Remote Laser Induced Fluorescence of Soils and Rocks

Abstract

The laser induced fluorescence spectroscopy was systematically utilized for remote sensing of different soils and rocks for the first time, to the best of our knowledge. Laser induced fluorescence spectroscopy measurements were carried out by the developed nanosecond LIDAR instrument with variable excitation wavelength (355, 532 and 1064 nm). LIDAR sensing of different Brazil soil samples have been carried out in order to construct a spectral database. The laser induced fluorescence spectra interpretation for different samples has been discussed in detail. The perspectives of LIDAR sensing of organic samples deposited at soils and rock have been discussed including future space exploration missions in the search for extraterrestrial life.

1. Introduction

Extensive anthropogenic activities and fast growth of industry resulted in harsh exploitation of environment. Nowadays, ecological problems became very important so environmental pollution monitoring in real time is crucial for early detection of ecological problems to decreasing the possible negative impact. Specifically, anthropogenic activities are responsible for numerous cases of degradation and damage of soils [1] due to both harsh exploitation by farmers as well as to leaks of hazardous compounds (liquid and gas) from nearby industrial areas. Development of new techniques and instrumentation for monitoring atmosphere, water objects, soils and ground vegetation is of high demand [2,3,4,5,6]. The laser remote sensing is a powerful technique for express/online diagnostics of distant targets. The LIDAR (Light Detecting and Ranging) is a key instrument for laser remote sensing and such instruments utilize pulsed lasers to interact with the remote target and induce elastically and inelastically backscattered photons which provide information on some properties of the remote target. In 1970–1990s, LIDAR instruments were rather bulk and heavy (>0.5 m³ and >50 kg) so they were installed at stationary platforms so the range of possible objects of interest was rather limited: in the majority of cases it was lower and upper atmosphere sensing. LIDAR systems were also installed onboard of helicopters or airplanes but high cost of such aircraft vehicle exploitation resulted in very limited LIDAR applications: scientific studies, military applications or proof-of-concept demonstrations. In late 2010s, cost-saving unmanned aircraft vehicles (drones) capable of carrying up to few kg became available. Nowadays, drone exploitation costs are low, so compact and sensitive LIDAR systems are of high interest for numerous applications such as ecological monitoring or plant field diagnostics in agriculture [4,5,7,8,9,10,11,12,13]. LIDAR measurements are also of high interest for satellite remote sensing since satellite data needs calibration for reliable interpretation for ocean and vegetation mapping [8,9,14,15,16,17,18].

Numerous objects have been extensively studied by LIDARs [4,9,14] but laser remote sensing of soils has not been systematically implemented so far [19,20,21]. Soils alteration by harsh exploitation or contamination cause biochemical changes in vegetation and hence quality of the produced food. Due to dynamic nature of soil contamination express and sensitive techniques are required for early monitoring in order to prevent intense or even inevitable problems. Fluorescence spectroscopy of soils has been extensively studied under laboratory conditions [22,23,24,25] and it was demonstrated that different pigments can be identified. Thus, soil microbiological activity and organic matter accumulation can be estimated and soil fertility can also be measured. In the present paper, we have developed a fluorescence LIDAR system for the remote sensing of soil and water surfaces as well as green vegetation from mobile carriers (helicopters, automobiles and ships). Both laboratory and field experiment results are discussed to illustrate the perspectives of laser remote sensing for express diagnostics of soils.

