3
$\begingroup$

It would be good to have satellites that gave current readings of skin temperature so that farmers could see if their crop was freezing or too hot, as well as for other global warming monitoring. Would this be possible, why isn't there one?

$\endgroup$
13
  • 1
    $\begingroup$ I would think that's what infrared satellite can be used for if there's no clouds. I'm pretty limited in satellite knowledge, but not sure if there's any band to specifically to give skin temperature if there's not. $\endgroup$ Apr 4, 2022 at 5:22
  • 1
    $\begingroup$ How would the skin temperature of a plant differ from the air temperature at the location of the plant? If there isn't a difference, it would be easier & cheaper for the farmer to establish a system of temperature sensor around the farm that could be monitored via a screen or via a computer in the farmhouse. The computer could sound an alarm if the temperature varies from specified parameters. $\endgroup$
    – Fred
    Apr 4, 2022 at 5:49
  • 1
    $\begingroup$ Ah, indeed, guess IR doesn't entirely go to true surface then. I do see a site over here in the western Hemisphere that offers land surface derived temp for GOES: weather.cod.edu/satrad/… But can't find anything for Europe, only land.copernicus.eu/global/products/lst... which seems to be backed from the EU and include Eurosat, but unfortunately, as can often be the way over there, such data is not free (as is the case with the bulk of the European weather model) :-/ Maybe someone has something better. $\endgroup$ Apr 4, 2022 at 9:18
  • 1
    $\begingroup$ @JeopardyTempest Even with MTG-FCI, most IR channels have a resolution of 2 km SSP (worse in Europe), with 2 channels at 1 km SSP, in the absence of clouds only. Probably not good enough for farmers. $\endgroup$
    – gerrit
    Apr 5, 2022 at 8:20
  • 1
    $\begingroup$ sciencedirect.com/science/article/abs/pii/003442579490023X $\endgroup$ Apr 5, 2022 at 14:47

2 Answers 2

4
$\begingroup$

It would be good to have satellites that gave current readings of skin temperature so that farmers could see if their crop was freezing or too hot

This is not generally possible in near real time ("live update"), but satellites may still offer some useful information. There are several problems:

  • Satellites can measure skin temperature using infrared sensors only in the absence of clouds. On average, 67% of Earth and 55% of land is covered in clouds (King et. al, 2013) (of course, this number various strongly with time and place). Therefore, much or most of the time, satellites cannot measure skin temperature instantly, but have to wait for a cloud-free moment. However, dry nights (no cloud and not much water vapour) cool down more than humid or cloudy nights, so maybe during the most at-risk nights the radiometres can still give a clue.
  • To monitor in near-real time, you need to use either geostationary satellites or a very large swarm of low Earth Orbit satellites.
    • The spatial resolution of infrared channels in geostationary satellites is too poor. Most geostationary imagers have resolutions of more than a kilometre. The current Meteosat generation has 3×3 km² at sub satellite point and closer to 4×6 km² in central Europe.
    • A very large swarm would be costly, and the infrared resolution would still be dozens of metres.
    • High resolution imagers in any case only regard a small area at any time.
  • In theory, microwave can be used to measure surface temperature in the presence of clouds (although for a thicker layer than infrared), but only with a good knowledge of surface emissivity (which is a function of soil moisture, among other things), and the spatial resolution of microwave sensors is much worse (>10 km from low earth orbit, non-existent from geostationary orbit).

A resolution of 1×1 km² from geostationary orbit would already be very ambitious, where 2×2 km² or more is more realistic. I'm not sure if this is good enough for farmers for discriminating the skin temperature of their crops from the skin temperature of the general area, which may be quite different (heat loss in a clear night strongly depends on surface material).

Farmers who wish to monitor their crop temperature in real time should probably just use infrared cameras installed at their property or possibly with drones constantly flying in the air. Whether this is cost-effective is beyond the scope of Earth Science SE.

as well as for other global warming monitoring

This is routinely done, but this does not need a high spatial resolution and it does not need to be in near real time.


