Research Technical Report
Effects of Artificial Light on Deep Sea Organisms: Recommendations for ongoing use of artificial lights on deep sea submersibles
Kochevar, R.E. (1998)
Monterey Bay Aquarium
866 Cannery Row, Monterey, CA 93940
Technical Report to the Monterey Bay National Marine Sanctuary Research Activity Panel, January, 1998
At a meeting of the Research Activity Panel of the Monterey Bay National Marine Sanctuary (MBNMS; 4/12/96), Dr. Richard Parrish presented concerns regarding potential damage to visual systems of deep sea animals following exposure to bright lights on submersible vehicles. A subcommittee was appointed to develop the following working document, to summarize the current state of knowledge on effects of artificial light on vision in deep sea animals, to assess (if possible) the likely impact of routine submersible operations on animal communities, and to offer suggestions for future action (e.g., research, etc.) pertinent to this topic.
The natural optic environment of the deep oceans consists of very dim, homochromatic, downwelling light, supplemented by animal bioluminescence. Most organisms inhabiting this environment possess highly specialized visual systems, which are sensitive to the small amount of light that is there. The sensitive nature of these visual systems might make such organisms vulnerable to damage from exposure to bright artificial lights of submersible vehicles. This document will outline the current state of knowledge on effects of artificial light on vision in deep sea animals; assess (where possible) the likely impact of routine submersible operations on animal communities; and offer suggestions for future research aimed at addressing the impacts of submersible use on visual organisms.
The majority of literature dealing with light in the deep oceans focuses specifically upon characterizing the visual systems of deep sea organisms, especially fish and crustaceans. These organisms have been of interest to the vision research community primarily due to the unique visual environment they inhabit.
Extensive studies concerning morphology of deep sea fish and crustacean eyes demonstrate a wide variety of adaptations to life in near-darkness. In fishes, such adaptations include gross changes in eye anatomy (Marshall 1979), as well as increased retinal photoreceptor sizes (Munk 1966) and increased levels of visual pigment (Denton and Warren, 1957). Several species of crustaceans exhibit analogous adaptations, including depth-related increases in rhabdom length and crystalline cone dimensions (review in Land, 1981; Hiller-Adams and Case, 1984), This combination of characteristics provides high light sensitivity required for low-light vision; however, it may also make them more susceptible to damage from bright light.
Numerous studies of deep sea fishes and crustaceans have demonstrated that most of these organisms contain retinal photopigments with peak sensitivities between 450 and 500 nm; roughly similar to the spectra of downwelling light and bioluminescent emissions (Denys and Brown, 1982; Frank and Case, 1988; Hiller-Adams et. al., 1988; Partridge et. al., 1988 ). Although a few fish species also appear to harbor visual pigments absorbing at longer wavelengths, these adaptations appear to be correlated with the simultaneous presence of unusual red photophores on the bodies of these animals, the function of which remains a matter of debate (Partridge et. al., 1989). Similarly, dual visual pigments in some crustaceans appears to be correlated with the presence of photophores in these species, although the specific role of such pigments remains unclear (Frank and Case, 1988).
Only a small number of studies have been published specifically addressing questions regarding the effects of light on deep sea organisms. These fall into two broad categories: Some focus on physiological effects of light exposure, observed as changes in morphology or neurophysiology of animal (mostly crustacean) eyes; while others focus on the use of artificial lights on trawling gear, which speak to effects of light upon animal (mostly fish) behavior. Both types of studies relate to the topic of submersible lights, and possible effects of their use on deep sea fauna.
The nature of light-induced damage to photoreceptors has been well documented in a variety of marine crustaceans. As with other visual systems, the inherent nature of photopigments allows for the possibility of ìphotobleachingî. Simply put, this phenomenon occurs when photopigments are exposed to light, which changes their chemical form in a reversible manner. This phenomenon will be immediately familiar to anyone who has stared at a colored object for several seconds, and then shifted their focus to a white page - resulting in the appearance of a ìshadow imageî in complementary colors to those of the original object, which persists for a few seconds before subsiding.
Because fish and crustaceans eyes lack irises, which in more complex eyes limit the amount of light striking the retina, the effects of acute and drastic increases in light intensity are likely to be much more profound in these organisms. Thus, we would expect photobleaching to occur rapidly, and to affect a high proportion of photopigment, among deep sea animals. However, the limited data available indicate that photopigments in these systems, as with other visual systems, regenerate with time (Hiller-Adams et al., 1988).
