CN102789004A - Satellite retrieval method for night rainfall probability - Google Patents
Satellite retrieval method for night rainfall probability Download PDFInfo
- Publication number
- CN102789004A CN102789004A CN2012102549673A CN201210254967A CN102789004A CN 102789004 A CN102789004 A CN 102789004A CN 2012102549673 A CN2012102549673 A CN 2012102549673A CN 201210254967 A CN201210254967 A CN 201210254967A CN 102789004 A CN102789004 A CN 102789004A
- Authority
- CN
- China
- Prior art keywords
- precipitation
- rainfall
- satellite
- probability
- night
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 239000011159 matrix material Substances 0.000 claims abstract description 24
- 238000001556 precipitation Methods 0.000 claims description 137
- 238000005259 measurement Methods 0.000 claims description 18
- 238000001228 spectrum Methods 0.000 claims description 17
- 239000000470 constituent Substances 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 abstract 1
- 230000003287 optical effect Effects 0.000 description 6
- 238000012216 screening Methods 0.000 description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 230000005855 radiation Effects 0.000 description 5
- 239000000428 dust Substances 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000002310 reflectometry Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- ZVNYJIZDIRKMBF-UHFFFAOYSA-N Vesnarinone Chemical compound C1=C(OC)C(OC)=CC=C1C(=O)N1CCN(C=2C=C3CCC(=O)NC3=CC=2)CC1 ZVNYJIZDIRKMBF-UHFFFAOYSA-N 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002040 relaxant effect Effects 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W2203/00—Real-time site-specific personalized weather information, e.g. nowcasting
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a satellite retrieval method for night rainfall probability, which comprises two steps of rain area division and rainfall estimation. A rain area division method comprises the steps of establishing an RPIM (rainfall possibility identification matrix) and then obtaining rainfall probability of all kinds of combination of brightness temperature Tb1 and brightness temperature Tb4-1 in the RPIM, wherein the RPIM is established based on an IR1-BTD41 two-dimensional spectral space. Therefore, an estimated rainfall value obtained through the rain area division method and a rainfall estimation method provided by the invention can be better related to and be more slightly deviated from an actual rainfall value, i.e. the estimated rainfall value is better consistent with the actually measured value, so the satellite retrieval method for night rainfall probability has the advantage of better application to multi-spectral satellite image night rainfall retrieval and nowcasting.
Description
Technical field
The present invention relates to a kind of satellite inversion method of rate of rainall, especially a kind of satellite inversion method to the rate of rainall at night belongs to the atmospheric science research field, is used for multispectral satellite image rainfall at night inverting and nowcasting.
Background technology
Stationary satellite precipitation at night inversion algorithm generally comprises following two steps, 1) the division rain belt, 2) the estimation rainfall; Mainly comprise following two steps and divide the rain belt: 1) screening cirrus and cirrus anvil from convection current stratus (being cumulonimbus), 2) (Kurino 1997 from warm cloud, to identify nimbostratus; Lensky and Rosenfeld 2003a; Luque et al. 2006).
As far back as 1978, Griffith et al. divided with regard to directly adopting cloud top bright temperature threshold value 253K to carry out the rain belt, but this comparatively dogmatic rain belt division methods is easy to thick cirrus erroneous judgement for the precipitation cloud, so scientists (Adler and Negri 1988; Vicente et al. 1998; Ba and Gruber 2001) on this basis, utilize to remedy measure spatial gradient of cloud-top temperature removal cirrus, to reach the effect of rain belt division preferably; And Inoue (1985,1987) then utilizes BTD
21(BTD
21Be a tunnel, it is worth T
B2-1Refer to passage IR
2With passage IR
1The infrared brightness temperature difference, IR
2Be the infrared channel of wavelength between 11.5 μ m-12.5 μ m, passage IR
1Refer to the infrared channel of wavelength between 10.3 μ m-11.3 μ m) discern cirrus, but the method only limits to tropical ocean (Inoue 1987).(2003a 2003b) has analyzed cloud particle effective radius R to Lensky and Rosenfeld
eWith the relation of radiation value, proposed to utilize BTD
41(BTD
41Be a tunnel, it is worth T
B4-1Infrared channel MIR and passage IR in the finger
1The infrared brightness temperature difference, middle infrared channel MIR is the infrared channel of wavelength between 3.5 μ m-4.0 μ m) value divide the method for rain belt, and think a moderate T
B4-1Possibly hint and have big R
eThe existence of precipitation cloud.Can know according to above analysis: in the research course that whole night, the rain belt was divided, scientists mainly still concentrates on and adopts the threshold value combination to declare knowledge, and therefore, still there are a certain distance in the effect and the actual monitoring situation of rain belt screening.
For rainfall estimation at night, method commonly used is mainly based on being worth (Ba and Gruber 2001 with bright temperature area in cloud top (Arkin and Meisner 1987) or bright temperature; Kuligowski 2002; Vicente et al. 1998) be variable curve-fitting method.
