CDXWRFcore - Historial de revisiones

Lluis.fita en 21:40 27 feb 2019

2019-02-27T21:40:00Z

← Revisión anterior		Revisión del 21:40 27 feb 2019
Línea 1:		Línea 1:
	= Core variables =		= Core variables =

	~~'''UNDER CONSTRUCTION'''~~

	These are the basic variables required by CORDEX. Most of them are standard fields and therefore tend to require simple calculations from the currently available variables from the WRF model. These variables are obtained by setting the pre-compilation flag CORDEXDIAG, and will appear in two different files: 3D variables at pressure levels (the WRF model internally interpolate them since it uses the η coordinate in the vertical) will appear in the output file with the 23rd stream (mainly called wrfpress), and the 2D variables in the module's output file wrfcdx.		These are the basic variables required by CORDEX. Most of them are standard fields and therefore tend to require simple calculations from the currently available variables from the WRF model. These variables are obtained by setting the pre-compilation flag CORDEXDIAG, and will appear in two different files: 3D variables at pressure levels (the WRF model internally interpolate them since it uses the η coordinate in the vertical) will appear in the output file with the 23rd stream (mainly called wrfpress), and the 2D variables in the module's output file wrfcdx.

Lluis.fita: /* sund: duration of sunshine */

2019-02-27T19:55:53Z

sund: duration of sunshine

← Revisión anterior		Revisión del 19:55 27 feb 2019
Línea 104:		Línea 104:
	This variable accounts for the the sum of the time for which the direct solar irradiance (downwelling short-wave radiation, rsds) exceeds 120 Wm<SUP>-2</SUP> (WMO, 2010a) implemented following equation 12 and provided in seconds. In order to provide an example of the correct implementation of this diagnostics preliminary results are shown in figure 2. The figure shows the `sund' values and compare them with the incoming solar radiation. It is shown how the `sund' values vary accordingly to the moment of the day with zero values during night (left panel) or persistent totally cloud covered regions (map at the right panel)		This variable accounts for the the sum of the time for which the direct solar irradiance (downwelling short-wave radiation, rsds) exceeds 120 Wm<SUP>-2</SUP> (WMO, 2010a) implemented following equation 12 and provided in seconds. In order to provide an example of the correct implementation of this diagnostics preliminary results are shown in figure 2. The figure shows the `sund' values and compare them with the incoming solar radiation. It is shown how the `sund' values vary accordingly to the moment of the day with zero values during night (left panel) or persistent totally cloud covered regions (map at the right panel)

	[[Archivo:~~test_sund~~.png]]		[[Archivo:sund.png]]

	where <I>SWDOWN</I>: downward shortwave radiation (W m<SUP>-2</SUP>), <I>δt</I>: time-step length (s).		where <I>SWDOWN</I>: downward shortwave radiation (W m<SUP>-2</SUP>), <I>δt</I>: time-step length (s).

	[[Archivo:~~sund~~.png\|frame\|50px\|Figure 2: Temporal evolution at S 62<SUP>◦</SUP> 4' 38.00", 4<SUP>◦</SUP> 58' 55.51" W (left panel) of shortwave downward radiation (rsds, red line, left y-axis) and sunshine duration for a 3-hourly 9freq (sund, stars, right y-axis). Sunshine length map on 2012-12-09 at 15 UTC. X denotes position of the temporal evolution]]		[[Archivo:test_sund.png\|frame\|50px\|Figure 2: Temporal evolution at S 62<SUP>◦</SUP> 4' 38.00", 4<SUP>◦</SUP> 58' 55.51" W (left panel) of shortwave downward radiation (rsds, red line, left y-axis) and sunshine duration for a 3-hourly 9freq (sund, stars, right y-axis). Sunshine length map on 2012-12-09 at 15 UTC. X denotes position of the temporal evolution]]

	=== tauuv: surface downward wind stress ===		=== tauuv: surface downward wind stress ===

