- Open Access
Passive kHz lidar for the quantification of insect activity and dispersal
© The Author(s) 2018
- Received: 19 October 2017
- Accepted: 6 April 2018
- Published: 30 May 2018
In recent years, our group has developed electro-optical remote sensing methods for the monitoring and classification of aerofauna. These methods include active lidar methods and passive, so-called dark-field methods that measure scattered sunlight. In comparison with satellite- and airborne remote sensing, our methods offer a spatiotemporal resolution several orders of magnitude higher, and unlike radar, they can be employed close to ground. Whereas passive methods are desirable due to lower power consumption and ease of use, they have until now lacked ranging capabilities.
In this work, we demonstrate how passive ranging of sparse insects transiting the probe volume can be achieved with quadrant sensors. Insects are simulated in a raytracing model of the probe volume, and a ranging equation is devised based on the simulations. The ranging equation is implemented and validated with field data, and system parameters that vary with range are investigated. A model for estimating insect flight headings with modulation spectroscopy is implemented and tested with inconclusive results. Insect fluxes are retrieved through time-lag correlation of quadrant detector segments, showing that insects flew more with than against the wind during the study period.
The presented method demonstrates how ranging can be achieved with quadrant sensors, and how it can be implemented with or without active illumination. A number of insect flight parameters can be extracted from the data produced by the sensor and correlated with complementary information about weather and topography. The approach has the potential to become a widespread and simple tool for monitoring abundances and fluxes of pests and disease vectors in the atmosphere.
- Near-field optics
- Remote sensing
- Dark field
- Modulation spectroscopy
Insects are a diverse group of animals with a large impact on human society. They cause a large amount of human deaths annually  and effect significant economic damage in forestry  and agriculture . Pesticides employed to combat this can have severe health effects for humans  and pollinators . Insect activity and movement patterns occur on fast timescales , are complex and species specific , and can be highly localized . As such, a high spatiotemporal resolution is required to monitor their movements, activity patterns, and interactions.
A number of methods have been developed to investigate the phenology of insects in situ. Vehicle-mounted sweep nets , human landing catch, electrocuting grids , and insect traps  are commonly employed for directly monitoring insect behavior and abundances. These methods can provide rich information on insects, but are laborious and yield relatively low counts. Air- and satellite-borne topographical remote sensing methods, including imaging and light detection and ranging (lidar), are commonplace. In these, indirect observations of insect activity are made through correlation with the profiled vegetation structure [12, 13]. By relying on the passage of aircraft or satellites, these indirect methods are restricted to a time resolution in the order of days or weeks, and satellite imaging typically has a spatial resolution of 30 × 30 m2. Radar entomology is a method wherein direct observations of insects are made through the transmission and measurement of backscattered microwave pulses [14–16]. Radar systems have a demonstrated species classification capability , but are unable to monitor horizontally close to ground due to clutter effects.
Lidar entomology is an emerging field in which laser transects across the landscape are monitored [18–20]. We have experience recording over a hundred thousand insect observations per hour , and species classification through modulation spectroscopy of insect wing beats has been demonstrated [11, 22–24]. Entomological lidar systems typically have a spatial resolution in the order of centimeters and a temporal resolution in the order of microseconds. Our group has also developed passive, so-called dark-field methods  based on sunlight illumination of the field of view (FOV). Remote dark-field methods yield comparable results to entomological lidar, but with much simpler instrumentation [26, 27]. By relying on sunlight, these methods are limited to daytime use and clear-sky conditions. Until now, passive dark-field methods have not been thought to provide range information.
Without range information, it is challenging to quantify the scattering cross section of a target. A number of methods have been developed to tackle this problem. Techniques such as nephelometry and flow cytometry  limit the probe volume to a point, allowing precise measures of the scattering properties of passing sparse particles in a limited volume. Other methods instead assess particle sizes in extinction mode. In digital in-line holography , the sparse intersection of zooplankton with the probe volume generates a diffraction pattern on a sensor. Through post-focusing, measures of size and position are obtained. These techniques utilize active laser illumination of the FOV in order to enable ranging. Compared to laser-based techniques, passive techniques benefit from reduced complexity, cost, weight, and power consumption. Laser eye safety considerations and radiation legislation are avoided, but operation is limited to daytime use and often clear-sky conditions. The strongest signal from aerofauna is obtained in backscatter mode, which puts constraints on the FOV orientation.