In addition, with respect to the tremendous development of precipitation inverting in the daytime, night, also there was bigger problem in the precipitation inverting.Because the infrared radiation that stationary satellite is measured is mainly from the cloud top, but quantity of precipitation more be with cloud in raindrop size distribution relevant.The relation of precipitation and cloud top infrared radiation is not easy to set up.By day, visible light (VIS) reflectivity can reflect the optical thickness information of cloud, in infrared (MIR; 3.5-4.0 μ m) reflectivity information can reflect water dust particle radii information (Nakajima and King, 1990).Can obtain good effect when utilizing their inverting precipitation on daytime.Night, reflectivity information can not obtain, and the information that can only rely on infrared channel itself is come inverting precipitation, and this has increased the difficulty of inverting at night undoubtedly.
Summary of the invention
The present invention is directed to the deficiency of prior art, the satellite inversion method of rate of rainall at a kind of night is provided, its primary technical purpose is to change existing rain belt division methods, this rain belt division methods be through set up precipitation probability declare know matrix RPIM after, obtain various bright temperature T
B1, bright temperature T
B4-1Be combined in precipitation probability and declare the precipitation probability of happening of knowing among the matrix RPIM, and precipitation probability is declared and known matrix RPIM and be based on IR
1-BTD
41The two-dimension spectrum space is set up; Less important technical purpose of the present invention is through obtaining certain specific bright temperature T
B1, bright temperature T
B4-1Be combined in precipitation probability and declare the precipitation probability of happening of knowing among the matrix RPIM, obtain this specific bright temperature T
B1, bright temperature T
B4-1The rainfall of combination is estimated.Therefore; The rainfall estimated value that is obtained through rain belt of the present invention division methods, rainfall method of estimation; Can have correlativity and less deviation preferably with the precipitation value of reality; Promptly this rainfall estimated value and measured value have consistance preferably, cause the present invention in rainfall at night inverting of multispectral satellite image and nowcasting, have the better application advantage.
For realizing above technical purpose, the present invention will take following technical scheme:
A kind of night rate of rainall the satellite inversion method, comprise that the rain belt divides and two steps of rainfall estimation, described rain belt partiting step is used for discerning the precipitation cloud of warm cloud, specifically may further comprise the steps: ⅰ, ⅰ, set up IR
1-BTD
41Two-dimension spectrum space---this IR
1-BTD
41The two-dimension spectrum space is based on passage IR
1Pairing bright temperature Tb1, channel B TD
41Pairing bright temperature T
B4-1Foundation forms; Wherein: passage IR
1Refer to the infrared channel of wavelength between 10.3 μ m-11.3 μ m, its value is T
B1, channel B TD
41Be a tunnel, it is worth T
B4-1Be middle infrared channel MIR and passage IR
1The infrared brightness temperature difference, middle infrared channel MIR is the infrared channel of wavelength between 3.5 μ m-4.0 μ m; ⅱ, based on IR
1-BTD
41The two-dimension spectrum space is set up precipitation probability and is declared knowledge matrix RPIM---at first with IR
1-BTD
41The two-dimension spectrum space is divided into the elementary cell of several 64*64, and the elementary cell of each 64*64 corresponds to a unit character space; Then, according to historical ground actual measurement precipitation value and the corresponding instantaneous bright temperature T of satellite of corresponding longitude and latitude
B1Observed reading, bright temperature T
B4-1Observed reading; Confirm the precipitation sample number and the non-precipitation sample number of constituent parts feature space; Precipitation sample number and non-precipitation sample number on the constituent parts feature space that passes through then to be obtained are confirmed the precipitation probability of happening of constituent parts feature space, can obtain precipitation probability and declare and know matrix RPIM; ⅲ, detect the Evaluation on effect index, confirm that suitable precipitation probability is a threshold value, carry out precipitation probability and declare the rain belt of knowing matrix RPIM and divide through qualitative assessment precipitation; The precipitation probability that ⅳ, basis have been set up is declared and is known matrix RPIM, obtains various bright temperature T
B1, bright temperature T
B4-1Precipitation probability of happening under the combination.
Setting up precipitation probability among the said step I i declares when knowing matrix RPIM, with bright temperature T
B1Be row, T
B4-1Be row.
Qualitative assessment precipitation detection Evaluation on effect index is precipitation detection probability POD or false alarm rate FAR or Heidke technical merit HSS among the said step ⅲ; Wherein:
In the formula;
is that satellite is estimated and observation does not all have a rainfall percentage,
be that satellite estimates at rainfall and observe percentage,
of no rainfall are that satellite is estimated no rainfall and observed the percentage that rainfall is arranged, and
is the satellite estimation and observes the percentage that rainfall is all arranged.