Lluis.fita: /* Vertically integrated variables */

2019-02-27T19:39:48Z

Vertically integrated variables

← Revisión anterior		Revisión del 19:39 27 feb 2019
Línea 183:		Línea 183:
	The instantaneous vertically integrated amount of water vapor (<I>prw</I>), liquid condensed water species (<I>clwvi</I>), ice species (<I>clivi</I>), graupel (<I>clgvi</I>), hail (<I>clhvi</I>) are the vertically integrated amounts of each species along the vertical column (density weighted) over each grid point. They are provided using the same implementation as those in <CODE>p_interp.F</CODE> - WRF tool for vertical interpolation. The general equation following WRF standard variables is:		The instantaneous vertically integrated amount of water vapor (<I>prw</I>), liquid condensed water species (<I>clwvi</I>), ice species (<I>clivi</I>), graupel (<I>clgvi</I>), hail (<I>clhvi</I>) are the vertically integrated amounts of each species along the vertical column (density weighted) over each grid point. They are provided using the same implementation as those in <CODE>p_interp.F</CODE> - WRF tool for vertical interpolation. The general equation following WRF standard variables is:

	[[Archivo:~~clivar~~.png]]		[[Archivo:vertint.png]]

	where <I>clvivar</I>: the column vertically integrated variable's CF-compliant name (<I>prw, clwvi, clivi, clgvi,</I> or <I>clhvi</I>), <I>MU</I>: perturbation dry air mass in column (Pa), <I>MUB</I>: base-state dry air mass in column (Pa), <I>g</I>: gravity (ms<SUP>-2</SUP>), <I>e_vert</I>: total number of vertical levels, WRFVAR: the water species' mixing ratio at each sigma level (kgkg<SUP>-1</SUP>), <I>DNW</I>: difference between two consecutive full-eta levels (-). Table 2 indicates the WRFVAR names associated with the clvivar names.		where <I>clvivar</I>: the column vertically integrated variable's CF-compliant name (<I>prw, clwvi, clivi, clgvi,</I> or <I>clhvi</I>), <I>MU</I>: perturbation dry air mass in column (Pa), <I>MUB</I>: base-state dry air mass in column (Pa), <I>g</I>: gravity (ms<SUP>-2</SUP>), <I>e_vert</I>: total number of vertical levels, WRFVAR: the water species' mixing ratio at each sigma level (kgkg<SUP>-1</SUP>), <I>DNW</I>: difference between two consecutive full-eta levels (-). Table 2 indicates the WRFVAR names associated with the clvivar names.

Lluis.fita: /* wsgsmax: maximum near-surface wind speed of gust */

2019-02-27T19:35:27Z

wsgsmax: maximum near-surface wind speed of gust

← Revisión anterior		Revisión del 19:35 27 feb 2019
Línea 148:		Línea 148:
	The wind gust accounts for the wind from upper levels that is projected to the surface due to instability within the planetary boundary layer. In the current version of the module two complementary methods of diagnosing the variable have been implemented (resultant winds are Earth-rotated). The choice between the two methods is done by the parameter labeled <CODE>wsgs_diag</CODE> (in cordex section), with the default value set to 1. The implemented methods are:		The wind gust accounts for the wind from upper levels that is projected to the surface due to instability within the planetary boundary layer. In the current version of the module two complementary methods of diagnosing the variable have been implemented (resultant winds are Earth-rotated). The choice between the two methods is done by the parameter labeled <CODE>wsgs_diag</CODE> (in cordex section), with the default value set to 1. The implemented methods are:
	* The Brasseur method [<CODE>wsgs_diag = 1</CODE>]: An implementation of wind gust considering Turbulent Kinetic Energy (TKE) estimates and stability defined by virtual temperature (θ v ) as indicated in equation 15 following Brasseur (2001). Implementation is adapted from a version already introduced in the CLimate WRF ([[http://www.meteo.unican.es/wiki/cordexwrf/SoftwareTools/ClWrf clWRF]]; Fita et al., 2010),		* The Brasseur method [<CODE>wsgs_diag = 1</CODE>]: An implementation of wind gust considering Turbulent Kinetic Energy (TKE) estimates and stability defined by virtual temperature (θ v ) as indicated in equation 15 following Brasseur (2001). Implementation is adapted from a version already introduced in the CLimate WRF ([[http://www.meteo.unican.es/wiki/cordexwrf/SoftwareTools/ClWrf clWRF]]; Fita et al., 2010),

