US Department of Agriculture, USDA Forest Service, Technology and Development Program Banner with Logos.
Images from various aspects of the T&D Program.
HomeAbout T&DT&D PubsT&D NewsProgram AreasHelpContact Us
Search T&D
   Go
  Search all USDA
Program Areas
Left Nav Bottom
T&D > Programs Areas > Inventory & Monitoring > Stress Detection Glasses Program Areas
Evaluation of Stress Detection Glasses

Entire Report

"Detect stress in your plants before it's too late". This is the type of statement used to market plant stress detection glasses. Do they work? How well do they work? Do they have applications in Forestry?

The glasses are available through several commercial outlets. They are marketed through these outlets by Dr. Robert Brock of Beaverton, OR. The glasses are marketed to producers of crops. They are advertised to spot, from a distance, stressed turf or field crops due to insect, disease or moisture stress. They would be an "early warning system" that can be used by field work crews. More recently they have been offered in forestry supply catalogues.

The glasses were developed by a NASA scientist in an attempt to identify camouflaged objects in forests, grass or jungle environments. The glasses filter out green color, while enhancing yellow and red wavelengths. Dr Brock's paper "Stress Detection Glasses, History and Additional Information", September 2002, details how the glasses function.

A search for information evaluating these glasses showed there has been little formal evaluation. Most information is anecdotal. A formal paper "Efficacy of NASA Plant Stress Detection Glasses for Pine Beetle Detection" by Rankin, Heath and Murtha details the use of the glasses during aerial surveys to locate bark beetle infestations. A comparison of mapping with and without the glasses indicated that significantly more beetle infestations were recorded when the glasses were worn.

The glasses are designed to block the color green and enhance other colors. They would possibly amplify changes (reductions) of green in the plant foliage brought about by a change in the chlorophyll when the plant is under stress. The glasses could feasibly make it possible to detect color changes in green that would not be visible to the unaided eye. This would have application for determining when vegetation is under stress and providing advanced warning so that action can be taken to relieve that stress by watering, fertilizing, removing insects or disease, whatever is causing the stress. It may also provide application during designation of trees to be removed for salvage because of fire, disease or insect attack. It may be possible with the glasses to determine, at an earlier stage, when a tree is starting to fade. This could be very useful in the spring or early summer when trees start to come under moisture stress and fading of crown begins.

STRESS DETECTION GLASSES
The stress detection glasses are available from several
suppliers and come in regular and clip-on styles.

Viewing
Examples of what vegetation looks like through the stress detection glasses. With actual observations using the glasses the contrast of vegetation changes seems to be more intense than it does through the camera lenses.


Normal view of tree showing
live and dead foliage.


View of same tree through stress
detection glasses. Dead/dying
foliage shows bright red.


Above and below - Normal view
of fading leaves.


Above and below - Same view with
glasses. Green leaves appear gray
violet.


Normal view of tree showing
live and dead foliage.


View of same tree through stress
detection glasses. Dead/dying
foliage shows bright red.

Evaluation of Glasses

The glasses were tested at the San Dimas Technology and Development Center. The purpose of the testing was to determine if plant stress could be detected with the glasses before it could be detected with the unaided eye.

Preliminary Evaluation
A preliminary evaluation was started in January, 2003 at the SDTDC. Herbicide (Roundup) was applied to 4 patches (approximately 18x18 inches) of naturally occurring weeds (2 grass and 2 broadleaf). Application of herbicide was at 10:00 am on January 22, 2003. Weather - Approximately 60 degrees, 50% relative humidity, and hazy. Most changes to the plants occurred within the first two or three days. Observations were made over several more days to take advantage of changing light conditions (hazy and sunny days).

Test areas were checked approximately every 2 hours during daylight/work day. After 24 hours evidence of stress started to show up in the grass but this was only evident upon close inspection with the glasses and probably would not have been noticed if the test areas were not identified. At about 28.5 hours the stress to the grass was becoming very noticeable with the glasses. Without the glasses the changes could be seen but were not as obvious at a glance. The glasses appeared to amplify the differences. The broadleaf weeds had some leaf curling at 28.5 hours. However the herbicide effects at this time were not very pronounced and would not have been observed at a glance, with or without the glasses.

Early morning and late afternoon observations seemed less effective than when the sun was higher in the sky (approximately 10am to 3pm for this time of year). On sunny days, during the midday hours, the glasses seemed to make fading in leaves much more evident. During overcast days the differences were similar throughout the day—probably a little easier to discern in the morning and evening hours but less pronounced at midday.

The preliminary evaluation showed that the glasses should be tested to consider sun angle (time of day) and sun position relative to the observer.