Described rainfall estimating step is used to estimate through the rain belt that partiting step is declared to know and be the quantity of precipitation in territory, precipitation cloud sector, and this rainfall estimates that RR is definite by following equation:
Wherein,
Be the pairing mean rainfall of bright temperature Tb1;
Be that precipitation probability is declared among the knowledge matrix RPIM by T
B1And T
B4-1The precipitation probability of happening of searching;
For the correction factor of rainfall estimation in real time, by T
B2-1Decision; T
B2-1Be passage IR
2With passage IR
1The infrared brightness temperature difference, passage IR2 refers to the infrared channel of wavelength between 11.5 μ m-12.5 μ m;
In the formula, K is temperature unit Kelvin.
Said mean rainfall
By based on bright temperature T
B1The fitting function match and get, the expression formula of this fitting function is following:
Unit be mm/10min; Bright temperature T
B1Unit be K.
Said historical ground actual measurement precipitation value is 10 minutes ground rain gage actual measurement rates of rainall at interval.
According to above technical scheme, can realize following beneficial effect:
1, the present invention adopts the passage IR of satellite spectrum
1With channel B TD
41Set up the two-dimension spectrum space, reason is channel B TD
41Bright temperature value not only with IR
1Bright temperature value have lower related coefficient, and the calculating through the radiation model is for spissatus (opticalthickness; 15), channel B TD
41Bright temperature value because of different water dust radius Rs
eAnd difference, i.e. BTD
41Bright temperature value can be used in distinguish spissatus.Compared with prior art, the division methods precipitation cloud that can carry out better in the warm cloud in this rain belt is declared knowledge.
2, rainfall method of estimation according to the invention adopts T earlier
B1Estimate mean rainfall, and then adopt
And
Adjust, wherein,
Be T
B1, T
B4-1Be combined in precipitation probability and declare the precipitation probability of happening of knowing among the matrix RPIM,
Then be that real-time rainfall is estimated correction factor, by T
B2-1Decision.Therefore, compared with prior art, the present invention can obtain rainfall in real time and estimate.
Therefore; The rainfall estimated value that is obtained through rain belt of the present invention division methods, rainfall method of estimation; Can have correlativity and less deviation preferably with the precipitation value of reality; Promptly this rainfall estimated value and measured value have consistance preferably, cause the present invention in rainfall at night inverting of multispectral satellite image and nowcasting, have the better application advantage.
3, the historical ground actual measurement precipitation value of the present invention's employing is 10 minutes ground rain gage actual measurement rates of rainall at interval; Be the present invention when night rainfall sample collection; Instantaneous measured value of satellite and ground, 10 minutes intervals rain gage actual measurement rate of rainall are set up statistical relationship; With in the prior art usual adopt with the instantaneous measured value of satellite and 1 hour at interval ground rain gage actual measurement rate of rainall set up statistical relationship and compare, the present invention can reduce rainfall sample collection at night error better, reason is: the cumulonimbus group that generates heavy rain moves with the speed of development very fast; During coupling; Actual measurement 1 hour rate of rainall in ground does not always drop in the vigorous cumulonimbus group of convection current, usually is in the cirrus district at cumulonimbus group edge on the contrary, or even the clear sky district in its place ahead.Therefore, actual measurement rainfall accumulative total period and moonscope are instantaneous approaching more, and the rainfall sample error will be more little.The present invention adopts 10min rain gage rate of rainall and the instantaneous measured value of satellite to mate because in the 10min, cloud cluster move and develop less, so 10min actual measurement rate of rainall generally all can have coupling preferably with cloud cluster position and development intensity.
Description of drawings
Fig. 1 is applicable to the change curve of the BTD that is calculated by SBDART of warm cloud about optical thickness;
Fig. 2 be night precipitation probability declare the synoptic diagram of knowing matrix RPIM;
Fig. 3 is that mean rainfall RR is about T
B1The fitting function curve map.
Embodiment
Accompanying drawing discloses the synoptic diagram of preferred embodiment involved in the present invention without limitation; Below will combine accompanying drawing that technical scheme of the present invention at length is described.
Spectroscopic data used herein derives from Japanese GMS multifunctional transport satellite of new generation (MTSAT)-2, and the definition of the passage that relates to or the bright temperature difference (BTD) is as shown in table 1.Satellite image such as all adopts at the longitude and latitude projection pattern, and the resolution of longitude and latitude all is 0.05 °.Have per half an hour of the satellite image once of the 7-9 month in 2010 to be used as research, wherein the image in August is used for forming algorithm, and other bimestrial images are as evaluation of algorithm
.
