measurements are better done by recording both fluorescence decay and elastic back-scattering signals from the 
same laser shot (or from two consecutive shots). The light source must satisfy several constraints, including 
directivity, pulse duration <100 ps as required for sub -nanosecond lifetime measurements, wavelength 
matching the absorption of photosynthetic pigments, eye safety and energy [7]. Excitation wavelengths near (but 
below) 400 nm is the best compromise between eye-safety requirements and the efficiency of chlorophyll 
excitation. An additional limitation appears when using short light pulses: since non linear effects (singlet- 
singlet annihilation) occurs when the number of photons exceeds ~2 lO 1 ^ photons/pulse/cm 2 (i.e. 
10|iJ/cm 2 /pulse at 355 nm) [22], the density of energy must be bound by this value. A telescope with a large 
aperture is necessary to collect efficiently both fluorescence and back-scattering responses. We describe in the 
following a new picosecond LIDAR system developed by the LURE laboratory for remote fluorescence lifetime 
measurements of plant canopies [23]. 
Fig. 3 shows a scheme of the system. A detailed description of the calibration tests of this LIDAR has 
been published [28]. The excitation unit consists in a flash-pumped frequency-tripled mode-locked Nd-Yag laser 
(Quanta Systems, Milan). The frequency was set to 10 Hertz, the pulse duration was = 50 ps (FWHM) and the 
energy up to 10 mJ per pulse. This laser was kindly lent to us by the IRSA laboratory of the JRC (Ispra, Italy). 
The collecting optics consists in a Fresnel lens of 380 mm diameter with a focal length of 400 mm. The optical 
system focuses the fluorescence into the detector through a set of filters. The fluorescence (670 <X< 750 nm) 
and the back-scattering (interference filter at 355 nm) signals are measured alternatively. The laser beam is 
made coincident with the optical axis of the Fresnel lens by two mirrors. The detector consists in a Sylvania 502 
crossed-field photomultiplier with a rise and fall time of <150 ps. It provides a signal of a few hundreds mV, 
when loaded with 50 Ohms. This signal is directly fed to a Tecktronix SCD 1000 transient digitizer and further 
transferred to an HP 9816 computer through an IEEE 488 interface. The whole system has a bandwidth of 1 
GHz. The detector is protected from the ambient light by a mechanical shutter that opens during a time window 
of =3 ms synchronized with the laser shot. 
5. EXPERIMENTAL RESULTS 
5.1. Effect of light intensity. 
FIG. 4. Fluorescence decay measured with the LIDAR system, in the dark (Fo) and under saturating light 
conditions (Fm) 
In order to test the ability of our LIDAR system to detect changes in the fluorescence quantum yield, 
experiments have been carried out at different light intensities. Figure 4 shows the fluorescence decay of a 
single maize leaf excited by a single-shot laser pulse. It also shows the instrumental response function recorded 
by looking at the back-reflected light, at the same wavelength as the excitation pulse. Although the laser 
intensity is adjusted near, but below the threshold for non-linear phenomena, no significant actinic effect is 
observed, as a result of the very short pulse duration used and the relatively large illuminated area (=35 cm z ). 
Thus the LIDAR system is able to work at distance, (15 m in the present experiment), even in the presence of a 
continuous saturating light, like a commercial pulsed fluorimeter at near contact Two different experimental