	[[Archivo:BrasseurUNS.png]]		[[Archivo:BrasseurUNS.png]]

	where <I>TKE</I>: Turbulent Kinetic Energy (m2s<SUP>-2</SUP>), <I>θ<SUB>v</SUB></I>: virtual potential temperature (K). <I>z<SUB>p</SUB></I> height of the considered parcel (m, maximum height which satisfies equation 15), <I>∆θ<SUB>v</SUB></I> (z): variation of <I>θ>SUB>v</SUB></I> over a given layer (Km<SUP>-1</SUP>).		where <I>TKE</I>: Turbulent Kinetic Energy (m2s<SUP>-2</SUP>), <I>θ<SUB>v</SUB></I>: virtual potential temperature (K). <I>z<SUB>p</SUB></I> height of the considered parcel (m, maximum height which satisfies equation 15), <I>∆θ<SUB>v</SUB></I> (z): variation of <I>θ>SUB>v</SUB></I> over a given layer (Km<SUP>-1</SUP>).
	* The AFWA method [<CODE>wsgs_diag = 2</CODE>]: An implementation adopted from the WRF module <CODE>module_diag_afwa.F</CODE> which calculates the wind gust that only occurs as a blending of upper-level winds <I>zagl</I> (around 1 km above ground; zagl(k<SUB>1000</SUB>) ≥ 1000 m, see equation 16) when precipitation intensity per hour is above a given maximum value, prate mm		* The AFWA method [<CODE>wsgs_diag = 2</CODE>]: An implementation adopted from the WRF module <CODE>module_diag_afwa.F</CODE> which calculates the wind gust that only occurs as a blending of upper-level winds <I>zagl</I> (around 1 km above ground; zagl(k<SUB>1000</SUB>) ≥ 1000 m, see equation 16) when precipitation intensity per hour is above a given maximum value, prate mm

	[[Archivo:wssblend_afwa.png]]		[[Archivo:wssblend_afwa.png]]

	where <I>va</I>: air wind (ms<SUP>-1</SUP>), <I>zagl</I>: height above ground (m), <I>k_<SUB>1000</SUB></I>: vertical level at which <I>zagl</I> is equal or above 1000 m. prate mm), va blend : blended wind (ms<SUP>-1</SUP>) <I>hr</I>: hourly precipitation rate (mmh<SUP>-1</SUP>)		where <I>va</I>: air wind (ms<SUP>-1</SUP>), <I>zagl</I>: height above ground (m), <I>k_<SUB>1000</SUB></I>: vertical level at which <I>zagl</I> is equal or above 1000 m. prate mm), va blend : blended wind (ms<SUP>-1</SUP>) <I>hr</I>: hourly precipitation rate (mmh<SUP>-1</SUP>)

Lluis.fita: /* evspsblpot: potential evapotranspiration */

2019-02-27T19:27:14Z

evspsblpot: potential evapotranspiration

← Revisión anterior		Revisión del 19:27 27 feb 2019
Línea 228:		Línea 228:

	See in figure 8 an example of the differences between both implementations. It shows how important is the correction introduced by Milly and its strong effect on the diurnal cycle. Basically, the correction permits potential evapotranspiration during night time and reduces its strength at noon (18 UTC corresponds to 12 local time). There is a generic diagnostic to overcome boundary layer scheme dependency in the calculation of the drag coefficient (see below in generic variables).		See in figure 8 an example of the differences between both implementations. It shows how important is the correction introduced by Milly and its strong effect on the diurnal cycle. Basically, the correction permits potential evapotranspiration during night time and reduces its strength at noon (18 UTC corresponds to 12 local time). There is a generic diagnostic to overcome boundary layer scheme dependency in the calculation of the drag coefficient (see below in generic variables).