2. Experiment Setup

A scheme of the developed LIDAR system is presented in Figure 1. A specially designed Nd:YAG laser (pulse duration 8 ns, pulse repetition rate up to 25 Hz, fundamental harmonic (λ = 1064 nm) pulse energy 600 mJ, second harmonic (λ = 532 nm) pulse energy 280 mJ, third harmonic (λ = 355 nm) pulse energy 100 mJ) became the core element of the LIDAR instrument. We utilized custom design frequency converters based on potassium dihydrogen phosphate (KDP) crystals to generate the required wavelength while colored glass filters cleaned up the second and third harmonics. The laser was specifically developed for harsh exploitation with strong vibrations to provide long-term stable generation while operating the laser onboard of an aircraft. The laser beam was directed by an aluminum coated plane mirror to the remote target (ground, water or vegetation). The backscattered photons were collected by the same mirror and reflected to the telescope (200 mm in diameter, aperture of 1/3) which focused the irradiation to the input slit of the polychromator (assembled as an autocollimator). The different diffraction gratings (300 and 1200 lines per mm) can be used depending on required spectral resolution. In the first diffraction order of the polychromator (300 lines/mm grating) the single-measurement spectral window was 525 nm with 0.5 nm spectral resolution, which was sufficient to record broadband fluorescence spectra of soils. By rotating the spectrograph’s diffraction grating spectral regions of interest can be chosen. In order to protect the detector from elastically scattered laser pulse we utilized notch filters installed at the polychromator input. Polychromator can be also used at higher orders for spectral gratings (up to 9th). In such case spectral resolution will be 0.022 nm per channel, so fine spectral lines in laser plasma can be distinguished, thus elemental analysis by laser induced breakdown spectroscopy can also be carried out [26]. A sensitive gated linear CCD camera (1024 pixels) was utilized for spectra quantification in 350–900 nm spectral range. The camera’s intensifier provided strong amplification up to 3 × 10⁴ so every 10 captured photons were quantified as one digital count. The minimum gate can be set to 10 ns thus the LIDAR can be exploited at any solar illumination conditions. The G-56 pulse generator was used to synchronize the Nd:YAG laser triggering and gated camera acquisition while a personal computer captures and stores the backscattered spectra. Based on the developed LIDAR parameters we estimated that ground areas about 1 m² can be sampled sequentially from a height of 300–500 m.

The developed LIDAR was designed to be utilized during field campaign in different regions of Brazil. In order to test the instrument in the laboratory we have measured a few hundred different soil and rock samples originating from various regions of Brazil. Measurements were performed under daylight conditions since gated laser induced fluorescence spectroscopy acquisition (gate 10–50 ns) were not affected by the sunlight.

3. Results

Examples of the fluorescence spectra induced by 355 nm excitation for soil samples are presented in Figure 2. For comparison, return signal spectra of seawater and ground vegetation are also shown. The fluorescence spectra of all the studied samples have characteristic peaks in blue and green bands. A systematic study of more than 500 samples including soils, vegetation and seawater has been carried out to construct laser induced fluorescence spectra database. Laser induced fluorescence spectra contain different characteristic bands in blue, green and red spectral regions depending on the sample. Concerning soil sample bands in fluorescence spectra were attributed to amino acids, nucleic acids and some of the coenzymes: the reduced form of nicotinamide adenine dinucleotide (NADH) and its phosphate (NADPH), flavins (flavin mononucleotide and flavin adenine dinucleotide) and the vitamins [27]. These compounds have a high fluorescence quantum yield so they are most intense in the spectra. Furthermore, phytofluene (hexahydrolycopene, partially hydrogenated pigment from carotenoid group) and perylene (aromatic hydrocarbon of five condensed benzene rings) also contribute to the blue band. The latter compound is found in arid soil and volcano eruption products [28,29]. The green spectral band was attributed to 1,4-dihydroxyanthraquinones [30], which are of microscopic fungi origin always presented in soils.

These results clearly demonstrated that the developed LIDAR can detect different organic compounds in soils, sediment rock, natural water, including living microorganism cells, green leaves of plants and sea plankton [10,31,32]. Preliminary data on the analyzed fluorescence spectra of soil, seawater, vegetation and synthetic samples are summarized in

Table 1. Fluorescence maxima of natural and synthetic samples, measured by the fluorescence LIDAR (λ = 355 nm excitation).

