Investigations on Frequency and Energy References for a Space-borne Integrated Path Differential Absorption Lidar

Kurzfassung

The Integrated Path Differential Absorption Lidar (IPDA) technique using hard target reflection the near IR has the potential to deliver CO2 column measurements from space with unprecedented accuracy which is a prerequisite to understand the sources and sinks of this dominant anthropogenic greenhouse gas. The observational needs, however, demand for very stringent system requirements, of which two were thoroughly investigated. The first is the online frequency accuracy. With a sub-100-kHz requirement for a system operating at 1.57µm, using a ~10000x wider, Doppler-broadened CO2 line as absolute reference appears difficult. At LTF, a compact frequency reference breadboard was designed using an alternative approach: A DFB laser emitting at 1.56µm is frequency-doubled and absolutely stabilized to a miniaturized Rubidium reference (780nm). The frequency stability of the 1.56-µm laser was measured by beating against a self-referenced comb, showing an Allan deviation of always less than 20kHz and below 2kHz between 1E3s and 2E5s. To bridge the gap between 1.56µm and the CO2 absorption lines at ~1.572µm, an optical frequency comb (OFC) was realized by modulating part of the DFB laser light with a waveguide electro-optical modulator enclosed in a Fabry-Pérot cavity. A second DFB laser is planned to be offset-locked to an adequate line of the OFC, to serve as reference for monitoring the frequency of the lidar transmitter laser, or as a stable source for injection seeding. From preliminary OFC stability measurements, the frequency stability of this offset-locked laser is not expected to noticeably deviate from that at 1.56µm. The second requirement, looked after by DLR, relates to the monitoring of the on-line to off-line ratio of outgoing pulse energies, whereby accuracy of the order of 1E-4 is needed. This is imposed by the IPDA technique which, contrary to the range-resolved differential absorption lidar (DIAL) technique, is not self-calibrating. The preferred set-up would use an identical detector for both lidar signal and pulse energy monitor to circumvent potential detector ageing effects. It is thus important to employ pick-up, attenuation, and transmit schemes that do not alter the energy ratio. For this purpose, integrating spheres are insensitive to beam pointing and intensity profile variations and can be fibre-coupled. However, given the coherence of the lidar transmitter, speckle effects prove to be the main limitation. It is shown that the speckle patterns at the exit port of integrating spheres of two subsequent pulses within a double-pulse sequence of ~400 µs are partly correlated. This correlation, which generally leads to systematic deviations in the energy ratio, needs to be reduced to meet the requirements. The use of larger detectors capturing a higher number of speckles leads to a lower systematic error, but calls for a departure from the concept of a single detector for receive and calibration branch. By using pyroelectric energy detectors at the exit port of integrating spheres the relative systematic error of the energy ratio of double pulse pairs could be reduced to ~3E-5 over time scales of a few hours. This work is supported by ESA and Swiss National Science Foundation.