Following long-term exposure to bright light, an irreversible phenomenon, distinct from photobleaching, has been found to occur. This phenomenon involves damage to the retinal tissue itself, and can be visualized using standard histological techniques. Although similar effects can be elicited in any visual system, it has been suggested that such damage may occur more quickly and at lower light levels among deep sea organisms (Meyer-Rochow, 1994; Nilsson and Lindstrom, 1983).
The few studies published addressing effects of light exposure on deep sea animals suggest that exposure to bright light can result in a variety of physiological responses, ranging from little or no effect, to temporary (recoverable) effects, to permanent blindness. In the most complete study we found, specimens of the deep sea isopod Cirolana borealis were exposed to light of different intensities (no light, 117 lx, 1250 lx, 2500 lx, daylight) for different periods of time (10 or 60 min.), and subsequently observed using elecroretinography and histology. This study demonstrated that morphologically and functionally the eyes are destroyed by exposure to light at levels at or above 1250 lx (4.9 W m-2), with individuals exposed to ìordinary room lightî (measured at 117 lx), exhibiting little damage after initial exposure of either 10 or 60 minutes, and nearly complete recovery after 12 hours in darkness (Nilsson and Lindstrom, 1983).
In a similar study, Norway lobsters (Nephrops norvegicus), collected at depths from 30-180m, were exposed to light ìfrom a single fluorescent tubeî for varying lenghts of time, and observed using histological techniques (Loew, 1976). This species was found to suffer complete rhabdom degeneration after 2.5 hours of exposure. However, comparison of these results with those of the study above is difficult since light intensities were not quantified. Furthermore, the extensive duration of exposure (2.5 hours being the minimum; far longer than in situ submersible observations) make this work largely irrelevant to potential impacts of submersible use.
In extensive studies concerning neurophysiology of vision in the midwater mysid Gnathophausia ingens, it has been found that this species exhibits full recovery of light response when measuring electroretinograms or extracellularly from visual interneurons following up to 15 minutes of exposure to ìordinary room lights,î provided that the specimen was allowed several hours of dark adaptation after being exposed to light. However, because these studies were not conducted specifically to study the effects of bright light exposure, these data have not been published to date (personal communication: J. Moeller).
Although studies of light impacts on deep sea animals are scarce, it seems that the following generalities based upon studies of shallow-living species likely hold true in the deep sea: The nature of response to light exposure varies among species, and appears to be dependent upon numerous factors, including spectral characteristics of the light and duration and intensity of exposure. There is also evidence to suggest that there is a high degree of variation among individual responses to light exposure within a given species, believed to result from a combination of factors including pre-adaptation to light, environmental temperature, diet, and levels of blood-borne substances such as serotonin, ascorbic acid, and small proteins (Meyer-Rochow, 1994).
Another factor, certainly relevant to impacts of submersible use, is that of animal behavior. This topic has been addressed in a few studies. In a series of experiments conducted to investigate the potential use of lights to enhance commercial trawling, M. R. Clarke and colleagues installed 70 W lights on rectangular midwater trawls and bottom trawls, and compared catches with and without illumination (Clarke and Pascoe, 1985). Their results showed that catch rates of midwater fishes increased with light at 600 m, 800 m, and 1500 m, with no significant differences observed at 200 m or 300 m. Catch frequencies of cephalopods also increased at 300 m and 800 m. Catch rates of decapod crustaceans, on the other hand, decreased at 200 m, 300 m, 800 m, and 1500 m (Clarke and Pascoe, 1985). Among the 79 species of midwater fishes caught, at least twelve species exhibited increased catch rates with light, while only one had lower catch rates (Swinney et al., 1986). And among benthic fishes, collected at relatively shallow depths (60 m), catch frequencies of three species increased with light and four decreased (Clarke et al., 1986).
The results of the trawling studies were attributed to three factors: 1 - Positive phototactic responses which brought more animals into the path of the trawls; 2 - Phototactic responses of some species which may in turn attract their predators; 3 - Dazzling and sensory blanking effects may reduce the animalsí ability to make avoidance responses (Pascoe, 1990). Unfortunately, none of these studies address the issue of long-term effects.
Laval and Baussant (1990) describe experiments conducted with the submersible Cyana, in which the effects of light on the deep scattering layer were measured. Working at 300 m, 1250 W lights on the submersible had a marked effect on the surrounding fauna, as recorded by an echo-sounder. With the lights on, the scattering layer dispersed in a broad area around the submersible. When the lights were turned off, the scattering layer re-appeared around the sub. Based on trawl sampling and direct observations, a principal component of these layers were abundant representatives of the fish genus Cyclothone (Laval and Baussant, 1990).
