SMOS catches rare events

Category : Data, L1

By Joaquin Munoz Sabater (ECMWF) and Fernando Martin Porqueras (IDEAS)

The 18 of April 2011 a strip with unusual values of brightness temperatures was detected in Western Sahara and Morocco. This feature looked strange enough to the Quality Check team Given at ESAC that they asked US what we thought about it. Many options were quickly discarded (RFI, Instrument problem etc…). The problem was that the measurements as shown on the QC picture below pointed towards high soil moisture which is not all that expected at this period of year in this area with such a linear shape.

morocco

Fig 1 QC picture on two orbits (Ascending and Descending) on April 18th 2011 from DPGS (X pol on the left and Y on the right)

However, between the 17 and 19 April 2011 a rare precipitation event took place in Western Sahara. Precipitations up to 10 mm, locally stronger, took place in a narrow, well defined band through the desert. This was caused by the influence of a low placed in the Azores which produced unusual precipitations in the Sahara area. This precipitation band moved progressively towards the North-East direction. Figures 1-4 show the cumulated total precipitation during 48 hours, in steps of 12 hours, starting the 16 April 2011. These figures clearly show the direction and cumulative values of the precipitation band. Non-coincidentally, there is a very good correlation between SMOS measured brightness temperatures (being much colder than its surroundings) and these precipitation figures, which explain these abnormal values. Figure 5 shows the ECMWF soil moisture analysis the 18 April 2011 at 12h00, for the first layer (7cm), which also displays this geographical strip being wetter (up to 20%) than its very dry surroundings. SMOS was able to catch very well this precipitation event.

step48 step49
step50 step51
SM1 SM2

These results demonstrates once more the good skills of the SMOS instrument to clearly catch rare precipitation events.

Comparison of SMOS Sea Surface Salinity with SSS from in situ drifters in the tropical Atlantic Ocean

Category : Cal/Val, L2

Comparison of SMOS Sea Surface Salinity with SSS from in situ drifters in the tropical Atlantic Ocean evidence that SMOS senses large salinity gradients, but contamination by land remains

J. Boutin, G. Reverdin, N. Martin, S. Morrisset (LOCEAN)

In order to dispose of salinity measured as close as possible to the sea surface for validating SMOS SSS (representative of the 1cm of the sea surface), numerous surface conductivity/temperature drifters have been deployed in the frame of the french CNES/TOSCA program (GLOSCAL project). Data recorded by these drifters can be viewed on: http://www.locean-ipsl.upmc.fr/smos/drifters/.

Two drifters were deployed this summer in the tropical Atlantic showing large contrasts of SSS linked to the fresh water discharged by the Amazone river and transported northward by the north Brazil current:

fig1

SMOS SSS retrieved with the L2 OS processor were colocated at +/-25km, +/-2 days from these drifters. SMOS SSS were averaged around each drifter data, the average being weighted by the error on individual retrieved SSS and by their mean resolution.

Figures below illustrate that on descending orbits, SMOS well sees the salinity contrast between the two drifters . Close to the coast, SSS on ascending orbits are clearly out of range.

fig2

Figure 2: Comparison of SMOS SSS (retrieved with L2 OS processor, model 1, green: on descending orbits; red: on ascending orbits) with SSS recorded by drifter Metocean 66270 (around 36psu) . West of 40W, SMOS SSS are noisier, especially on ascending orbits, illustrating that a pollution by land is still present.

Fig3

Figure 3: Comparison of SMOS SSS (retrieved with L2 OS processor, model 1) with SSS recorded by drifter Metocean 63220 (SSS between 29 and 32psu), that remains close to the coast. While on descending orbits (green points), SMOS SSS are consistent with the drifter measurements, SSS on ascending orbits are well above the expected values.

SMOS is getting closer to the mark

Category : Cal/Val, L2

SMOS is currently undergoing Cal Val Activities and in this framework at CESBIO we are busy-with colleagues everywhere to compare SMOS data with ground data.

In particular, with colleagues in the US we have looked at the SCAN data and several very interesting results are emerging. Below a comparison done by Ahmad AlBitar at CESBIO – CATDS between SMOS and SCAN network data.