Formal Evaluation
The evaluation of the glasses was undertaken starting July 14th, 2003. High temperatures were in the 90 to 100 degree range. Humidity varied but was high for this period of time in southern California. It was mostly sunny but there were periods of hazy weather. Eight species of plants native to California were used in the test. These were manzanita, sedge, bush anemone, grass, ponderosa pine, toyon, monkeyflower, and ceanothus. Individuals of each type were stressed by withholding water (drought stress) or by applying a light dose of herbicide. The plants were evaluated by six people, several times a day, over a period of four days. Persons evaluating the plants had forestry field experience ranging from none to over 30 years. The plants were viewed from four angles (north, east, south, and west). The plants were evaluated each time and at each angle with and without the glasses.

Method

Most plants were in five-gallon containers, except the grass and sedge which were in one-gallon containers. Each plant species was assigned a group (A through H). Each plant was assigned a number within the group. One or two plants of each species were lightly treated with pre-mixed Roundup and one or two plants of each species had water withheld to simulate drought. The plants had not been watered over the previous weekend. Starting on July 14th the plants not being tested for drought stress were watered daily. The evaluators did not know which plants were being stressed.

Evaluation was based on a comparison of other plants within the same group but not between groups. Each plant was recorded as having no change, minor/slight change, or obvious changes (assigned values of 0, 1, or 2, respectively). There were some initial observable differences due to the natural variation in plants and the fact that there were dead or dying leaves or branches on some plants. The test was started immediately after initial treatment at a time when none of the treatment effects would be showing up. This allowed for natural variation to be recorded and changes evaluated from these base observations. The testing was terminated after four days because the effects of the herbicide treatments were fully evident. The effects of drought stress were evident in the least drought resistant plants in the larger containers and in the small plant containers (dry out more quickly). The most drought resistant plants did not show much evidence of stress but it was felt that there was enough data for purposes of the test.

Results

The results of the evaluation were tabulated and comparisons made of the results with and without the glasses. Each day was broken into five time periods to evaluate for sun angle. Data was also separated by direction of viewing (north, east, south, or west) and further segregated by the location of the sun to the observer.

Observations by time period
The data represented in Figure 1 (next page) indicates that in the first two days there were more changes observed with the glasses than without them. This appears to reverse for the last two days where changes observed without the glasses exceeded changes observed with the glasses. This indicates that the glasses do aid initially in the detection of stressed plants and are better than the naked eye. Of the 10 time periods in the first two days the changes detected with the glasses exceeded those without the glasses 8 out of 10 times. In the last two days the changes detected with the glasses exceeded those without the glasses only 1 out of 9 times.

The differences in the first two days compared to the last two days can be explained by how the glasses work. The glasses make green vegetation appear gray or black but make yellows and orange brighter. This would make the vegetation with diminished chlorophyll appear bright orange. Brown or gray vegetation is not enhanced by the glasses. Therefore, over a period of time, the vegetation that initially appeared bright when it is stressed later blends in with the darker colors that the glasses turn the green vegetation.

Change observed by time period.
Date/Time Without
glasses
With
glasses
7/14 to 0900
2
6
to 110
19
21
to 1400
41
55
to 1530
25
36
to 1700
65
58
7/15 to 0900
83
75
to 1100
90
103
to 1400
201
224
to 1530
26
30
to 1700
71
92
7/17 to 0900
170
170
to 1100
256
252
to 1400
293
236
to 1530
127
113
to 1700
65
80
 
129
106
 
318
279
 
341
313
 
245
223
 
0
0
Figure 1. Each day was divided into 5 periods. Due to work schedules of the participants it was not feasible to get the same number of samplings in each time period. Though the data with and without glasses is directly comparable within a time period on a given day, no comparison should be made to data between time periods. Numbers represent the total tally of assigned values (0, 1, or 2) related to observed changes in a time period.

Sun angle in relation to observer
The data represented in Figure 2 (next page) indicates that when wearing the glasses there is little difference between facing the sun and having the sun at your back. However, discussion with the samplers indicates that the brightness of the color difference is more intense when the sun is at your back.

Sun angle in relation to observer.
Date Glasses Facing
the sun
Sun
behind
Sun
over-head
**
Sun
to
side
(left
or
right)
*
E
AM
W
PM
W
AM
E
PM
NESW Midday S/N AM S/N PM
7/14/2003 Without 5 19 4 19 10.25 6 26
With 5 21 8 20 13.75 7 26.5
7/15/2003 Without 36 25 46 19 50.25 45.5 26.5
With 41 28 44 28 56 46.5 33
7/16/2003 Without 99 46 109 48 73.25 109 49
With 91 44 114 43 59 108.5 53
7/17/2003 Without 109 60 112 63 85.25 113 61
With 94 54 94 54 78.25 98.5 57.5
* Total divided by 2 because of 2 directions.
** Total divided by 4 because of 4 directions.
Figure 2. The data was segregated by time periods of daylight hours with samplings up to 11 o’clock representing AM and samplings after 2 o’clock representing PM. From 11 to 2 o’clock is considered midday where the sun is overhead. Since the stress in the treated plants increased as time passed each day, the data between AM, midday, and PM are not comparable. What can be compared are the data within the same time period by direction and the data with and without glasses within the same time period and direction.