No.	λ, nm	Sample Characteristic
1	475–480	6,7-dimethyl pterin adsorbed on kaolinite
2	475–480	NADH on kaolinite
3	475–480	Complex of abiogenic pterins (8:3:1, 185 °C, 8 h)
4	475–480 and 500–520	Same, but 6 h
5	475–480	X anthopterin from mixture 8:3:1 (6 h)
6	480–520, broad band	Pterins + flavin in a wet synthetic product (8:3:1, 8 h)
7	550	Riboflavin on kaolinite
8	530	1,4-dihydroxyanthraquinones from Penicillium funiculosum (on kaolinite)
9	475–480	Lipid extract from brown oak leaves
10	475–480	Dry leaf (powder) of Amaranthus cruentus
11	475–480 and 500–520	Green leaf (NADH+pterins)
12	475–480 and 500–520	Dark soil, Moscow Region (pterins)
13	480–520, wide band	Pleistocene algal fossil rock, Crimea, 100,000 yr
14	475–480	Juvenile ashes of the Tyatya Volcano eruption of 1973 (perylen)
15	520 and 550	Latosole from the Amazon area, Brazil
16	475 and 550	Green leaf Pao Brazil
17	475 and 550	Seawater, Brazil

. According to the presented data, both pterin and flavin compounds can be identified, but these two classes of coenzymes are principally important for living organisms. Therefore, this instrument can be employed for the remote sensing search of life traces (chemical evidence “prints” of ancient life in Precambrian rock, organic materials of space origin etc.) in future extraterrestrial missions.

In order to estimate the capability of LIDAR “life markers” detection the following experiment has been carried out. Two 1 cm² areas were sampled by the LIDAR for the ancient fern print at rock plate (see Figure 3a): inside and outside it. The distance between their centers did not exceed 1.5 cm. Figure 3b shows the corresponding laser induced fluorescence spectra normalized to the return signal at λ = 355 nm (elastic scattering of the probing beam). Preliminary experiments have revealed that the spectra do not vary significantly within a probed area both outside and inside the print. However, Figure 3b exhibits an additional signal intensity in the range 540–560 nm when probing the relic print. This spectral band enhancement for the fern print can be explained by the condensed π-electron aromatic structures whose analogs are soil humins and rock kerogens. Detection and selection of such a signal during laser remote sensing of paleoarcheologic places of interest will indicate the areas which should be extensively searched in order to reveal the traces of ancient vegetation (and civilization). Of particular interest is that the same technique can be utilized for life markers detection during extraterrestrial missions on Mars, Venus or Jupiter’s moon Europe [33,34,35,36]. It should be noted that fluorescence spectra features transform with time as the organic pigment molecules undergo decomposition. Laser induced fluorescence spectra features comparison can be potentially applied as a remote technique for estimating organic pigments age.

Table 2. Fluorescence compounds (pigments) in soil and their spectral characteristics.

Pigment	Excitation, nm	Emission, nm	Solvent
Benzene rings (anthracenes) and heterocycles	260–280	340–360	Benzene, ethanol
Phytofluene (hexahydrolycopene)	350–370; in the excitation spectrum are bands at 310, 350 and 370 nm	415	Hexane
Perylene	420; in the spectrum are bands at 370, 385, 405 and 434 nm	440 and 465	Hexane, heptane
Flavonols (3-hydroxyflavones)	370	470–520	Ethanol
	440	520–540	Ethanol + H3BO3 + citrate
2-hydroxy-5-methyl-1,4-benzoquinone	370–390	480–500	Ethanol; in 0.1N KOH absorbs at 510 nm
Leucopterin	355	430	Ethanol, pH 9.0; (2-amino-4,6,7-trihydroxypteridine)
Xanthopterin, isoxanthopterin (2-amino-4,6-dihydroxypteridine and 2-amino-4,7-dihydroxypteridine)	375	460–480	Same
NADH and NADPH in living microorganism cells and leaves	340	455, 480	Water, pH 7.0
Flavins (riboflavin, lumiflavin, lumichrome; flavin mononucleotide and flavin adenine dinucleotide)	370 (or 450)	500–530	Ethanol, CHCl3
1,4-dihydroxyanthraquinones	370	500–540	Ethanol CH3 COOH
	500	640–660	A complex with (CH3COO)2Mg in ethanol
Chlorines (related to chlorophylls)	420	672, 715	Ethanol
Phycoerythrins	500	580–630	Water
Phycocyanobilins	600	660–715	Water
Phycobilins-655	580–600	665	2,6-lutidine, amyl alcohol
4,9-dihydroxyperylene-3,10-quinone	500–520	660–670	In concentrated H2SO4