SMOS ECMWF and SCAN

This figure shows a comparison between 3 sources of measurements. In the top panel Ahmad has plotted the absolute soil moisture from ECMWF (blue) Scan (site 2001 in green) and SMOS (red). SMOS still underestimates the others but we can see that the temporal behaviour is very well captured. This is very promising for some applications for SMOS.

The middle panel shows the retrieved vegetation opacity. It starts in May only as it was then that the proper algorithm was implemented at ESAC.

Note also, in the lower panel where the temperatures plotted the evolution from the first days to today. This clearly emphasizes the improvements carried out on the calibration etc (mainly at level 1 thus). Of course the temperature retrieval is not very often carried out (see ATBD) so only a few points are present but all those for July august are completely embedded in the ECMWF and SCAN curves.

Just to show the SMOS ability to capture dry downs, Ahmad has stretched the scales of the three data sets beween 0 and 1 (the so called wetness index!) as shown below. With this trick, we see how well SMOS captures the events.

Absolute (top) and strectched (bottom) soil moisture / wetness indices

Absolute (top) and strectched (bottom) soil moisture / wetness indices

Of course we can still see the improvements a time goes since level 1 and level 2 teams did work a lot during commissioning phase and after to improve the calibration and algorithms. So the real part to be studied is from end of May to today.

I must also say that this example was obtained after selecting SCAN SNOTEL sites (Northern US as we are also looking at freeze thaw/ snow) which were « nice » (i.e., rather flat and homogeneous, minimum forest cover, low RFI), and – of course – I selected one of the best of Ahmad’s plots!

Comparison of SMOS Tb with model1 Tb

Category : Commissioning phase, L1

1. Overview
One ascending orbit of dual polarization (No. 251) and one descending orbit of full polarization (No. 388) are chosen for comparisons of Tb_measurement and Tb_model1 under antenna frame.

 

The ascending orbit of dual polarization (No. 251) and three zones for comparisons of Tb.
Figure 1: The ascending orbit of dual polarization (No. 251) and three zones for comparisons of Tb.

Information of this orbit is given in table 1.

Table 1
Ascending/Descending: A
Orbit No: 251
Validity time: 1119T132314 ~ 1119T141715
Data version: 321_002

2. Ascending orbit of dual polarization (No. 251)
The zone of the ascending orbit is shown in Figure 1.There are 100 snaps (ID:2515408~2515527, Latitude: -44.5~-37.4(23.7)) in the blue box, 101 snaps (ID: 2515287~2515407) in the red box, and 69 snaps in the yellow box (ID:2515263~2515345). Is this zone, there is the Easter Island (27°08’S,109°22’W), which is in size of 163.6 km² (about 10% of SMOS’s foot-print), as shown by the red point in Figure 2.

During that orbit, large wind speed variability occurs (Figure 2):

 

Figure 2: Wind speed (orbit 251)
Figure 2: Wind speed (orbit 251)

2.1. For 201 snaps in both blue and red boxes
Comparisons of Tb_measurement and Tb_model1 of 201 snaps in the blue and red boxes are shown in this section. There are 100 snaps for X polarization and 101 snaps for Y polarization.
Figure 3 shows:
3.a Top: The biases between Tb_measurement and Tb_model1
3.a Bottom: The standard deviations of Tb_measurement-Tbmodel 1
3.b Top: The standard deviations of Tb_measurement
3.b Bottom: The standard deviations of Tb_model1

 

Figure 3.a: Comparisons between Tb_measurement and Tb_model1
Figure 3.a: Comparisons between Tb_measurement and Tb_model1
Figure 3.b: Comparisons between Tb_measurement and Tb_model1
Figure 3.b: Comparisons between Tb_measurement and Tb_model1

The standard deviations of Tb_measurement-Tbmodel 1 in the AFFOV appears lower than the standard deviation of Tb_measurement: for each point in the FOV, our model brings information consistent with the measurements!

X. Yin – J. Boutin, IPSL LOCEAN

download wordpress themes