			[[Archivo:test_evspsblpot.png\|frame\|50px\|Figure 8: Evolution (a, in y-log scale) of potential evapotranspiration by bulk (yellow) and Milly92 (blue) generic (dashed lines) methods at S 4<SUP>◦</SUP> 58' 55.524", 62<SUP>◦</SUP> 4' 37.92" W (blue cross in b). On 2015-11-18 15 UTC, potential evapotranspiration following Milly correction (b), differences between both methods (c, evspblpot<SUB>Milly92</SUB>-evspblpot<SUB>bulk </SUB>), differences between both generic methods (d, evspblpot<SUP>gen</SUP><SUB>Milly92</SUB>-evspblpot<SUP>gen</SUP><SUB>bulk</SUB>), differences between Milly method and its generic counterpart (e, evspblpot<SUB>Milly92</SUB>-evspblpot<SUP>gen</SUP><SUB>Milly92</SUB>) and differences between bulk method and its generic counterpart (f, evspblpot<SUB>bulk</SUB>-evspblpot<SUP>gen</SUP><SUB>bulk</SUB>)]]

	== Generic variables ==		== Generic variables ==

Lluis.fita: /* Vertically integrated variables */

2019-02-27T19:22:12Z

Vertically integrated variables

← Revisión anterior		Revisión del 19:22 27 feb 2019
Línea 212:		Línea 212:
	Note that clgvi and clhvi are part of the `Tier1' level and are only accessible if pre-compilation variable CDXWRF is set to 1. See section 2.3 for more detail. In order to provide an example of the correct computation of the diagnostics, results at a given grid point are shown in figure 7. It is shown that the total precipitable water (prw) correctly corresponds to the density weighted vertical integration of the water content along the column of air.		Note that clgvi and clhvi are part of the `Tier1' level and are only accessible if pre-compilation variable CDXWRF is set to 1. See section 2.3 for more detail. In order to provide an example of the correct computation of the diagnostics, results at a given grid point are shown in figure 7. It is shown that the total precipitable water (prw) correctly corresponds to the density weighted vertical integration of the water content along the column of air.

	[[Archivo:test_prw.png\|frame\|50px\|Figure 7: On 2012-12-09 at 15 UTC at S 62 ◦ 4 0 38.~~00”~~, 4 ◦ 58 0 55.~~51”W~~ (left): water path (prw, vertical straight line in mm top x-axis), vertical profile of water vapour (qv, line with full circles in kgkg −1 bottom x-axis), water path at each level (line with crosses). Map of water path (right) on 2012-12-09 at 15 UTC, red cross shows where the vertical accumulation is retrieved		[[Archivo:test_prw.png\|frame\|50px\|Figure 7: On 2012-12-09 at 15 UTC at S 62<SUP>◦</SUP> 4' 38.00", 4<SUP>◦</SUP> 58' 55.51" W (left): water path (prw, vertical straight line in mm top x-axis), vertical profile of water vapour (qv, line with full circles in kgkg −1 bottom x-axis), water path at each level (line with crosses). Map of water path (right) on 2012-12-09 at 15 UTC, red cross shows where the vertical accumulation is retrieved]]

	=== evspsblpot: potential evapotranspiration ===		=== evspsblpot: potential evapotranspiration ===

Lluis.fita: /* Vertically integrated variables */

2019-02-27T19:21:24Z

Vertically integrated variables

← Revisión anterior		Revisión del 19:21 27 feb 2019
Línea 212:		Línea 212:
	Note that clgvi and clhvi are part of the `Tier1' level and are only accessible if pre-compilation variable CDXWRF is set to 1. See section 2.3 for more detail. In order to provide an example of the correct computation of the diagnostics, results at a given grid point are shown in figure 7. It is shown that the total precipitable water (prw) correctly corresponds to the density weighted vertical integration of the water content along the column of air.		Note that clgvi and clhvi are part of the `Tier1' level and are only accessible if pre-compilation variable CDXWRF is set to 1. See section 2.3 for more detail. In order to provide an example of the correct computation of the diagnostics, results at a given grid point are shown in figure 7. It is shown that the total precipitable water (prw) correctly corresponds to the density weighted vertical integration of the water content along the column of air.