This paper adopted Chinese Anhui (29 ° of 18'-34 ° of 52'N of latitude scope, 114 ° of 56'-119 ° of 37'E of longitude scope) more than 300 automatically the station rain gage in ground
minute rate of rainall data that satellite data was write down in the identical period.These precipitation data are used as the reference value of check satellite inverting.
| ? | Wavelength(μm) | Channel Label | Value Label | Value Description |
| 1 | 10.3~11.3 | IR1 | T b1 | BT |
| 2 | 11.5-12.5 | IR2 | T b2 | BT |
| 3 | 6.5-7.0 | WV | T b3 | BT |
| 4 | 3.5-4.0 | MIR | T b4 | BT |
| 5 | 3.5-4.0 | MIR d | ρ | The Reflectance on daytime |
| 6 | 0.55-0.90 | VIS | A | Albedo |
| 7 | - | BTD21 | T b2-1 | BTD of IR2 and IR1 |
| 8 | - | BTD31 | T b3-1 | BTD of WV and IR1 |
| 9 | - | BTD41 | T b4-1 | BTD of MIR and IR1 |
Table 1
Table 1 discloses the definition of passage, bright temperature BT or bright temperature difference BTD in the satellite spectrum picture, in the table: infrared channel 1 (IR1,10.3 μ m-11.3 μ m); Infrared channel 2 (IR2,11.5 μ m-12.5 μ m); Vapor channel (WV, 6.5 μ m-7.0 μ m); Middle infrared channel (MIR, 3.5 μ m-4.0 μ m); Visible channel (VIS, 0.55 μ m-0.9 μ m).
Set up IR
1
-BTD
41
The two-dimension spectrum space
The application at first selects IR
1Bright temperature value as IR
1-BTD
41The row in two-dimension spectrum space then, in each channel value or BTD value of satellite spectrum picture remainder, are picked out and IR
1Value has the channel value or the BTD value of low related coefficient.Through calculating the related coefficient between each channel value or the BTD value, can find out: IR
2, WV, MIR and BTD
31Value and IR
1Value all has very high related coefficient, is used as IR
1-BTD
41The provisional capital in two-dimension spectrum space is not suitable for; Optional have only BTD
21And BTD
41, then, discuss through the radiation model, further to BTD
21, BTD
41Select.
As everyone knows, the precipitation cloud often has very big optical thickness because the convection current development is vigorous, and the while cloud will form precipitation and must have the very raindrop of long radius, this explanation optical thickness (τ) and water dust radius (R
e) can be as the criterion of precipitation generation.Different water dust radius (R
eAs 5 or 20 μ m), the cloud radiomimesis result of different optical thickness at 0.55 μ m (τ from 0 to 30) is presented among Fig. 1.On scheming, obviously can find out BTD
21And BTD
41All receive τ and R
eInfluence.But for the very big cloud of optical thickness (τ>15), BTD
21Value approach 0, almost with R
eIrrelevant with τ.This shows BTD
21Distinguish spissatus (such as thick cumulus and stratus are screened) from convective cloud is disabled.Spissatus BTD
41Be worth it because of different R
eAnd it is different.Therefore, the final IR that selects of this research
1And BTD
41Combination set up RPIM.
Set up precipitation probability and declare knowledge matrix RPIM
At first with IR
1-BTD
41The two-dimension spectrum space is divided into the elementary cell of several 64*64, and the elementary cell of each 64*64 corresponds to a unit character space; Then, according to historical ground actual measurement precipitation value and the corresponding instantaneous bright temperature T of satellite of corresponding longitude and latitude
B1Observed reading, bright temperature T
B4-1Observed reading; Confirm the precipitation sample number and the non-precipitation sample number of constituent parts feature space; Precipitation sample number and non-precipitation sample number on the constituent parts feature space that passes through then to be obtained; Confirm the precipitation probability of happening of constituent parts feature space, can obtain precipitation probability as shown in Figure 2 and declare knowledge matrix RPIM.
Declare knowledge matrix RPIM according to precipitation probability shown in Figure 2, for the Yun Eryan that is warmer than 230K, the high value of precipitation probability of happening is in T
B4-1Approach 0 zone.This and Lensky and Rosenfeld (2003a, 2003b) the screening rain belt is selected for use is-T of 1 ~ 4K
B4-1Scope is more or less the same.(the T just in the upper left corner of RPIM
B4-1Less than-10K, T
B1Zone less than 230K) have the high value of a strange precipitation probability district, this is to be caused by the defective that satellite is calibrated.
Utilize this precipitation probability to declare knowledge matrix RPIM and can analyze various T
B1And T
B4-1The possibility that rainfall under the combination takes place, and then distinguish rain belt and rainless region.With different rainfall probabilities is that threshold value is divided the rain belt, and the rain belt size of being analyzed will be different.Excessive or too small in order to prevent the rain belt, confirm that a suitable precipitation probability threshold value is declared and know the rain belt.
Qualitative assessment precipitation detects the Evaluation on effect index: precipitation detection probability (POD), false alarm rate (FAR) and Heidke technical merit (HSS), and it defines as follows:
Wherein,
is that satellite is estimated and observation does not all have a rainfall percentage,
be that satellite estimates at rainfall and observe percentage,
of no rainfall are that satellite is estimated no rainfall and observed the percentage that rainfall is arranged, and
is the satellite estimation and observes the percentage that rainfall is all arranged.
