BRDF Applications in Semiarid Grassland Monitoring with the AVHRRs

---

---

Why bother with the AVHRR? Its a meteorological sensor isn't it?

With a number of all-new, all-singing-and-dancing sensor systems either in orbit or about to become operational, why bother with a sensor which was originally designed mainly to look at clouds? There are many reasons :

• First, geographical coverage is exceptional. With an orbital period of ca. 102 minutes every location on the globe is sampled at least twice in any 3-day period and there are usually two instruments operating at any one time : currently one on board NOAA-12 in an AM orbit and another on board NOAA-14 in a PM orbit; NOAA-15 (NOAA-K prior to launch) with a third generation AVHRR sensor was put into orbit on 13th May 1998 and will shortly take ovr from NOAA-12.
  
  
• Second, temporal coverage is exceptional both in terms of the length of time AVHRRs have been flown (since the late 1970s) and in terms of the revisit period. High resolution sensors cannot compete on this basis; a revisit period of 16 days is too long to wait for a scene which may be 50% cloud! Furthermore, we can't go back in time with new sensors, which means that long term retrospective monitoring is only possible with the AVHRR.
  
  
• Third, the sensor is operated by NOAA and is paid for by the US taxpayer; data from it is very inexpensive and is archived by ground receiving stations worldwide. Furthermore, the NOAA/NASA POES satellite programme is ongoing and new AVHRRs will continue to be launched for the foreseeable future (see POES Launch Schedule and here; note also that post-2000 co-ooperation between NOAA and Europe in sustaining the meteorological satellite series has been agreed; see here) and also this overview.
  
  
• Fourth, since the AVHRR scans at large angles relative to the surface (>56 degree local incidence angles) and since there are two such instruments operating in (local) morning and afternoon orbits, it allows us to sample the surface BRDF, albeit sparsely, with a distribution which is good enough for the purposes of inverting simple linear BRDF models. The range of solar zeniths obtained using the AM sensor extends to over 70 degrees at northern latitudes in the (local) summertime and this is important for model inversion.
  
  
• Fifth, the AVHRR provides a means of testing candidate BRDF models in real large-scale applications, which is not yet possible with other sensors (although this will soon change).