	Figure 7: On 2012-12-09 at 15 UTC at S 62 ◦ 4 0 38.00”, 4 ◦ 58 0 55.51”W (left): water path (prw, vertical straight line in mm top x-axis), vertical profile of water vapour (qv, line with full circles in kgkg −1 bottom x-axis), water path at each level (line with crosses). Map of water path (right) on 2012-12-09 at 15 UTC, red cross shows where the vertical accumulation is retrieved		[[Archivo:test_prw.png\|frame\|50px\|Figure 7: On 2012-12-09 at 15 UTC at S 62 ◦ 4 0 38.00”, 4 ◦ 58 0 55.51”W (left): water path (prw, vertical straight line in mm top x-axis), vertical profile of water vapour (qv, line with full circles in kgkg −1 bottom x-axis), water path at each level (line with crosses). Map of water path (right) on 2012-12-09 at 15 UTC, red cross shows where the vertical accumulation is retrieved

	=== evspsblpot: potential evapotranspiration ===		=== evspsblpot: potential evapotranspiration ===

Lluis.fita: /* Vertically integrated variables */

2019-02-27T19:20:19Z

Vertically integrated variables

← Revisión anterior		Revisión del 19:20 27 feb 2019
Línea 211:		Línea 211:

	Note that clgvi and clhvi are part of the `Tier1' level and are only accessible if pre-compilation variable CDXWRF is set to 1. See section 2.3 for more detail. In order to provide an example of the correct computation of the diagnostics, results at a given grid point are shown in figure 7. It is shown that the total precipitable water (prw) correctly corresponds to the density weighted vertical integration of the water content along the column of air.		Note that clgvi and clhvi are part of the `Tier1' level and are only accessible if pre-compilation variable CDXWRF is set to 1. See section 2.3 for more detail. In order to provide an example of the correct computation of the diagnostics, results at a given grid point are shown in figure 7. It is shown that the total precipitable water (prw) correctly corresponds to the density weighted vertical integration of the water content along the column of air.

			Figure 7: On 2012-12-09 at 15 UTC at S 62 ◦ 4 0 38.00”, 4 ◦ 58 0 55.51”W (left): water path (prw, vertical straight line in mm top x-axis), vertical profile of water vapour (qv, line with full circles in kgkg −1 bottom x-axis), water path at each level (line with crosses). Map of water path (right) on 2012-12-09 at 15 UTC, red cross shows where the vertical accumulation is retrieved

	=== evspsblpot: potential evapotranspiration ===		=== evspsblpot: potential evapotranspiration ===

Lluis.fita: /* wsgsmax100: daily maximum wind speed of gust at 100 m */

2019-02-27T19:18:56Z

wsgsmax100: daily maximum wind speed of gust at 100 m

← Revisión anterior		Revisión del 19:18 27 feb 2019
Línea 172:		Línea 172:
	The user can also select the height at which the estimation is computed throughout the namelist parameter <CODE>z100m_wind</CODE> [100 m as default value].		The user can also select the height at which the estimation is computed throughout the namelist parameter <CODE>z100m_wind</CODE> [100 m as default value].

	Figure 6 shows different preliminary results using the three different approximations. It is illustrated (a panel) how wind-gusts are larger than the 10-m diagnostic winds, and also the difference is larger when using Monin-Obukhov how method compared to the two others methods. Certain problems (too small Monin-Obukhov length) are recognized when applying Monin-Obukhov for extrapolating wind at 100 m, which is shown in panel b, where wind gusts appear to be strong as 80 ms<SUP>-1</SUP> . Therefore a user is advised to use this method with care.		Figure 6 shows different preliminary results using the three different approximations. It is illustrated (a panel) how wind-gusts are larger than the 10-m diagnostic winds, and also the difference is larger when using Monin-Obukhov how method compared to the two others methods. Certain problems (too small Monin-Obukhov length) are recognized when applying Monin-Obukhov for extrapolating wind at 100 m, which is shown in panel b, where wind gusts appear to be strong as 80 ms<SUP>-1</SUP>. Therefore a user is advised to use this method with care.