Table 2 has shown 76146 precipitation data has been used the rainfall that different probability draws as threshold value or the analysis result of non-rainfall.
Table 2
Table 2 shows that along with the raising of threshold value, though false alarm rate significantly reduces, the precipitation detection probability is also reducing, and HSS presents a parabolic type.Be decided to be threshold value 50% proper, HSS is 0.357 in this case, and FAR is 0.552, and POD reaches 0.528.
When selecting 50% probability threshold value for use, all clouds that are warmer than 253K all are divided into non-precipitation cloud.The very low nimbostratus in some cloud tops is missed in this division.If because directly adopt T
B1<253 threshold value, POD is 0.692,30.8% precipitation sample does not detect in addition.The cloud top is cooler than the setting of all thinking the precipitation cloud of 230K, has in fact abandoned the screening to dense cirrus.If can there be effective ways to screen dense cirrus, must improve the precision of screening rain belt.
According to Ba and Gruber (2001) screening rain belt at night, HSS is 0.245.
Rainfall is estimated
The precipitation method of estimation of this paper is similar to the method that Ba and Gruber (2001) proposes, and also is earlier by T
B1Estimate mean rainfall, by several factor adjustment, concrete function does again
Wherein,
is mean rainfall, obtains (Fig. 3) by match.
Also to consider ambient humidity when estimating precipitation.Ba and Gruber (2001) utilizes the product PWRH of precipitable water and humidity to adjust estimation quantity of precipitation.PWRH just can obtain once in per 3 hours; In order to estimate precipitation in real time, the present invention adjusts the factor
with humidity and replaces this index.
By T
B2-1Decision, the concrete definition as follows:
T
B2-1Reflected the total precipitable water of land face (Eck and Holben 1994; Sobrino et al. 1999), it is rational being used for defining the humidity adjustment factor.
In addition, the present invention declares when knowing matrix RPIM setting up precipitation probability, and said historical ground actual measurement precipitation value is 10 minutes ground rain gage actual measurement rates of rainall at interval.Be that the present invention sets up statistical relationship with ground, 10 minutes intervals rain gage actual measurement rate of rainall and the instantaneous measured value of satellite, thereby have coupling preferably with cloud cluster position and development intensity.
When carrying out that half an hour, precipitation was estimated,, can obtain per 10 minutes continuous cloud atlas once, 3 width of cloth have just been obtained halfhour precipitation by the distribution of rainfall additions in 10 minutes that the cloud atlas inverting obtains through cloud cluster mobile trend track algorithm.Owing to considered moving of cloud, this mode can estimate dynamically that the precipitation cloud covers the time and intensity variation of survey station, has therefore significantly improved the precision that half an hour, precipitation was estimated.
In addition, the present invention will adopt root-mean-square-deviation
and related coefficient
that described rainfall is estimated to assess.
Wherein,
(8)
In the formula,
and
is respectively satellite and estimates rate of rainall and rain gage actual measurement rate of rainall.
In addition, consider and settle in an area precipitation (Scofield 1987 to factor affecting such as, landform for wind-engaging; Vicente et al. 2002), and moonscope and rain gage observed pattern is inconsistent, and the present invention adopts " relatively radius R c " notion.Use survey station to be the center, with " relatively radius " be radius the inverting precipitation area the satellite inverting rate of rainall of approaching actual rate of rainall represent the estimated value at this station.This paper is set at 1,3,5 pixel respectively with Rc.
Assess ten minutes used samples of precipitation and amount to 13954,
Be 0.7212 mm (10 min)
-1Assessment result (Fig. 4) shows that equation (4) is over-evaluated weak precipitation easily, underestimates precipitation.During Rc=1, have a large amount of weak precipitation [<2.5 mm (10 min)
-1] sample is over-evaluated becomes above precipitation [>5 mm (10 min)
-1], and all superpower precipitation [>12 mm (10 min)
-1] then all underestimated.Weak precipitation is over-evaluated mainly because cirrus is handled as the precipitation cloud is caused, and when increasing Rc, thisly over-evaluates phenomenon and has obtained alleviation.Significantly do not improved along with relaxing of evaluation condition but superpower precipitation underestimates, this is equation (4) restriction.Because even the bright temperature in cloud top is 190K, under the environment of humidity (
=2) precipitation of, estimating according to equation (4) also has only 15 mm (10 min)
-1
However, during Rc=3, cc has still reached 0.776.During Rc=5, cc further brings up to 0.85.