Because the stressed plants changed over time it is not reasonable to compare the midday readings with other data taken at different times. What this information shows is that during the initial stages of plant stress (first two days) there is a better chance of detecting stress with the glasses.

The data indicates that having the sun at the side increases the chances of observing plant stress compared to either facing the sun or having the sun at your back. This does not appear consistent with the verbal opinions of the observers. This difference is probably attributable to layout of the plants being sampled. The east/west axis of the layout was approximately twice as long as the north/south axis. This probably caused screening of some plants by others in front of them. The layout of the plant groups is indicated below.

North
West Group A Group B Group C Group D East
Group E Group F Group G Group H
South

Follow-up field evaluation
The SDTDC testing of stress detection glasses was on small plants. In essence all observations were looking straight ahead or down. What differences would there be when looking up into a tree canopy? The San Bernardino National Forest was visited and the glasses were evaluated on trees affected by drought/bark beetles.

Small tree at eye level

Stress detection glasses increase the color contrast of dead/dying foliage. Looking toward the sun reduces the contrast. Viewing with the sun to your back makes the color contrast of dead/dying vegetation the brightest.

Large trees looking up

Looking up into crowns at a 30 to 45 degree angle is similar to the results at eye level. However, when the sun is higher in the sky the contrast is reduced if looking at the tops of the trees. This is probably because you are again looking toward the sun.

Landscape scale viewing

On a landscape scale, dead/dying trees really stand out compared to the naked eye. When the glasses are removed the trees can still be picked out, but not as easily. The stress detection glasses could be very useful when trying to estimate mortality at long distances, such as from ridge to ridge.

General observation
The stress detection glasses will help pick up changes to the color of the foliage. As needles or leaves start to fade before normal cast-off they become very evident when using the glasses. Insect damage such as needle miners becomes a lot more evident. Trees exhibiting chlorotic foliage from mineral deficiency or disease will appear brighter through the stress detection glasses.

Just because there is an observable effect does not necessarily mean that a tree is dying. The stress detection glasses could be useful in helping detect possible mortality during the early stages of bark beetle or other insect attack. However, they should only be used to draw your attention to the possibility of a problem. The tree should not be designated for removal based on the stress detection glasses alone. Closer inspection should be used to determine what is causing the color change.

Opinions of Samplers and Others

The people doing the sampling provided verbal and written observations and opinions of the stress detection glasses. Other persons that have tried the stress detection glasses provided some additional input/opinions. These are summarized below.

Sampler's Comments

It became evident that the dead or dying plants had a more orange or red appearance.

The signs of mortality were more evident without glasses than with the use of the glasses.

On the first day of the experiment I was able to detect stress in two plants that I could not see with the naked eye. As the test went on, many of the plants showed noticeable signs of stress with the glasses, and no signs without the glasses.

Group D showed plants having signs of mortality but this was also detected without the glasses.

....when the sun was at your back and directed on to the plants it enhanced the colors more.

The glasses did enhance the mortality but the mortality could be detected without the glasses.

The usefulness of the glasses for detecting stress varied by type of plant, time of day, direction, and cloud cover.

Group F showed more yellowing but showed better without the glasses.

The colors were more enhanced with the glasses.

There were certain times of the day that the glasses seemed most effective. During the early afternoon and late evening the glasses worked better than they did in the mid-afternoon. The light colors were more apparent in the early morning and late afternoon. There was not this distinct difference when it was the middle of the day.

Overall the glasses helped me detect stress in the plants that I was not able to see with the naked eye.

Although the plant stress could be observed at about the same time with and without the UV glasses, the stress was more apparent with the glasses for some plant types.

My observation is that a user may find these glasses helpful in detecting mortality. However, there will be a learning curve in seeing these differences.

These glasses may only be good for specific species.

The glasses tended to conceal stress in plants with yellow-green leaves.

I would only use these glasses in conjunction with other science.

I think the glasses are a useful tool for some types of plants, not necessarily to detect stress sooner, but to see it easier.

Others

Bill Radtke, Michigan Department of Natural Resources. Surveying a variety of causal agents in both hardwood and conifer forests.

