2001 Red Edge Inflection Point vs. Ozone

     Since Forest Watch began in the 1991-92 school year, the scientific focus of the program has been to monitor the effects of ground level ozone on the health of white pine trees. To accomplish that goal, we have assessed a specific set of trees for the duration of the program. These trees are at five specific sites in NH, listed below Figure 1. Foliage samples from trees at each school have been collected yearly and assessed for ozone related symptoms.

A subset of each sample is sent to the University of New Hampshire where they are scanned by a spectrometer called the Visible InfraRed Intelligent Spectroradiometer (VIRIS). The VIRIS gives us almost 600 bands of data to work with and this allows us to derive many important index values that relate to the health of the trees. The main index that has been used is the red edge inflection point (REIP), which is a first derivative value calculated along the red edge region of the spectrum (680-750 nm). Plants with less healthy chlorophyll will show a lower REIP as the red edge moves toward the blue part of the spectrum and plants with more chlorophyll will have higher REIP values as the red edge moves toward the longer wavelengths. Figure 1 shows the familiar REIP / Ozone graph.

Figure 1 : Red Edge Inflection Point (REIP) vs. Average Maximum Monthly Ozone values (June – August).

The familiar Red Edge Inflection Point (REIP) vs. Ozone graph (Figure 1) shows the trends that have occurred since the summer of 1991. The REIP data for 1991-1999 came from the six original schools that have been involved since 1992. For the 2000 and 2001 needle years, one of those schools was not able to submit data, so the REIP was calculated using five schools for those two years:

Yearly ozone is compared to REIP values from 1991-2001. Originally there were six schools used in the analysis, but for 2000 and 2001, only five of those schools were able to submit samples. A list of schools used in the analysis is shown below the graph. The yearly REIP values were calculated as the average of all trees for each school. Also listed are the seven NH ozone stations where the ozone data was measured during the growing season. The ozone data was acquired from the New Hampshire Department of Environmental Services in an hourly format. Each yearly value is the result of averaging the maximum monthly values of June, July, and August.

From 1991-1998 there was a strong inverse correlation between ozone and REIP. In 1999, that correlation did not continue. Ozone and REIP both increased dramatically. Some hypotheses for the unusual results included: 1) instrument change resulted in inconsistent values, 2) due to dry conditions in the summer of 1999, stomates closed during the peak ozone periods, and 3) wet conditions in the spring of 2000 resulted in a recovery of the 1999 needles. After performing some tests, we decided that hypothesis #2 was the most reasonable explanation. The 2000 needles showed very little difference in health, even though the ozone levels were very low. The high REIP value in 2000 followed the previous trend of low ozone and high REIP, but this still did not explain the unusually high 1999 value. More data was needed.

The 2001 needles have now been added to the graph and you can see that there is a slight drop in REIP, corresponding with a distinct increase in ozone levels. In fact, ozone levels in 2001 turned out to be comparable to 1999 values. The REIP value for 2001 was 720 nm, down from 722 nm in 2000.

Although the decrease in REIP from 722 to 720 does not seem very big, it is a drop!! Looking at the values for each instrument, it appears that there is more variability between values from 1998-2001 (new VIRIS) than from 1991-1997 (old VIRIS). This might suggest that there is more sensitivity in the new instrument. However, the error bars suggest that there is less variability between schools for each year. In any case, the new instrument has not measured the REIP levels expected with elevated ozone levels. More testing needs to be done with the VIRIS to see what the sensitivity is and what other indexes might prove useful. If the sensitivity of the instruments is different, it may be better to plot separate graphs for needles measured with the old VIRIS and for those measured with the new VIRIS. If the new instrument is giving higher values, which it seems to be, they still should be comparable in terms of the relative changes per year. We are experimenting with several other indexes (See Figures 6 & 7) as well to see what else may provide a good health comparison with changing ozone levels. Stay tuned!!

The graph in Figure 2 shows yearly average REIP values for all participating NH schools plotted against ozone levels from NH stations. As expected, the REIP values are very similar to the values of the original six schools. There are some minor differences such as the 1997 REIP value, which showed an increase in health with higher ozone levels while the graph of the original schools illustrated a decrease in health with the increase in ozone. The correlation appears to be stronger with just the original schools in the analysis.

Figure 2: Ozone vs. REIP – All NH Schools

Similar graphs will be produced later this year for other states when ozone data has been compiled for each state.

This information is also in the 2001-02 Forest Watch Databook.If you have questions about the REIP vs. Ozone analyses please contact Forest Watch at 603-862-1792 or email this web site.