The precipitation of mid latitudes mainly is divided into 3 types: Stratiform Cloud Precipitation, strong convection precipitation and intervenient mixing precipitation.Amount in July, 2010 and September in 13 precipitation, mixing a precipitation example is 7, and a Stratiform Cloud Precipitation example is 1, and a strong convection precipitation example is 5.The result who utilizes ITCAT (half an hour, precipitation was estimated) estimation precipitation at three kinds of nights is discussed respectively below.
Assessment to mixing the precipitation estimated result shows that ITCAT can be competent at the estimation to this type precipitation fully.Even point-to-point statistics (Rc=1) estimates that the related coefficient of precipitation and actual precipitation also can surpass 0.64 (the highest reaches 0.71), during Rc=3, the scope of rmsd is at 1-1.87 mm (0.5h)
-1, cc is at least 0.83.If further enlarge Rc, then related coefficient almost can both reach 0.9.
Mixing precipitation is the most common type of precipitation, has accounted for 53.8% of all precipitation example.ITCAT has realized mixing the estimation of precipitation, shows that ITCAT is an acceptable to the estimated result of the most of precipitation of mid latitudes.
ITCAT can Stratiform Cloud Precipitation have good estimation, when Rc=3, estimates that the related coefficient of precipitation and actual precipitation has reached 0.88, and the deviation of mean value has only 0.03 mm (0.5h)
-1, rmsd is 0.32 mm (0.5h)
-1During Rc=5, related coefficient even can reach 0.97, rmsd also has only 0.16 mm (0.5h)
-1Though in July, 2010 and September, the individual example of pure stratiform clouds has only one, this result shows, takes all factors into consideration precipitation intensity, coverage and duration, can obtain accurate Stratiform Cloud Precipitation amount and estimate.
Claims (6)
- One kind night rate of rainall the satellite inversion method, comprise that the rain belt divides and two steps of rainfall estimation, it is characterized in that described rain belt partiting step is used for discerning the precipitation cloud of warm cloud, specifically may further comprise the steps: ⅰ, set up IR 1-BTD 41Two-dimension spectrum space---this IR 1-BTD 41The two-dimension spectrum space is based on passage IR 1Pairing bright temperature Tb 1, channel B TD 41Pairing bright temperature T B4-1Foundation forms; Wherein: passage IR 1Refer to the infrared channel of wavelength between 10.3 μ m-11.3 μ m, its value is T B1, channel B TD 41Be a tunnel, it is worth T B4-1Be middle infrared channel MIR and passage IR 1The infrared brightness temperature difference, middle infrared channel MIR is the infrared channel of wavelength between 3.5 μ m-4.0 μ m; ⅱ, based on IR 1-BTD 41The two-dimension spectrum space is set up precipitation probability and is declared knowledge matrix RPIM---at first with IR 1-BTD 41The two-dimension spectrum space is divided into the elementary cell of several 64*64, and the elementary cell of each 64*64 corresponds to a unit character space; Then, according to historical ground actual measurement precipitation value and the corresponding instantaneous bright temperature T of satellite of corresponding longitude and latitude B1Observed reading, bright temperature T B4-1Observed reading; Confirm the precipitation sample number and the non-precipitation sample number of constituent parts feature space; Precipitation sample number and non-precipitation sample number on the constituent parts feature space that passes through then to be obtained are confirmed the precipitation probability of happening of constituent parts feature space, can obtain precipitation probability and declare and know matrix RPIM; ⅲ, detect the Evaluation on effect index, confirm that suitable precipitation probability is a threshold value, carry out precipitation probability and declare the rain belt of knowing matrix RPIM and divide through qualitative assessment precipitation; The precipitation probability that ⅳ, basis have been set up is declared and is known matrix RPIM, obtains various bright temperature T B1, bright temperature T B4-1Precipitation probability of happening under the combination.
- 2. according to the satellite inversion method of claim 1 rate of rainall at said night, it is characterized in that, set up precipitation probability among the said step I i and declare when knowing matrix RPIM, with bright temperature T B1Be row, T B4-1Be row.
- 3. according to the satellite inversion method of claim 1 rate of rainall at said night, it is characterized in that qualitative assessment precipitation zone detection Evaluation on effect index is precipitation detection probability POD or false alarm rate FAR or Heidke technical merit HSS among the said step ⅲ; Wherein:
In the formula; is that satellite is estimated and observation does not all have a rainfall percentage,be that satellite estimates at rainfall and observe percentage, of no rainfall are that satellite is estimated no rainfall and observed the percentage that rainfall is arranged, and
is the satellite estimation and observes the percentage that rainfall is all arranged.