This work investigates whether optical ranging of sparsely distributed insects intersecting the probe volume of a quadrant sensor in the near-field can be achieved. A ranging equation based on a raytracing model is introduced and tested on field data. The estimated range to observed insects is used to evaluate the scattering processes in and along the probe volume. In addition to evaluating the ranging capabilities of quadrant sensors, we investigate the validity of a previously proposed model suggesting a relationship between the flight heading of insects and the frequency contents of the received signal . Finally, we test the method’s capability of profiling vertical and horizontal insect fluxes.
Numerical and analytical considerations
The detection scheme consists of a quadrant photodiode (QPD) in the image plane of a Newtonian telescope. The QPD is focused at a known distance, rfoc, which coincides with the position of a black termination cavity. As such, the QPD is unfocussed at the aperture but is gradually focused along the FOV toward rfoc. Expressed differently, the FOV of the QPD segments overlap entirely at the telescope aperture, and are gradually separated with distance until imaged sharply at rfoc. In this setup, the FOV is assumed to be evenly illuminated by the sunlight and is therefore equivalent to the probe volume. The properties of the imaging system are illustrated in Fig. 1.
The observed insects are small compared to the probe volume (corresponding to point sources when illuminated by homogeneous sunlight).
Throughout the duration of an observation, the insect velocity vector is constant.
The coaxial movement of the observed insects is limited compared to the length of the probe volume.
Aside from rfoc, there are other setup and observation parameters that affect the ranging accuracy. Equation 1 includes the quotient τ/Δt and is therefore invariant to the flight speed of insects. τ is equal to 0 at the aperture and Δt/2 at termination (see Fig. 1d). Δt depends on the cruise altitude of insects at close range due to the round aperture of the telescope, but is invariant to cruise altitude at far range due to the quadratic shape of the sensor. The quotient τ/Δt is equal to 0 at the aperture and increases linearly with range to 1/2 at termination, indicating that the ranging equation as a whole is invariant to cruise altitude. The climb and heading angles with which insects transit the probe volume can affect the results of Eq. 1. However, this effect is minute since the insect transit distance along the optical axis is negligible in comparison to the length of the probe volume (assumption 3).
Field measurements and insect observation properties
The sunlight propagates through the atmosphere before impinging on the FOV of the detector. As such, turbulence and other atmospheric phenomena can cause rapid variations in the optical background. By filtering the signal with a sliding median, the optical background was obtained. The noise level was obtained as the difference between the optical background and a sliding minimum filtered signal. A detection threshold with signal-to-noise ratio SNR = 2 was set.
As seen in Fig. 9, no insect observations at far ranges go below the 15 m/s line, whereas at closer ranges they do. This is an effect of the warping nature of the probe volume. The flight speed is calculated from the effective FOV diameter, which is a center-of-mass approximation. At short range, some insects transit the FOV with a shorter trajectory than the effective diameter, thereby ending up below the line. These insects do not necessarily fly faster than the ones at far range, but remain in the probe volume for a shorter time due to the shorter trajectory. It is also noted that insects with a high incident angle remain longer in the FOV. Therefore, their flight speed cannot be accurately gauged without knowledge of their heading. Presumably, a majority of the observations did not have a flight heading perpendicular to the FOV, and therefore only the shortest transit times are indicative of actual flight speed.
In this work, we demonstrated a novel method for extracting range information from insect observations in passive dark-field quadrant measurements. The derived ranging equation was applied to field observations of insects and validated. The range-dependent system sensitivity was investigated. A previously suggested model using modulation spectroscopy was proven insufficient for determining insect flight headings. Insect transit times at different ranges were compared to discreet flight speeds. The insect flight headings were also investigated using time-lag correlation, demonstrating the method’s capability of profiling vertical and horizontal insect fluxes. It was shown that most insects transiting the probe volume during the study period were flying laterally with the wind.