Other Pertinent Information
In addition to the limited published literature on this topic, there is a continually growing body of relevant information in the form of direct observations of deep sea animals observed, collected, and/or maintained in captivity. Although these observations are typically not published (or when they are, specific information regarding vision is not included), they are still potentially valuable in helping to understand the impacts of artificial lights on deep sea animals.
Several species of deep-sea fish have been successfully maintained in the laboratory (summarized by Smith and Baldwin, 1997). The general experience of researchers who have worked with these animals is that while artificial light may temporarily overload the visual system, virtually all fishes seem to recover their visual capabilities after a few hours or days if they are kept in the dark (personal communications: K. L. Smith, J. J. Childress, B. H. Robison).
In recent laboratory studies of two zoarcid fishes from deep water in Monterey Bay (Melanostigma pammelas and Lycodapus mandibularis), exposure to artificial lights did lead to apparent visual blanking, however after spending as little as an hour in darkened aquaria, they regained their behavioral responses to visual stimuli (personal communication: B. H. Robison). Furthermore, both of these species have been observed to thrive for months or even years in captivity, while being maintained under ìnormalî (i.e., fluorescent tube) laboratory lighting (personal communication: G. Van Dykhuizen).
Other species of fishes, cephalopods, and crustaceans have also been maintained for extended periods in the laboratory, and have been found to retain behavioral responses to light. For example, several specimens of the fangtooth (Anoplogaster cornuta) have been collected, using both midwater trawls and ROV. In all cases, once in captivity they exhibited responses to direct illumination, even after being exposed briefly to sunlight, laboratory illumination, and/or prolonged observation with the ROV (personal communication: G. Van Dykhuizen). Similar observations have been made on cephalopods (Chiroteuthis, Galiteuthis, Gonatus, Histioteuthis, Vampyroteuthis infernalis; personal communications: B. Robison and K. Reisenbichler), and crustaceans, including the deep sea mysid Gnathophausia ingens which has been observed to be capable of prey capture after weeks or months of captivity, during which they were frequently exposed to laboratory light (personal communication: G. Van Dykhuizen).
Recently (September, 1997) a symposium on the sensory systems of deep-sea animals was held as part of the international Deep-Sea Biology Symposium, in Monterey. When the question of the recovery of optical pigments of deep sea fishes (after exposure to artificial lighting) was put to the assembled experts, the general response was simply that the studies necessary to provide an answer have not been done. Dr. Ronald Douglas (Dept. of Optometry & Visual Sciences, City University, London) did comment that the only wavelengths of light liable to cause irreversible damage to optical pigments were those in the ultraviolet range, and that these wavelengths are not likely to be produced at significant levels by white-light sources (R. Douglas, personal communication).
Potential for Impact
The incomparable access to the deep sea afforded by Monterey Submarine Canyon, coupled with the presence of a large, active research community in the Monterey Bay area, has made this area one of the most intensively studied deep sea ecosystems in the world. It has been visited by the Russian Mir submersibles, and been the site of several Alvin dives. Since the development of the ROV Ventana at MBARI, some sites in Monterey Canyon have been visited several times per month; and the frequency of submersible dives in the canyon will increase as MBARIís ROV Tiburon is brought on-line for routine dive operations.
Although submersible operations certainly hold the potential to impact deep sea animal communities, it is not currently possible to ascertain exactly what the potential impacts might be. This arises from two major factors: 1 - as we have seen, our knowledge about the effects of light on deep sea animals is severely limited. This topic has been investigated in only a few species, and even these studies are of limited value because the conditions under which they were performed are not similar to the conditions likely encountered in situ during submersible operations. 2 - Although we know quite a lot about the specifications of the submersibles currently in use (e.g., spectral and luminance characteristics of on-board lighting), the specific way in which these tools are used varies on a daily basis, as do the conditions under which dive operations are performed. Given these factors, it is not even possible to calculate the volume of water likely to be impacted on any given day. In order to do so, one would need to know the light transmissivity of the water, which varies with time and location; the relative velocity of the submersible, which may range from stationary observation to full-speed flight, and which changes moment-to-moment on all dives; and the intensity and position of each light, which may also change between and even during dive operations.
With all that having been said, it bears pointing out that a submersible suspended in the midwater produces a cone of light, typically projecting forward from the vehicle. At a distance of a few meters, regardless of the number of lights, a submersible acts as directional point-source of light, whose intensity falls off logarithmically with distance. The volume of this cone, the only region in which light is present, is miniscule in comparison to the unlit volume surrounding the submersible in every other direction.