M. D. King, S. Platnick, W. P. Menzel, S. A. Ackerman and P. A. Hubanks, "Spatial and Temporal Distribution of Clouds Observed by MODIS Onboard the Terra and Aqua Satellites," in IEEE Transactions on Geoscience and Remote Sensing, vol. 51, no. 7, pp. 3826-3852, July 2013, doi: 10.1109/TGRS.2012.2227333. Accessed online 5 April 2022.

$\endgroup$
8
  • 1
    $\begingroup$ "On average, in the order of 70% of Earth is covered in clouds." - Citation needed. $\endgroup$
    – Erik
    Apr 5, 2022 at 11:53
  • 2
    $\begingroup$ @Erik Citation added (67% rather than 70%, close enough). $\endgroup$
    – gerrit
    Apr 5, 2022 at 12:01
  • 1
    $\begingroup$ One thing worth pointing out is that those worried about borderline freezes will often be favored to get data, as temperatures drop off (and frost forms) best in clear skies. Here in Florida our danger nights to crops are almost always clear. Experientially don't know if that's entirely true in places that are often nearer the wraparound moisture nearer the low pressure up more into the midlats, though imagine it is, because the first fronts of the year would feasibly tend to have lows more poleward likewise. And of course such areas will be less a question of if than of when. $\endgroup$ Apr 5, 2022 at 17:26
  • 1
    $\begingroup$ Also I think I tend to agree with trond in the end that 1 km resolution can still be quite good for farmers to get a rough idea of impacts, though of course may miss microscale features and such. $\endgroup$ Apr 5, 2022 at 17:29
  • 1
    $\begingroup$ @JeopardyTempest That's true, clear nights are colder nights. Maybe my answer is too pessimistic. I've changed the formulation a bit. For a huge farm in (say) Brazil perhaps it's valuable (although I don't know if the information "your crops may freeze in the next 2 hours" is actually actionable for such a farm, but that's another question). $\endgroup$
    – gerrit
    Apr 5, 2022 at 19:49
2
$\begingroup$

Not, because the costs of satellite image provider are excessive, and the same idea can be exploited with drones (and in fact it is being commercially exploited, as it can be seen by accessing this report https://www.businessinsider.com/agricultural-drones-precision-mapping-spraying?r=US&IR=T ).

I suggest to have a look at this study: https://www.researchgate.net/publication/341253627_Applications_of_UAV_Thermal_Imagery_in_Precision_Agriculture_State_of_the_Art_and_Future_Research_Outlook to understand where we are in 2022 in terms of implementation of remote sensing and precision agriculture (we are quite ahead wrt to what OP question implies).

$\endgroup$
5
  • $\begingroup$ You mention drones being commercially exploited, it would help more to give examples. Also may help to give a rough vision to the costs being excessive, rather than assume everyone would immediately agree with that idea. Not saying your wrong, but details help give credence and support to ideas like these :-) $\endgroup$ Apr 5, 2022 at 17:33
  • 1
    $\begingroup$ @JeopardyTempest added references $\endgroup$
    – EarlGrey
    Apr 5, 2022 at 21:43
  • $\begingroup$ Did you see about the Sentinel 3 ESA satellites? They are global and give a resolution of 0.3'C and 300m, 420-1030Nm range $\endgroup$ Apr 6, 2022 at 2:02
  • $\begingroup$ @EarlGrey awesome, that really lays weight behind your position, and is the kind of answer I think we in like math and science Stack Exchanges really enjoy, I sure do. Thanks for taking that next step :) $\endgroup$ Apr 6, 2022 at 4:23
  • $\begingroup$ @LifeInTheTrees there were some momentum around Sentinel-2, with even some infrastructure aimed specifically at agriculture, like this: github.com/Sen2Agri I am not updated in Sentinel 3, they provide daily data but if there are clouds the moment they should be above your field it is no-data day (so to be complemented by more expensive, at least on paper, drones? I do not have an opinion). $\endgroup$
    – EarlGrey
    Apr 6, 2022 at 6:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.