The obtained ranging Eqs. 1 and 1.1 depend on the dimensionless quotient τ/Δt and are therefore independent of the shape of the probe volume. This quotient is equal to 0 at the aperture due to the FOV overlap, and scales linearly with range to 1/2 at the termination. As such, both circular and quadratic sensors and apertures can be utilized without affecting the ranging properties.
In order to properly evaluate the system sensitivity with range, the insect flight headings have to be known. Due to their elongated shape, insects display different optical cross sections when observed and illuminated from different angles. Target classification can be accomplished through modulation spectroscopy, and would also aid in the evaluation of the system sensitivity. Observing the same insect species at the same angle but at different distances will allow calibration of the system, which in turn would aid in the quantification of insect sizes.
The heading investigation through modulation spectroscopy yielded both confirmative and non-confirmative results in this study. It has been observed in a laboratory setting that the specular, polarization-maintaining reflexes from insect wings contribute in particular to the higher harmonics of f0, but also contribute significantly to the lower harmonics. Depending on the relative phase between the specular components and the diffuse wing beats, this contribution can interfere either constructively or destructively. The phase of specular reflexes in the wing-beat cycle depends on the angle of illumination. This angle will shift throughout the day in passive measurements, but is constant in lidar measurements.
Obtaining flight heading information from insects transiting the probe volume would also enable the quantification of insect flight speeds. In the present study, it was demonstrated that a circular probe volume cross section introduces uncertainties in flight speed estimations. This can be overcome by utilizing a quadratic aperture, yielding a quadratic FOV cross section along the entire probe volume. Using a circular detector would yield a circular FOV at the termination, reintroducing the uncertainty at far ranges. Therefore, quadratic sensors and apertures are preferable to circular ones in this application.
There are a number of possible uses for the presented method. It can be implemented horizontally to profile insects in active or passive mode along transects across the landscape. By utilizing an infrared laser with a wavelength where the atmosphere does not scatter sunlight, the method can also be implemented vertically. Another approach to vertical profiling would be to fix the aim of the setup at the Polaris star, which would ensure sunlight impinging on the probe volume at an approximately normal angle at all times. The setup could then be implemented vertically in passive mode. Detection schemes with multiple wavelength- or polarization bands could also be envisioned in both active and passive mode.
We conclude that quadrant sensors can be used to determine the range to organisms or particles transiting the probe volume, and that fluxes can be quantified. We also conclude that the previously suggested model for determining insect flight headings using modulation spectroscopy is insufficient and needs further work. In particular, the effect of specular reflexes on the strength and phase of lower harmonics of f0 needs to be investigated.
The demonstrated ranging method has the potential to become a widespread tool for monitoring insect abundances and fluxes, particularly in farming environments. Similar to DOAS, which is a widespread method for monitoring gas concentrations in urban environments , the demonstrated method is inexpensive and simple to implement. Despite its ease of use and low power requirements, it yields rich information on insect observations. By retrieving range information, insect concentrations and fluxes are resolved in space and time and can be correlated to weather and topography. The method also yields modulation spectra that have a demonstrated capability of species classification. Our method can be adapted and scaled to other geometries where objects, organisms, or particles transit the probe volume. It can also be employed with active illumination, making it a versatile technique that can be used in many different fields of research.
MB designed the study, collected the data, and initiated the simulations. SJ finalized the simulations, analyzed the field data, and wrote the manuscript. Both authors took equal part in devising the passive ranging Eq. (1). Both authors read and approved the final manuscript.
We thankfully acknowledge the support and collaboration from Susanne Åkesson which enabled the project, Sandra Török for her participation in the field measurements as well as her previous work and documentation of the dataset, and Alem Gebru, Adam Bäckman, Carsten Kirkeby, Ádám Egri, Annika Söderman and Maren Wellenreuther for their assistance during the measurements.