- 4. according to the satellite inversion method of claim 1 rate of rainall at said night, it is characterized in that described rainfall estimating step is used to estimate through the rain belt that partiting step is declared to know and be the quantity of precipitation in territory, precipitation cloud sector, this rainfall estimates that RR is definite by following equation:
Wherein,Be bright temperature T B1Pairing mean rainfall;
Be that precipitation probability is declared among the knowledge matrix RPIM by T B1And T B4-1The precipitation probability of happening of searching;
For the correction factor of rainfall estimation in real time, by T B2-1Decision; T B2-1Be passage IR 2With passage IR 1The infrared brightness temperature difference, passage IR 2Refer to the infrared channel of wavelength between 11.5 μ m-12.5 μ m;
In the formula, K is temperature unit Kelvin. - 5. according to the satellite inversion method of claim 1 rate of rainall at said night, it is characterized in that said mean rainfall By based on bright temperature T B1The fitting function match and get, the expression formula of this fitting function is following:Unit be mm/10min; Bright temperature T B1Unit be K.
- 6. according to the satellite inversion method of claim 1 rate of rainall at said night, it is characterized in that said historical ground actual measurement precipitation value is 10 minutes ground rain gage actual measurement rates of rainall at interval.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210254967.3A CN102789004B (en) | 2012-07-20 | 2012-07-20 | Satellite retrieval method for night rainfall probability |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210254967.3A CN102789004B (en) | 2012-07-20 | 2012-07-20 | Satellite retrieval method for night rainfall probability |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102789004A true CN102789004A (en) | 2012-11-21 |
| CN102789004B CN102789004B (en) | 2015-02-04 |
Family
ID=47154453
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210254967.3A Active CN102789004B (en) | 2012-07-20 | 2012-07-20 | Satellite retrieval method for night rainfall probability |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102789004B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160192A (en) * | 2015-09-17 | 2015-12-16 | 浙江大学 | TRMM (Tropical Rainfall Measuring Mission) satellite rainfall data downscaling method based on M5-Local |
| CN108874734A (en) * | 2018-04-25 | 2018-11-23 | 中国科学院国家空间科学中心 | A kind of Global Land Surface Precipitation inversion method |
| CN108897074A (en) * | 2018-04-25 | 2018-11-27 | 中国科学院国家空间科学中心 | A kind of global ocean precipitation inversion method |
| CN109297605A (en) * | 2018-10-09 | 2019-02-01 | 北京大学 | A kind of Surface Temperature Retrieval method with Thermal Infrared Data infrared based in |
| CN109992746A (en) * | 2019-03-11 | 2019-07-09 | 中国气象科学研究院 | Atmosphere water substance total amount, total precipitable water and its corresponding precipitation efficiency calculation method |
| CN111060992A (en) * | 2019-12-27 | 2020-04-24 | 国家卫星气象中心(国家空间天气监测预警中心) | Equal-weight conjugate precipitation detection method and system for satellite-borne microwave dioxygen detection channel |
| CN111123408A (en) * | 2019-11-26 | 2020-05-08 | 深圳震有科技股份有限公司 | Method, system and storage medium for predicting precipitation distribution based on GIS |
| CN112462369A (en) * | 2020-10-30 | 2021-03-09 | 国家卫星气象中心(国家空间天气监测预警中心) | Method and device for microwave imager to invert sea atmosphere rainfall |
| CN113552656A (en) * | 2021-07-26 | 2021-10-26 | 福建农林大学 | Rainfall intensity monitoring method and system based on outdoor image multi-space-time fusion |
| CN113591387A (en) * | 2021-08-05 | 2021-11-02 | 安徽省气象台 | Huber norm constraint-based satellite data inversion precipitation method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655564A (en) * | 2009-09-15 | 2010-02-24 | 中国农业科学院农业资源与农业区划研究所 | Method for inversing surface temperature and emissivity from MODIS data |
-
2012
- 2012-07-20 CN CN201210254967.3A patent/CN102789004B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655564A (en) * | 2009-09-15 | 2010-02-24 | 中国农业科学院农业资源与农业区划研究所 | Method for inversing surface temperature and emissivity from MODIS data |
Non-Patent Citations (3)
| Title |
|---|
| XIAO-YONG ZHUGE ET AL: "Rainfall retrieval and Nowcasting Based on Multispectral Satellite Images. Part I: Retrieval Study on Daytime 10-Minute Rain Rate", 《2011 AMERICAN METEOROLOGICAL SOCIETY》 * |
| 杨引明等: "应用GMS5多通道云图估计降水技术的研究", 《气象》 * |
| 罗利民: "夜间观云的辅助方法", 《陕西气象》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160192B (en) * | 2015-09-17 | 2017-12-26 | 浙江大学 | TRMM satellite rainfall data NO emissions reduction methods based on M5 LocalR |
| CN105160192A (en) * | 2015-09-17 | 2015-12-16 | 浙江大学 | TRMM (Tropical Rainfall Measuring Mission) satellite rainfall data downscaling method based on M5-Local |