summarizes the results for laser induced fluorescence spectroscopy with different excitation wavelengths. According to the presented table, excitation by 355 nm wavelength is not optimal in some cases for soil component identification. For example, detection of tetrapyrrole pigments (chlorophylls, phycoerythrins and phycocyanins) is not reliable for 355 nm excitation since these pigments have a low quantum yield at this wavelength and their emission can be totally overlapped with that for flavins and 1,4-dihydroxyanthraquinones. However, in this case, the second harmonic (λ = 532 nm) of the Nd:YAG laser can be efficiently used as a probing beam as it was previously demonstrated for laser remote sensing of the upper ocean [3,4].

Undoubtedly, humic materials (humic and fulvic acids) also contribute to the soil fluorescence spectra. However, their emission is masked by that from vegetative and microbial pigments if they exist in the sample as it was shown previously for humic acids (HA) and their fractions fluoresce in alkaline solutions [37,38,39]. According to our data, HA (purified from low-molecular fluorescent impurities) have a blue-green emission with maxima at 465 and 540 nm wavelengths. Nearly the same spectral characteristics are inherent to soluble vegetative lignine and model humin-like polymers [40]. Fluorescence of humic acids and humin is related to the excitation of π-electron systems in aromatic condensed rings whose amount can even increase during sample pyrolysis [41].

Intense blue-green fluorescence bands in the studied soil samples are attributed to pterin [41] and flavin [42,43] coenzymes, which had a fluorescence similar to that for Euglena gracilis seaweed [44]. The LIDAR was able to detect pigments immediately onsite in its native form. The conventional laboratory techniques require time consuming sampling, transportation, chromatographic separation and purification, so the original sample can be slightly or moderately damaged or modified during this period. For example, the fluorescence spectra of pigments in native form have a moderate red shift (by 10–20 nm) compared to their spectra in an organic solvent [45,46,47,48,49]. The “host” effect can affect not only the afterglow maximum position but also the quantum yield, when the pigment system of cells exhibits intermolecular migration of excitation. For instance, if the samples were not properly transported and handled then fluorescent compounds can transfer the absorbed quantum energy to chlorophyll, exciting or amplifying its fluorescence. The model membrane-like structure containing phycobilin and flavin pigments exhibits the transfer of light energy absorbed by phycobilin in the red band to the flavin reactivity center [44]. These effects can complicate the interpretation of fluorescence spectra, therefore additional chemical analysis is required in some cases.

4. Conclusions

The principle of LIDAR instrument is based on remote sensing by laser photons which induce fluorescence and backscattered photons that provide information on the object of interest. Here, we presented a multi-wavelength pulsed LIDAR instrument which was used for a systematic study of laser induced fluorescence of different soils. Due to high sensitivity and optimal choice of pumping wavelength, the developed LIDAR is able to identify the majority of most important soil pigments as well as cell pigment systems (pterins and flavins) allowing one acquiring virtually instantaneously the information on pigments in natural objects. We have demonstrated that the LIDAR instrument is an effective and powerful tool for remote monitoring of soil and water contamination by different agents (oil and oil products, organic industrial production pollutants, heavy metals, etc.). Alternatively, the LIDAR technology is perspective for searching ancient organic matter deposits and in life traces detection in ongoing and future extraterrestrial missions [50,51]. It should be noted that the two color (in/outside Chl-a absorption band) analysis of Mars images acquired by satellite multispectral cameras is very informative for preliminary estimation of perspective landing sites with organic pigments [52,53] and surface humidity [54]. A new generation of compact eye-safe LIDAR instruments [55,56,57,58] is of growing interest for development of numerous applications and control systems which can be safely exploited in inhabited places (city streets, airport, train station, etc.) without any risks of eye damage.