During benthic operations, the volume illuminated by the submersible is much more limited, since the ìlight coneî produced by the submersible lights impinges upon the substrate, rather than traveling out through the water. However, benthic operations represent a potentially different type of impact than dives in the midwater; because sites on the sea floor are inherently static, and many of the benthic organisms are sessile, there exists the distinct possibility of benthic organisms being exposed repeatedly to light (particularly at sites which are visited frequently), and for extended periods of time. Although the vast majority of these sessile organisms are non-visual, and thus not likely to be affected by light, some visual benthic and epi-benthic organisms, particularly those specifically associated with substrates or communities under intensive study (e.g., hydrothermal vents, cold seeps, etc.), may be at an increased level of risk relative to their midwater counterparts.
Another consideration is the use of light sources other than those used to illuminate the submersibleís surroundings. These currently include lasers, which are often used in pairs or other multiple-source arrays, to make various physical measurements; but may include other, more sophisticated high-intensity light technologies (such as LIDAR systems; see Gauldie et. al., 1996 for review) in the future. Although the potential volume of impact of devices currently in use are quite small relative to the volume illuminated by the operating lights, the high intensity light produced from lasers may be more damaging to organisms unfortunate enough to be struck directly in the eye by the laser beam. This risk may be increased with the use of a LIDAR system, which would likely use higher power lasers scanning or sweeping through a larger volume. Although LIDAR systems are not currently available for deep sea submersible deployment, aircraft- and shipboard-mounted LIDARs are now commercially available, so it is likely just a matter of time before they are adapted to be deployed at depth.
In order to provide a definitive management scheme for the use of artificial lights in the deep sea, the following information would be required: for each and every species of concern, we would need to have a three-dimensional data space, consisting of wavelength, intensity, and time of exposure. Within this data space would be threshold surfaces, showing tolerance limits for temporary (photobleaching) and permanent (retinal) damage. Using such a dataset, it would be possible to construct lighting designs and operational procedures allowing submersible observation of deep sea animals at manageable levels of risk.
Unfortunately, such a dataset does not presently exist for any deep sea organism; nor is it likely to exist. The number of specimens required for such a study to be carried out, at sufficient resolution and with sufficient levels of confidence to be meaningful, is simply beyond the scope of reason. Furthermore, this number of specimens of a single species, plus the amount of by-catch inevitably associated with such a collection, would incur far greater losses to a population than routine submersible use.
An alternative to this scenario would be to assume that the spectrum of light to which an animal would be exposed is that of submersible lights currently in use, and that they are running at full intensity in gin-clear water (a ìworst-caseî scenario). With these assumptions in place, the dataset upon which management decisions would be based is reduced to two dimensions: intensity (or distance from the light source, since the light fall-off is well-characterized for commercially-available light sources), and time of exposure. With such a dataset in place for a given species, a researcher encountering a specimen with an ROV or manned submersible could manage the encounter with some knowledge of the effect that the encounter is likely to have, at a given distance and for a given time, and make informed decisions about how to proceed.
Although such a scheme may be desirable, there still exists the practical question of how much time and effort, and how many specimens, would be required to carry out the necessary research to provide a useful dataset. For each species of concern, and using a modest number of light intensities and minimal replicates, tens to hundreds of individual specimens (collected and maintained alive, and in total darkness) would be required. Although such numbers of common species might be readily obtained, similar numbers of more rare species would likely involve months or years of routine collection; and the discarded by-catch would again likely far exceed the numbers of animals likely to be affected through routine submersible operations.
Thus we are left with the following facts: 1) There is not currently enough information available to carry out a robust risk analysis for submersible use. 2) The steps required to obtain the necessary information are not feasible for most species.
It is clear that the current state of knowledge is insufficient to devise a sound management policy for the use of artificial lights in the deep sea. However, available information suggests that deep sea fauna may be affected by exposure to bright artificial lights, underlining the need for research on this topic. We therefore make the following recommendations:
- Ongoing use of manned and unmanned submersibles should not be curtailed. Although they might present a potential hazard to deep sea fauna, the potential volume of impact is miniscule relative to the habitat volume, and is certainly less than the volume impacted by other research and commercial trawling activities.
- A reasonable investigation could be made by researchers utilizing submersibles by collecting organisms exposed to high-intensity lights (and/or lasers, if they are used) for extended periods of time. Distance from the submersible and time of exposure could be noted, and eye, eye-cup, retinal, or ommatidial samples from such animals could be processed for histology and provided to research facilities capable of analyzing them for histological damage. (Dr. Ronald Douglas, from the Dept. of Optometry & Visual Sciences, City University, London, has expressed an interest in obtaining and processing such specimens.) Through this process, the scientific value of specimens likely affected by bright lights may be increased, and baseline data could be collected which would contribute to our overall understanding of the impacts of submersible use, thus putting us in a better position to develop more robust risk-management strategies.