They did seem to be useful. After a period of acclimation and comparison I did learn to interpret the information garnered via the glasses.

Our application requires the constant transfer of information to a quad map and soon to a computer tablet. The stress detection glasses are so dark that constant removal or lifting of the clip-on lenses is required. This problem could be eliminated by leaving the bottom half or "bifocal" area of the lenses clear.

Bill Befort, Minnesota Department of Natural Resources. Sketch-mapping during aerial surveys.

Entire landscapes viewed from the air are far from uniform; there is a very wide range of perfectly normal variation and contrast, unrelated to plant damage, that must be instantly and continuously recognized and factored out by the observer as he flies along. The symptoms of vegetation damage are not the strongest contrasts in the general landscape, but some of the subtlest, and they are also far from uniform. Because stress detection glasses substantially alter the visual appearance of the landscape, they add a great deal to the baseline mental interpretation workload that the observer must handle merely to stay oriented and factor out inessentials, while any enhancement they provide to vegetation stress symptoms is likely to be lost in the clutter. In brief, they create more problem than they solve.

I've settled on amber (not yellow) "blue-blocker" sunglasses for sketch-mapping.

Keith Windell, Missoula Technology and Development Center. Evaluating fire damaged trees.

In the very short period of time I had them on (about 5 or 10 minutes per session) my eyes never did totally adjust to the color scheme (rose) you see through the glasses. I started to feel disoriented (and even sick) so discontinued using them.

The dead needles in the tree crowns stood out but at 50 to 100 feet I didn't feel like I could make a decent call on whether the trees were mortally stressed or not.

Although an experienced user of these glasses would no doubt be more confident with their decisions (leave vs. cut) I think there will always be a high degree of subjectivity in this approach....

Conclusions

The value of the stress detection glasses appears to be in the way they amplify a person's color perceptions. Determining what the observations mean will take some experience by comparisons with known plant stresses. Comparing different types of vegetation may be difficult. Changes in vegetation that is naturally a yellowish color may be difficult to detect. The glasses amplify color changes but give no indication of what is causing the change. As an example, natural color changes in leaves at the beginning of fall are significantly enhanced but the vegetation is probably healthy.

Some users find the glasses somewhat disorienting or upsetting. Others seem to have no problems, at least using them for short periods of time.

The ability of the eye to detect plant stress problems seems to increase with experience. A novice may not notice minute changes that the seasoned veteran may think are obvious. For this study, testing was of short duration. The ability to detect changes with the glasses may increase with more experience.

The glasses do appear to have utility for forestry work. Some experience with the glasses appears to help when evaluating what one sees. They may not provide magic answers as to whether vegetation is about to die but they do amplify a person's perceptions. They are a relatively inexpensive, small, and lightweight tool. They can easily be carried to the field and can aid in quickly picking out vegetation changes.

Comparison of Stress Detection with Other Sunglasses

Though not part of the original project, the NASA stress detection glasses were compared to other readily available sunglasses. It has been noted that amber lenses can detect stress related changes in plants. Are these detectable changes comparable to the NASA glasses? Do other colors work?

Four pair of sunglasses were compared to the NASA glasses where the changes in the plant were already apparent. Colors evaluated were blue, green, amber and brown.

The colored glasses appear to enhance color differences that are the same color as the lenses. The blue and the green did not seem to do much to enhance visibility of changes in plants due to stress/mortality. The amber glasses did the best job of the sunglasses particularly where the foliage turned orange or brown. The brown glasses did amplify some changes but not to the degree of the amber glasses. Wade (1997) found that among the most effective glasses were those that block the blue and haze that comes from the scattering of blue light under hazy conditions. He also found that the inexpensive variety of blue-blocking glasses was as effective as the expensive one.

However, compared to the NASA stress detection glasses, none of the sunglasses detected as wide a variety of change nor did they adequately detect early changes like the NASA glasses did.

Schultz (2001) tested 19 lenses of various colors, including the stress detection glasses evaluated in the paper. His test used ponderosa pine. He found that the stress detection glasses showed no particular advantage at detecting off-color ponderosa pine. He also found that the stress detection glasses were too dark to conveniently or safely wear while driving or walking under forested conditions.

References
  1. Brock, R., 2002, Stress Detection Glasses, History and Additional Information
  2. Rankin, L., Heath, J., and Murtha, P., 2000 Proceedings 22nd Canadian Symposium Remote Sensing
  3. Schultz, D., 2001, Evaluation of Lenses to Detect Stress (FHP Eval. No. N01-06 Letter of USDA Forest Service
  4. Wade, G., 1997, Amber Sunglasses Enhance Foliage Color Discrimination USDA Forest Service paper, NE-INF-131-97, from the Northeast Forest Experiment Station
Sections