| CN108874734A (en) * | 2018-04-25 | 2018-11-23 | 中国科学院国家空间科学中心 | A kind of Global Land Surface Precipitation inversion method |
| CN108897074A (en) * | 2018-04-25 | 2018-11-27 | 中国科学院国家空间科学中心 | A kind of global ocean precipitation inversion method |
| CN109297605A (en) * | 2018-10-09 | 2019-02-01 | 北京大学 | A kind of Surface Temperature Retrieval method with Thermal Infrared Data infrared based in |
| CN109992746A (en) * | 2019-03-11 | 2019-07-09 | 中国气象科学研究院 | Atmosphere water substance total amount, total precipitable water and its corresponding precipitation efficiency calculation method |
| CN111123408A (en) * | 2019-11-26 | 2020-05-08 | 深圳震有科技股份有限公司 | Method, system and storage medium for predicting precipitation distribution based on GIS |
| CN111060992A (en) * | 2019-12-27 | 2020-04-24 | 国家卫星气象中心(国家空间天气监测预警中心) | Equal-weight conjugate precipitation detection method and system for satellite-borne microwave dioxygen detection channel |
| CN111060992B (en) * | 2019-12-27 | 2021-09-10 | 国家卫星气象中心(国家空间天气监测预警中心) | Equal-weight conjugate precipitation detection method and system for satellite-borne microwave dioxygen detection channel |
| CN112462369A (en) * | 2020-10-30 | 2021-03-09 | 国家卫星气象中心(国家空间天气监测预警中心) | Method and device for microwave imager to invert sea atmosphere rainfall |
| CN113552656A (en) * | 2021-07-26 | 2021-10-26 | 福建农林大学 | Rainfall intensity monitoring method and system based on outdoor image multi-space-time fusion |
| CN113552656B (en) * | 2021-07-26 | 2022-05-24 | 福建农林大学 | Rainfall intensity monitoring method and system based on outdoor image multi-space-time fusion |
| CN113591387A (en) * | 2021-08-05 | 2021-11-02 | 安徽省气象台 | Huber norm constraint-based satellite data inversion precipitation method and system |
| CN113591387B (en) * | 2021-08-05 | 2023-09-01 | 安徽省气象台 | Satellite data inversion precipitation method and system based on Huber norm constraint |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102789004B (en) | 2015-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102789004A (en) | Satellite retrieval method for night rainfall probability | |
| Salomon et al. | Validation of the MODIS bidirectional reflectance distribution function and albedo retrievals using combined observations from the aqua and terra platforms | |
| Wang et al. | Operational data fusion framework for building frequent landsat-like imagery | |
| Huang et al. | Detection of Asia dust storms using multisensor satellite measurements | |
| Mao et al. | Angle effects of vegetation indices and the influence on prediction of SPAD values in soybean and maize | |
| CN106407656A (en) | Retrieval method for aerosol optical thickness based on high resolution satellite image data | |
| Tan et al. | A comparison of radiometric correction techniques in the evaluation of the relationship between LST and NDVI in Landsat imagery | |
| Zhang et al. | Estimation of forest leaf area index using height and canopy cover information extracted from unmanned aerial vehicle stereo imagery | |
| Duncan et al. | Comparison of Late Pleistocene and modern glacier extents in central Nepal based on digital elevation data and satellite imagery | |
| Sankey et al. | Landsat-5 TM and lidar fusion for sub-pixel juniper tree cover estimates in a western rangeland | |
| Hu et al. | Estimation of surface upward longwave radiation using a direct physical algorithm | |
| Wang et al. | A multilayer cloud detection algorithm for the Suomi-NPP Visible Infrared Imager Radiometer Suite (VIIRS) | |
| Rittger | Spatial estimates of snow water equivalent in the Sierra Nevada | |
| Hu et al. | Influence of emissivity angular variation on land surface temperature retrieved using the generalized split-window algorithm | |
| Thomas et al. | Spatial modelling of the fraction of photosynthetically active radiation absorbed by a boreal mixedwood forest using a lidar–hyperspectral approach | |
| Lee et al. | New approach for snow cover detection through spectral pattern recognition with MODIS data | |
| Xia et al. | Modeling daily temperatures via a phenology-based annual temperature cycle model | |
| Bian et al. | An angular normalization method for temperature vegetation dryness index (TVDI) in monitoring agricultural drought | |
| Deng et al. | An approach for reflectance anisotropy retrieval from UAV-based oblique photogrammetry hyperspectral imagery | |
| Jang et al. | Thermal‐water stress index from satellite images | |
| Renc et al. | Spatial structure of the surface heat and cold islands in summer based on Landsat 8 imagery in southern Poland | |
| Smekens et al. | Observation of SO2 degassing at Stromboli volcano using a hyperspectral thermal infrared imager | |
| Lehnert et al. | A new high spatial resolution low stratus/fog retrieval for the Atacama Desert | |
| Meng et al. | Hot dark spot index method based on multi-angular remote sensing for leaf area index retrieval | |
| Hesketh et al. | A review of remote sensing of tropical dry forests |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |