In 2019 the Canadian Space Agency initiated development of a dedicated wildfire monitoring satellite (WildFireSat) mission. The intent of this mission is to support operational wildfire management, smoke and air quality forecasting, and wildfire carbon emissions reporting. In order to deliver the mission objectives, it was necessary to identify the technical and operational challenges which have prevented broad exploitation of Earth Observation (EO) in Canadian wildfire management and to address these challenges in the mission design. In this study we emphasize the first objective by documenting the results of wildfire management end-user engagement activities which were used to identify the key Fire Management Functionalities (FMFs) required for an Earth Observation wildfire monitoring system. These FMFs are then used to define the User Requirements for the Canadian Wildland Fire Monitoring System (CWFMS) which are refined here for the WildFireSat mission. The User Requirements are divided into Observational, Measurement, and Precision requirements and form the foundation for the design of the WildFireSat mission (currently in Phase-A, summer 2020).

Details on the first linear arrays of 512x3 VO x microbolometers operating in space are reported. Arrays of this format are suited for remote sensing where relative motion between the spacecraft and target provides an inherent scanning mechanism. To take full advantage of the linear format, the array is built on a custom readout electronics that enables simultaneous integration of all pixels for scanning periods of up to 140 ms. The output signal from each pixel is digitized to 14 bits using a voltage-to-frequency conversion mechanism. Two arrays, integrated into two spectrally distinct radiometric packages, provide for coregistration of infrared images in three bands centered at 3.8, 10.85, and 11.85 μm for the retrieval of fire and sea surface temperatures. Analysis of the downlinked data confirms the reliable in-orbit operation and consistency with pre-launch characteristics for both arrays. Algorithms have been developed to perform post processing and absolute radiometric calibration of images in all bands. Image deconvolution using Wiener filtering was found effective in recovering the signal loss incurred in the active pixels when observing high temperature events. The in-flight gain and offset values were evaluated for all pixels by means of deep space measurements and cross calibration with reference spaceborne sensors. Preliminary assessment of the images calibrated using these values showed that they are in agreement with those retrieved from GOES sensor.

Focal planes of 80x60 VO x microbolometers with pixel pitch of 104 µm were developed in support of the remote sensing of ice clouds in the spectral range from 7.9 to 50 µm. A new design that relies on the use of central posts to support the microbolometer platform was shown effective in minimizing the structural deformation usually occurred in platforms of large area. A process for goldblack coating and patterning of the focal plane arrays was established. It was found that the goldblack absorbs more than 98 % and 92 % of incident light respectively at wavelengths shorter and longer than 20 µm. Moreover, a spectral uniformity of better than 96 % was achieved in all spectral channels required for the measurements. The noise figures derived from the data acquired over short periods of acquisition time showed the evidence of a correlation with the format of the addressed sub-arrays. This correlation was not observed in the data acquired over long periods of time, suggesting the presence of low frequency effects. Regardless of the length of acquisition time, an improvement of noise level could be confirmed when the operating temperature was increased. The dependence of responsivity on sub-array format and operating temperature was investigated. The noise equivalent power derived from this study was found to be in the range from 45 to 80 pW, showing that the far infrared focal plane arrays are suited for use in the intended application.

No abstract

No abstract