Thermal, Nonthermal, and
Total Flare Energies
Brian R. Dennis
RHESSI Workshop
Locarno, Switzerland
8 – 11 June, 2005
Separating Thermal & Nonthermal
Temporal - gradual vs. impulsive
Spatial - coronal vs. footpoint
Spectral - exponential vs. power-law
Spectral – iron-line complexes
- always thermal!!!?
Difficulties with Continuum
26 April 2003 Flare Time Profile
A0 A1 A3 A1 A0
All Detectors
▬ 3 - 6 keV
▬ 6 – 12 keV
▬ 12 – 25 keV
▬ 25 – 50 keV
Time for
spectrum
RHESSI Count-rate Spectrum
Flux ratio vs. Temperature
(Caspi & Lin, 2005)
Emissivity vs. Temperature
(Caspi & Lin, 2005)
Fe-line Equivalent Width
26 April 2003
CHIANTI
Coronal Abundances
Ionization Fraction
Less FeXXV than the calculations predict.
Antonucci (1987 – SMM/BCS)
Mazzotta et al. (1998)
Conclusions
Fe & Fe/Ni complexes are real.
Fe centroid energies vary with T & count
rate.
Fe to Fe/Ni ratio varies with T.
– Different dependency for different flares.
Fe equivalent width varies with T
– Data in A1 attenuator state most reliable.
– Up to 50% less FeXXV than Mazzotta et al.
predict (Phillips).
Eagerly await XSM spectra for comparison.
Flare vs. CME Energy
Flare thermal energies:
– SXR-emitting plasma (GOES & RHESSI)
– Radiated energy (GOES)
– Conducted energy (GOES & RHESSI)
– Total Solar Irradiance increase (SORCE)
Flare nonthermal energies
– Electrons from HXRs (Holman)
– Ions from gamma-rays (Share)
CME kinetic energy
– (LASCO – Gopalswamy)
Thermal Plasma
The thermal energy content of the thermal
plasma:
Uth = 3 ne V kT = 3 k T [EM f Vapparent]1/2 erg
f is the filling factor (assumed to be 1)
Emission measures (EM) and temperatures (T)
obtained from both RHESSI and GOES soft X-ray
observations.
The source volumes (V) were obtained from
RHESSI 12 – 25 keV images
V = f Vapparent = f A3/2
A is the area inside the contour at 50% of the peak value.
Figure 1. RHESSI image at the impulsive peak of the 2 Nov. 2003 flare.
Contours: blue: 12 – 25 keV (50%), magenta: 50 – 100 keV (30 & 70%)
Radiated Energy
The energy radiated from the thermal plasma over
all wavelengths:
Lrad = EM frad(T) ergs s-1
frad(T) is the Chianti radiative loss function
assuming coronal abundances.
Total radiated energy from the flare plasma –
Ltotal = n[ Lrad(t) *Dt ] erg
where the sum is over the duration of the SXR flare.
Mazzotta et al. (1998) ionization equilibrium
Radiative Energy Loss – frad (erg cm3 s-1)
Temperature (K)
Figure 2. Radiative losses vs. plasma temperature.
Conductive Cooling
The conductive losses – Lcond – were estimated assuming
classical conduction
Lcond = A k0 T5/2 VT 4 A/l k0 T7/2 erg s-1
where k0 = 10-6 erg cm-1 s-1 K-7/2
the classical Spitzer coefficient
A is the loop cross-sectional area in cm2
l is the loop half length.
A, l, and T can be determined from RHESSI images.
However, since there is so much uncertainty in estimating
this cooling component, no values are included in this
analysis.
X8.3 flare
2 Nov. 2003
GOES
SXR
Data
X8.3 flare
2 Nov. 2003
GOES
SXR
Data
CME vs. Flare Energies
SXR-Emiting Plasma - Ltotal Peak Thermal Energy (Upeak) Electrons
Ions TSI Increase (SORCE) Equipartition
10000.0
SORCE / TIM
28 October 2003
CME Kinetic Energy (1030 ergs)
1000.0
4 November 2003
21 April 2002
23 July 2002 3 Nov. 2003
100.0
10.0
1.0
0.1
0.01 0.1 1 10 100 1000
Flare Energy Component (1030 ergs)
CME vs. Flare Energies
SXR-Emiting Plasma - Ltotal Peak Thermal Energy (Upeak) Electrons
Ions TSI Increase (SORCE) Equipartition
10000.0
SORCE / TIM
28 October 2003
CME Kinetic Energy (1030 ergs)
1000.0
4 November 2003
21 April 2002
23 July 2002 3 Nov. 2003
100.0
10.0
1.0
0.1
0.01 0.1 1 10 100 1000
Flare Energy Component (1030 ergs)
Radiated vs. Thermal Flare Energies
100.000
10.000
Radiated Energy (1030 ergs)
1.000
0.100
Ltotal
0.010
LX,total
0.001
0.1 1.0 10.0 100.0
Peak Thermal Energy, Upeak (1030 ergs)
Conclusions
Flare and CME energies are correlated for the Oct/Nov 2003
period.
Total Flare and CME energies comparable to within a factor of 10.
Peak energy in SXR-emitting plasma is only ~1% of total flare
energy in some cases.
Energy radiated by SXR-emitting plasma is only ~10% of total
flare energy in some cases.
Energy in nonthermal electrons and ions can be a large fraction of
the total flare energy.
Dominant flare energy in impulsive phase may be electrons
and/or ions leading to early peak in total solar irradiance increase
seen with SORCE/TIM.
Some of the measured radiant energy of flare may result from a
decrease in the opacity of the lower chromosphere caused by a
decrease in the H– concentration (Fontenla, private
communication).