattering signal is written with a certain sampling step corresponding to the required spatial resolution. Moreover, the growth of the length of the track-sensing leads to a disproportionate growth of the power source and the dynamic range of the incoming signal. It also causes the multiple scattering

The new approach is based upon the use of a low-power radiation source (for example, a source of white light) within the specified parameters of the gating. The dark pulse of the continuous light source has a duration equal to usual laser pulse lidar (about 10-8 c). The time interval between the dark pulses is close to the time of the radiation propagation in an area where we can neglect the multiple scattering. The digitisation of the remote-sensing signal can be performed with standard digital systems (a constant gate) as well as with

I propose to restore the average characteristics of the medium to long sections of a length close to the length of the track measurements. This will significantly increase the accuracy of the reconstruction of the properties of the real heterogeneous medium. Signal processing assumes the creation of the registration system with an increasing time-step gate of the

The comparative calculation of the required radiation power was held for a given signal/noise ratio for different average atmosphere extinction coefficients σ = 10-2,..., 1 km-1

For high transparency (σ = 10-2 km-1), the maximum length of the zone of measurement is chosen equal to the length of the layer of a dense atmosphere, significantly affecting the scattering signal (Lmax = 30 km). To muddy the atmosphere, this distance is set by the condition that the optical depth does not exceed τ = 2σL = 10. This allows us to consider the scattering of the signal with an accuracy of 0.05% and to neglect the signal over large distances. Single scattering occurs with the condition τ = 2σL <3 (Kovalev, V. A., 1973; Ablavskij, L. M. and Kruglov, P. A., 1974). The calculations were made on the assumption of single scattering. The data obtained is used only so as to illustrate the detected trends

method for analysing the structure of inhomogeneities.

effects which can be difficult to take into account.

incoming signal (the one-dimensional case).

(the old and new systems).

(τ> 3).

systems based on the proposed approach (an increasing gate).

**2. Methodological approach** 

**2.1 Measurements** 

The following processing scheme was assumed: the initial signal (as a time function) the generalised structure of a signal an elementary cell of the signal structure. The multiplication of such cells allows the complete restoration of the characteristic structures in the supervised space.

The indicator of the time stability of the signal structure was the dispersion of the components of the elementary cell of a signal structure. If the dispersion exceeds an interval between elements of the revealed cell then the structure is unstable. The correlation of the generalised frequency structure of a horizontal signal and the generalised parameter which fixes the thermodynamic stability of the environment is a characteristic sign of the selforganising of the environment.

Looking at Remote Sensing the Timing of an

useful signals on an equal basis.

digitised signal (∂II ≠ const ≤ ∂I = const).

situations is provided by Table 2.

the measurement error δ = 10% (const).

type II.

Organisation's Point of View and the Anticipation of Today's Problems 221

essential point here is the rise of the level of the recorded background illumination with the increasing duration of the strobe. The scattering signal increases from the strobe to strobe –

The background illumination level is significantly higher than the corresponding internal noise receiver (in.ns = 0.1 for ti = 0.4 ms). It exceeds the signal of system I, with a point, but is comparable with the level of the scattering of the signal of system II (τ < 3). In this case, the accuracy of the scattering signal and the background are similar, and they can be used as

In fact, we have a mixture of two signals - the scattering signal and the background signal. Their value increases from strobe to strobe and the first of them (S) rises to a certain level

In these circumstances, the accuracy of the scattering signal increases when S/N = const (1

Thus, there is a new dependence - ∂ (t) which was previously unavailable for system I . Table 1 lists the measurement error depending upon the distance ls (n) corresponding to the

**L (km) 1 2 3 4 5 10**  δ,% 3,7 1,7 1,3 1,2 1,1 0,8

Model calculations showed that the measurement accuracy of the scattering signal for system II is several times higher than the measurement accuracy for system I. This means that for the same radiation power of remote systems, greater measurement accuracy is achieved for systems of type II through special time organisation and its recording of the

We can talk about the actual incompleteness of the concept of the signal/background ratio for the registration systems of type II when the strobe length (a single reference signal) depends upon the position of the laser pulse on a remote line sensing. Moreover, it is possible that the signal/background ratio is less than unity but that the measurement accuracy remains high. This is possible when the signal/internal noise ratio (S/in.ns) and the background/internal noise ratio (b/in.ns) is much higher than 1. An example of such

**L(km) 1 2 3 4 5 10**  S/N 1,23 0,81 0,65 0,55 0,49 0,33 Table 2. Signal/background ratio, depending upon the length of the strobe (km) and where

The obtained simulation results suggest that the measurement accuracy was higher than expected, if only to carry out the calculation of the signal/background ratio for systems of

in general – to the so-called maximum accumulated signal (Kovalev, V. A., 1973).

(Wmax) whilst the second of them (b) increases linearly with time and indefinitely.

because a strobe the length of the time of registration is increasing.

interval gating ts (n), if S/N = 10 = const, for σ = 0.1 km-1, nb = 50, ti = 0.4 s.

Table 1. The measurement error decreases with increasing interval gating.

The basis for the reception of new results is a series of works on the laser sounding of the atmosphere in stable nightime conditions. This has allowed the development of certain methods for the structural-statistical processing of an initial remote signal. The aim is to reveal the signs of the steady organisation of the frequency structure of environmental inhomogeneities.

The generalised regular structure comes from the summary of the sequence of the discrete readouts. They were received by the scanning of the investigated volume of the environment in a horizontal plane to a set of directions and with the set angular permission (Polkanov, Y. A. et al., 1989).

During the following stage, the signal is represented in the form of a regular structure of local maxima and minima. There was a separate analysis of the 'plus' and 'minus' structures (Polkanov, Y. A. and Kudinov. V. N., 1989).

These components behave as whole object and are registered as a uniform regular structure (type harmonious) only in the case of a steadily vertical stratified environment. When the infringement of the stability of the stratification of environmental communication between the 'plus' and 'minus' structures decreases, they become increasingly independent of one another other. The degree of such dependence can be characterised by a certain numerical parameter (Polkanov, Y. A. et al., 1991; Polkanov, Y. A. et al., 2009).

The thermodynamic stability of the environment and its stratification can be characterised numerically by a special generalised parameter on the basis of Richardson's number. With the infringement of the thermodynamic stability of the environment, this parameter adopts wavy characteristics on a vertical plane. The length of such a 'wave' with the falling of the environmental stability was decreased.

The integrated regular structure of vertical thermodynamic distribution is an indicator of such stratification of the environment.

It is possible to speak about the communication of the optical structure horizontal stability with the vertical stability of the thermodynamic structure of the environment and its stratification as being an indicator of such stability (Polkanov, Y. A. et al. 1989).

Besides this, the infringement of the stability of the environment leads to the infringement of the stability of the revealed structure and the occurrence of obvious anomalies within the structure (Polkanov, Y. A. et al., 1991; Polkanov, Y. A. et al., 2008) whose behaviour can provide information on the direction of the reorganisation (self-organisation) of the environment.