The authors declare that they have no competing interests.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Ethics approval and consent to participate
Financial support was given by research grants to Susanne Åkesson from the Swedish Research Council (621-2013-4361) and Lund University. This work was supported by the Swedish Research Council Linnaeus grants (349-2007-8690 & 349-2006-121) and Lund University to the Centre for Animal Movement Research (CAnMove) and to the Lund Laser Centre (LLC). The work is also supported by the Swedish Research Council through a U-forsk grant (348-2014-3481). No funding body had any role in the design of the study, data collection, analysis, interpretation or writing of the manuscript.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided 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 Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Organization WH: World Malaria Report. 2015.Google Scholar
- Das AJ, Stephenson NL, Davis KP. Why do trees die? Characterizing the drivers of background tree mortality. Ecology. 2016;97:2616–27.View ArticlePubMedGoogle Scholar
- Oliveira CM, Auad AM, Mendes SM, Frizzas MR. Crop losses and the economic impact of insect pests on Brazilian agriculture. Crop Prot. 2014;56:50.View ArticleGoogle Scholar
- Vrijheid M, Casas M, Gascon M, Valvi D, Nieuwenhuijsen M. Review: environmental pollutants and child health—a review of recent concerns. Int J Hyg Environ Health. 2016;219:331–42.View ArticlePubMedGoogle Scholar
- Sánchez-Bayo F, Goulson D, Pennacchio F, Nazzi F, Goka K, Desneux N. Are bee diseases linked to pesticides? A brief review. Environ Int. 2016;89:7–11.View ArticlePubMedGoogle Scholar
- Zhu S, Malmqvist E, Li W, Jansson S, Li Y, Duan Z, Svanberg K, Feng H, Song Z, Zhao G, et al. Insect abundance over Chinese rice fields in relation to environmental parameters, studied with a polarization-sensitive CW near-IR lidar system. Appl Phys B. 2017;123:211.View ArticleGoogle Scholar
- Dao A, Yaro AS, Diallo M, Timbine S, Huestis DL, Kassogue Y, Traore AI, Sanogo ZL, Samake D, Lehmann T. Signatures of aestivation and migration in Sahelian malaria mosquito populations. Nature. 2014;516:387–90.View ArticlePubMedPubMed CentralGoogle Scholar
- Guan Z, Brydegaard M, Lundin P, Svanberg S, Wellenreuther M, Runemark A, Svensson EI. Insect monitoring with fluorescence lidar techniques: field experiments. Appl Opt. 2010;49:5133–42.View ArticlePubMedGoogle Scholar
- Silver JB. Mosquito ecology: field sampling methods. Berlin: Springer; 2008.View ArticleGoogle Scholar
- Majambere S, Massue DJ, Mlacha Y, Govella NJ, Magesa SM, Killeen GF. Advantages and limitations of commercially available electrocuting grids for studying mosquito behaviour. Parasit Vectors. 2013;6:1.View ArticleGoogle Scholar
- Kirkeby C, Wellenreuther M, Brydegaard M. Observations of movement dynamics of flying insects using high resolution lidar. Sci Rep. 2016;6:29083.View ArticlePubMedPubMed CentralGoogle Scholar
- Eklundh L, Johansson T, Solberg S. Mapping insect defoliation in Scots pine with MODIS time-series data. Remote Sens Environ. 2009;113:1566–73.View ArticleGoogle Scholar
- Müller J, Brandl R. Assessing biodiversity by remote sensing in mountainous terrain: the potential of LiDAR to predict forest beetle assemblages. J Appl Ecol. 2009;46:897–905.View ArticleGoogle Scholar
- Drake VA, Reynolds DR. Radar entomology: observing insect flight and migration. Wallingford: CABI Publishing; 2012.View ArticleGoogle Scholar
