The early martian evolution—Constraints from basin ...rcoe/eart290C... · Icarus 195 (2008) 45–60 ... region is found at the edge of the inner, outer Solar System or in between,

Icarus 195 (2008) 45–60www.elsevier.com/locate/icarus

The early martian evolution—Constraints from basin formation ages

S.C. Werner

Geological Survey of Norway (NGU), Leiv Eirikssons vei 39, N-7491 Trondheim, Norway

Received 16 January 2007; revised 20 November 2007

Available online 6 January 2008

Abstract

Impact basin formation ages give insight into the early evolution of a planet. The martian basins Hellas, Isidis and Argyre provide an importanttime-marker for the cessation of the magnetic dynamo and the crustal thickness distribution, both established before 4 Ga ago. No martian surfacesare older than 4.15 Ga based on crater count statistics, and all are younger than the oldest lunar ones. I show that the heavy bombardment periodon the Moon and Mars evolved similarly, but endogenic processes have removed the oldest martian basin record. The basin-forming projectilepopulation appears to be different from the impactor population observed today in the inner Solar System. It is yet uncertain whether the heavybombardment period is cataclysmic or characterized by the decaying flux of planetary formation.© 2008 Elsevier Inc. All rights reserved.

Keywords: Mars; Cratering; Mars, surface; Moon, surfaces

1. Introduction

The record of large impact basins on different planetary bod-ies allows us to compare the characteristics of the impactor fluxand its size–frequency distribution during the heavy bombard-ment period and to study the possible tail end of planetary for-mation. Dating the basins provides flux estimates for the largestbasin-forming bodies as well as being used as age markers forcorrelating geophysical observations such as the cessation ofthe martian dynamo or the formation of the crustal dichotomy.For the Moon, the large basins were produced no later thanabout 3.8 to 3.9 Ga ago and a similar situation is expectedfor Mars, following the marker-horizon idea (Wetherill, 1975).This idea is based on the assumption that Solar System bod-ies have undergone similar impact histories since the planetaryformation. Whether the lunar basins were formed as a result ofa spiking or rapidly declining bombardment rate, whether thisobservation is restricted to the Earth–Moon-system or is ob-servable Solar-System wide, and whether the projectile-sourceregion is found at the edge of the inner, outer Solar Systemor in between, are constantly debated issues regarding the un-derstanding of planetary surface evolution and Solar Systemdynamics. For a recent summary, see Chapman et al. (2007).

E-mail address: [email protected].

0019-1035/$ – see front matter © 2008 Elsevier Inc. All rights reserved.doi:10.1016/j.icarus.2007.12.008

In this paper, the formation ages of the large martian im-pact basins are estimated, using the derived martian productionfunction (Ivanov, 2001), and applying the martian crateringchronology model (Hartmann and Neukum, 2001) and subse-quently compared to lunar basin formation ages. Using this setof production function and chronology model implies that thebombardment rate is monotonically declining during the heavybombardment period and the projectile population is the samefor Mars and the Earth–Moon system. In addition it is assumedthat the shape of the time-averaged crater-production distribu-tion has not changed over time, with the exception of largestdiameter range of basin-forming projectiles. The applicabilityof such an approach will be discussed. The martian basin agesare used to constrain the timing of important processes andevents in the martian geological evolution.

2. The early martian cratering record

Large martian impact basins appear randomly distributed inthe heavily cratered highlands, but a few can be found in adja-cent lowland units. In this study, I attempted to identify unitsthat best represent the formation age for a particular basin.Therefore, I selected a relatively narrow band around the craterrims, considered as a zone of the ejecta blanket. Continuousejecta blankets extend about one crater radius from the crater

http://www.elsevier.com/locate/icarus

mailto:[email protected]

http://dx.doi.org/10.1016/j.icarus.2007.12.008

46 S.C. Werner / Icarus 195 (2008) 45–60

Fig. 1. Overview of the martian basins (solid-line circles), for which crater counts were performed, and some others (dashed line), which are discussed in the text orare commonly known.

rim (Moore et al., 1974), and they are sufficiently thick to coverearlier formed craters. Therefore, the ejecta blanket is appro-priate for dating the impact event itself. While the interior ofmost basins cannot be considered as pristine and crater countswould not reveal the true formation age, the ejecta blanket isprimarily subject to erosion. For the largest basins it has beenshown that melt pools possibly take a few tens of millions ofyears to solidify sufficiently enough to record later cratering(Reese et al., 2002; Ivanov, 2006). Later modification will fillthe basin interior, e.g. mare volcanism on Moon or mass trans-portation on geologically active planets. Individual basin ejectablankets, however, can be lobate and asymmetric. Therefore,I remapped these areas individually for the 19 largest impactbasins on Viking MDIM 2 images, supported by MOLA topo-graphic and MOCWA image data. The global distribution of the19 basins is given in Fig. 1, and the detailed mapped areas areshown for each individual basin in Fig. 2. The image resolution(231 m/pixel) is sufficiently high to count craters larger thanone kilometer in diameter, and to get a representative age. Mostof the later geologic activity (mainly erosion) as well as anycontamination of secondary cratering has the least effect in thelarge crater-size range. For the formation-age determination,craters larger than 15 km in diameter were considered. Tanaka(1986) pointed out that the crater densities and crater distribu-tions for the oldest Noachian units on Mars are representativeonly above crater diameters of 16 km. Strong crater obliterationis observed (e.g. Tanaka, 1986) reflected in the measurements,and expressed by a flat cumulative crater distribution even whencompared to a minus-two distribution (Fig. 3).