- At benthic sites, it may be possible to evaluate possible impacts on a more gross (i.e., community) scale by monitoring changes in species abundance and diversity at similar sites, which are visited at differing frequencies (e.g., a cold seep which is visited six times per year would be compared to a similar seep which is visited only once a year). Furthermore, in order to test the hypothesis that light has an impact, such studies should specifically compare changes in abundance and diversity of visual vs. non-visual species.
- Funding should be made available to pursue studies directly relevant to this issue. Basic information regarding tolerances to submersible lights, even among the commonest species of animals, is sorely lacking; research aimed at providing such information would be feasible in a laboratory equipped to carry out such work.
Clarke, M.R. and P.L. Pascoe, 1985. The influence of an electric light on the capture of deep-sea animals by a midwater trawl. J. Mar. Biol. Assoc. U. K., 65:373-393.
Clarke, M.R., P.L. Pascoe, and L. Maddock, 1986. Influence of 70 watt electric lights on the capture of fish by otter trawl off Plymouth. J. Mar. Biol. Assoc. U. K., 66:711-720.
Denton, E.J. and F.J. Warren, 1957. Photosensitive pigments in the retinae of deep-sea fish. J. Mar. Biol. Assoc. U.K., 36:651-662.
Denys, C.J. and P.K. Brown, 1982. Euphausiid visual pigments. J. Gen. Physiol. 80:451-472.
Frank, T.M. and J.F. Case, 1988. Visual spectral sensitivities of bioluminescent deep-sea crustaceans. Biol. Bull. 175:261-273.
Gauldie, R.W., S.K. Sharma and C.E. Helsley, 1996. LIDAR applications to fisheries monitoring problems. Can. J. Fish. Aquat. Sci. 53:1459-1468.
Hiller-Adams, P. and J.F. Case, 1984. Optical parameters of euphausiid eyes as a function of habitat depth. J. Comp. Physiol. A, 154:307-318.
Hiller-Adams, P., E.A. Widder and J.F. Case, 1988. The visual pigments of four deep-sea crustacean species. J. Comp. Physiol. A. 163:63-72.
Land, M.F., 1981. Optics of the eyes of Phronima and other deep-sea amphipods. J. Comp. Physiol. 145:209-226.
Laval, P. and T. Baussant, 1990. Effect of the lights from an approaching submersible on the 15kHz deep scattering layer in the Ligurian Sea (Mediterranean). C. R. Acad. Sci. Paris, t. 311, Series III:181-186.
Loew, E.R., 1976. Light, and photoreceptor degeneration in the Norway lobster, Nephrops norvegicus (L.). Proc. R. Soc. Lond. B. 193:31-44.
Marshall, N.B., 1979. Developments in deep sea biology. Blanford Press, 566 pp.
Meyer-Rochow, V.B., 1994. Light-induced damage to photoreceptors of spiny lobsters and other crustaceans. Crustaceana 67(1):95-109.
Munk, O., 1966. Ocular anatomy of some deep-sea teleosts. Dana Report 70:1-62.
Nilsson, H.L. and M. Lindstrom, 1983. Retinal damage and sensitivity loss of a light-sensitive crustacean compound eye (Cirolana borealis): Electron microscopy and electrophysiology. J. Exp. Biol. 107:277-292.
Pascoe, P.L. 1990. Light and capture of marine animals. In: Herring, P.J., Campbell, A.K., Whitfield, M. and Maddock, L. (eds). Light and Life in the Sea. Cambridge Univ. Press, 357 pp.
Partridge, J.C., S.N. Archer and J.N. Lythgoe, 1988. Visual pigments in the individual rods of deep-sea fishes. J. Comp. Physiol. A. 162:543-550.
Partridge, J.C., J. Shand, S.N. Archer, J.N. Lythgoe, and W.A.H.M. van Groningen-Luyben, 1989. Interspecific variation in the visual pigments of deep-sea fishes. J. Comp. Physiol. A. 164:513 - 529.
Smith, K.L. and R.J. Baldwin, 1997. Laboratory and in situ methods for studying deep-sea fishes. In: Randall,D.J. and Farrell, A.P. (eds), Fish Physiology, vol. 16: Deep-Sea Fishes. Academic Press, San Diego.
Swinney, G.N., M.R. Clarke, and L. Maddock, 1986. Influence of an electric light on the capture of deep-sea fish in Biscay. J. Mar. Biol. Assoc. U. K., 66:483 - 496.