- Chapman JW, Drake VA, Reynolds DR. Recent insights from radar studies of insect flight. Annu Rev Entomol. 2011;56:337–56.View ArticlePubMedGoogle Scholar
- Melnikov V, Leskinen M, Koistinen J. Doppler velocities at orthogonal polarizations in radar echoes from insects and birds. IEEE Geosci Remote Sens Lett. 2014;11:592–6.View ArticleGoogle Scholar
- Drake VA. Distinguishing target classes in observations from vertically pointing entomological radars. Int J Remote Sens. 2016;37:3811–35.View ArticleGoogle Scholar
- Brydegaard M, Gebru A, Svanberg S. Super resolution laser radar with blinking atmospheric particles—application to interacting flying insects. Prog Electromagn Res. 2014;147:141–51.View ArticleGoogle Scholar
- Repasky KS, Shaw JA, Scheppele R, Melton C, Carsten JL, Spangler LH. Optical detection of honeybees by use of wing-beat modulation of scattered laser light for locating explosives and land mines. Appl Opt. 1839;2006:45.Google Scholar
- Shaw JA, Seldomridge NL, Dunkle DL, Nugent PW, Spangler LH, Bromenshenk JJ, Henderson CB, Churnside JH, Wilson JJ. Polarization lidar measurements of honey bees in flight for locating land mines. Opt Express. 2005;13:5853–63.View ArticlePubMedGoogle Scholar
- Brydegaard M, Gebru A, Kirkeby C, Åkesson S, Smith H. Daily evolution of the insect biomass spectrum in an agricultural landscape accessed with lidar. EPJ Web Conf. 2016;119:22004.View ArticleGoogle Scholar
- Moore A, Miller RH. Automated identification of optically sensed aphid (Homoptera: Aphidae) wingbeat waveforms. Ann Entomol Soc Am. 2002;95:1–8.View ArticleGoogle Scholar
- Potamitis I, Rigakis I. Novel noise-robust optoacoustic sensors to identify insects through wingbeats. IEEE Sens J. 2015;15:4621–31.View ArticleGoogle Scholar
- Chen Y, Why A, Batista G, Mafra-Neto A, Keogh E. Flying insect classification with inexpensive sensors. J Insect Behav. 2014;27:657–77.View ArticleGoogle Scholar
- Runemark A, Wellenreuther M, Jayaweera HHE, Svanberg S, Brydegaard M. Rare events in remote dark-field spectroscopy: an ecological case study of insects. IEEE J Sel Top Quantum Electron. 2012;18:1573–82.View ArticleGoogle Scholar
- Brydegaard M, Merdasa A, Gebru A, Jayaweera H, Svanberg S. Realistic instrumentation platform for active and passive optical remote sensing. Appl. Spectrosc. 2016;70(2):372–85.View ArticlePubMedGoogle Scholar
- Gebru A, Rohwer E, Neethling P, Brydegaard M. Investigation of atmospheric insect wing-beat frequencies and iridescence features using a multispectral kHz remote detection system. J Appl Remote Sens. 2014;8:083503.View ArticleGoogle Scholar
- Darzynkiewicz Z, Roederer M, Tanke HJ. Cytometry: new developments. Amsterdam: Elsevier Academic Press; 2004.Google Scholar
- Jericho SK, Garcia-Sucerquia J, Xu W, Jericho MH, Kreuzer HJ. Submersible digital in-line holographic microscope. Rev Sci Instrum. 2006;77:043706.View ArticleGoogle Scholar
- Brydegaard M. Towards quantitative optical cross sections in entomological laser radar—potential of temporal and spherical parameterizations for identifying atmospheric fauna. PLoS One. 2015;10:e0135231.View ArticlePubMedPubMed CentralGoogle Scholar
- Malmqvist E, Jansson S, Török S, Brydegaard M. Effective parameterization of laser radar observations of atmospheric fauna. IEEE J Sel Top Quantum Electron. 2016;22:1–8.View ArticleGoogle Scholar
- Török S. Kiloherts electro-optics for remote sensing of insect dispersal. Lund: Lund University; 2013.Google Scholar
- Platt U, Stutz J. Differential optical absorption spectroscopy: principles and applications. Berlin: Springer; 2008.Google Scholar