			[[Archivo:test_wsz100.png\|frame\|50px\|Figure 6: 100 m wind estimates. Comparison between upper-level winds and estimation on 2012-12-09 at 15 UTC at S 62<SUP>◦</SUP> 4' 38.00", 4<SUP>◦</SUP> 58' 55.51" W (a): 3h-maximum eastward wind (red) at 100 m by power-law (uzmax<SUP>pl</SUP>, star marker), Monin-Obukhov theory (uzmax<SUP>mo</SUP>, cross) by logarithmic law (uzmax<SUP>ll</SUP>, sum) 10-m wind value (uas, filled triangle) and upper-level winds (ua, filled circles with line), also for the northward component (green). Temporal evolution of wind speed (b) with all approximations and upper-level winds at the closest vertical level at 100 m (on log-y scale, <z>= 107.1 m on average). Maps of three estimations: power-law (c), Monin-Obukhov (d), logarithmic-law (e) with the blue cross showing the point of previous figures. Vertical distribution of winds at the given point in Wind rose-like representation (f)
	Figure 6: 100 m wind estimates. Comparison between upper-level winds and estimation on 2012-12-09 at 15 UTC at
	S 62 ◦ 4 0 38.~~00”~~, 4 ◦ 58 0 55.~~51”W~~ (a): 3h-maximum eastward wind (red) at 100 m by power-law (uzmax pl , star marker),
	Monin-Obukhov theory (uzmax mo , cross) by logarithmic law (uzmax ll , sum) 10-m wind value (uas, filled triangle)
	and upper-level winds (ua, filled circles with line), also for the northward component (green). Temporal evolution of
	wind speed (b) with all approximations and upper-level winds at the closest vertical level at 100 m (on log-y scale,
	z = 107.1 m on average). Maps of three estimations: power-law (c), Monin-Obukhov (d), logarithmic-law (e) with the
	blue cross showing the point of previous figures. Vertical distribution of winds at the given point in Wind rose-like
	representation (f)

	=== Vertically integrated variables ===		=== Vertically integrated variables ===

Lluis.fita: /* wsgsmax100: daily maximum wind speed of gust at 100 m */

2019-02-27T19:16:21Z

wsgsmax100: daily maximum wind speed of gust at 100 m

← Revisión anterior		Revisión del 19:16 27 feb 2019
Línea 173:		Línea 173:

	Figure 6 shows different preliminary results using the three different approximations. It is illustrated (a panel) how wind-gusts are larger than the 10-m diagnostic winds, and also the difference is larger when using Monin-Obukhov how method compared to the two others methods. Certain problems (too small Monin-Obukhov length) are recognized when applying Monin-Obukhov for extrapolating wind at 100 m, which is shown in panel b, where wind gusts appear to be strong as 80 ms<SUP>-1</SUP> . Therefore a user is advised to use this method with care.		Figure 6 shows different preliminary results using the three different approximations. It is illustrated (a panel) how wind-gusts are larger than the 10-m diagnostic winds, and also the difference is larger when using Monin-Obukhov how method compared to the two others methods. Certain problems (too small Monin-Obukhov length) are recognized when applying Monin-Obukhov for extrapolating wind at 100 m, which is shown in panel b, where wind gusts appear to be strong as 80 ms<SUP>-1</SUP> . Therefore a user is advised to use this method with care.


			Figure 6: 100 m wind estimates. Comparison between upper-level winds and estimation on 2012-12-09 at 15 UTC at
			S 62 ◦ 4 0 38.00”, 4 ◦ 58 0 55.51”W (a): 3h-maximum eastward wind (red) at 100 m by power-law (uzmax pl , star marker),
			Monin-Obukhov theory (uzmax mo , cross) by logarithmic law (uzmax ll , sum) 10-m wind value (uas, filled triangle)
			and upper-level winds (ua, filled circles with line), also for the northward component (green). Temporal evolution of
			wind speed (b) with all approximations and upper-level winds at the closest vertical level at 100 m (on log-y scale,
			z = 107.1 m on average). Maps of three estimations: power-law (c), Monin-Obukhov (d), logarithmic-law (e) with the
			blue cross showing the point of previous figures. Vertical distribution of winds at the given point in Wind rose-like
			representation (f)

	=== Vertically integrated variables ===		=== Vertically integrated variables ===