Fig. 3 shows the crater size–frequency distributions of the19 basins in comparison to isochrons calculated from the poly-nomial martian production function (Ivanov, 2001), and ap-plying the martian cratering chronology model (Hartmann andNeukum, 2001). The ages are listed in Table 1. The listed errorsare according to the statistical uncertainties of the measure-

ments and the least-square fitting error. Uncertainties in thechronology model (Hartmann and Neukum, 2001) are arguedto be of a factor of two. This implies a possible systematicshift towards older or younger ages, respectively. While agesbelow about 3 Ga translate linearly due to a roughly constantcratering rate, the decaying flux dominates the period discussedhere. Considering an uncertainty of a factor of two in the craterfrequency, the age variation is much smaller. Table 1 includesthe maximum and minimum systematically shifted age for eachmeasurement. For these 19 basins with detectable ejecta blan-kets, the measured ages are found with the exception of one tobe older than 3.7 Ga. That is expected following the marker-horizon idea of Wetherill (1975).

The oldest surface areas on Mars, roughly the martian south-ern highlands, e.g. Noachis Terra (crater count result Npl1 inFig. 3, details in Table 19), were formed between 4.0 to 4.1 Gaago during the heavy bombardment period (Werner, 2005). Onthe basis of counts of unambiguously defined craters, no areaswith older surface formation ages have been found (Werner,2005). Some basins such as Flaugergues or Ladon (comparebasins listed by Tanaka et al., 1992), where ejecta could notbe identified due to obvious resurfacing processes could be asold as the highland units themselves. These basins are situ-ated in the heavily cratered highland unit with an average ageof about 4.1 Ga (Werner, 2005), or occupy large areas of thelowland units with basin rims coinciding with the dichotomyboundary outline (e.g. Utopia and Acidalia). Crater counts inthe giant polygon terrain centered in the Utopia and Acidaliabasins provide in basement ages around 3.7 and 3.8 Ga (Werner,2005), unlikely to estimate the basin formation stage, but laterresurfacing (Tanaka et al., 1992). Crater counts in units adja-cent to the Utopia basin to the southwest, i.e. coinciding withthe highland–lowland boundary (north-west of the Isidis basin),yield surface ages of 3.98 Ga (Werner, 2005). Clearly distin-guished basins are not found in the northern lowlands with

Martian basin ages 47

Fig. 2. Overview of the mapped units for all 19 martian basins, which were used for counting craters.

48 S.C. Werner / Icarus 195 (2008) 45–60

Fig. 2. (continued)


Fig. 3. The crater size–frequency distributions measured for the martian impact basin. The numbers are given in Tables 3–22. Isochrons calculated for a polynomialexpression (Ivanov, 2001) and a minus-two distribution applying the martian cratering chronology model (Hartmann and Neukum, 2001) are plotted. For comparison,the inset show the polynomial and the minus-two approach for a diameter range between 1 and 500 km, equated at the diameter D = 1 km and an age of 4.2 Ga.

the exception of the relatively young and fresh-looking Lyotcrater. Earlier considerations suggested a stratigraphic positionof Lower Amazonian for Lyot (Tanaka, 1986), while the re-sult of this work shows a formation age of 3.4 Ga that impliesa Late Hesperian age. Similar-sized basins, i.e. Kepler, Gusev,and Lowell, formed during the heavy bombardment period, aswell as all other martian basins, which average between 3.7 and4.1 Ga (Table 1 and Figs. 3 and 4). Lyot’s morphology (peak-ring crater) is transitional to basins, and possibly its impactordoes not belong to the basin-forming projectile population.

The shape of the crater size–frequency distribution as wellas many other aspects of the crater-based age-determinationmethod are debated. Frey et al. (2002) and Frey (2006) countedso-called quasi-circular depressions at a diameter D � 200 km,many of which are clearly seen in the MOLA elevation data,but generally not visible in available imagery. These authorsinterpreted these depressions as remnants of strongly eroded(highlands) or deeply buried (lowlands) craters produced earlyin the martian history. Fig. 3, additionally, shows the 4.1- and4.2-Ga isochrons for a “minus-two” distribution, which Frey(2006) used to translate his measurements into absolute ages.The inset in Fig. 3 shows a detailed comparison between a“minus-two” distribution and the polynomial expression for

the diameter range between 1 and 500 km. Both distribu-tions equal at 1 and 250 km, and hence, the results are al-most directly comparable. Measurements by Frey (2006) sug-gest that the buried lowlands are older than the visible high-land surface where crater count ages are based on cratersclearly recognized by their morphology. The comparison be-tween my measurements and the proposed ages by Frey (2006)fit neatly together for the basin formation ages (Fig. 3). Themartian evolutionary history has been described substantiallyin earlier investigations based on crater counts, and is sup-ported by general stratigraphic considerations (e.g. Tanaka,1986).

3. The lunar basin formation ages

On the Moon, 42 impact basins are listed (Wilhelms, 1987).For 32 of the lunar basins, Wilhelms (1987) summarized crater-frequency measurements that were performed on the ejecta de-posits. Neukum (1983) did crater counts on a subset of thesebasins. For ten of the oldest pre-Nectarian basins crater countsare non-existent, but Wilhelms (1987) established stratigraphicrelationships between these basins. Stratigraphically, the oldestbasins are the South Pole-Aitkin basin on the far-side and Pro-

50 S.C. Werner / Icarus 195 (2008) 45–60

Table 1List of the martian impact basins, and the resulting ages from this study

Name Agein Ga

Chronologyuncertainty

N (D � 1 km) N (D � 10 km) N (D � 200 km) Areain km2

Crater (smaller than 230 km)Gusev 4.02 ± 0.02 4.12 0.0424 6.06e−4 8.90e−7 40957

3.91Lowell 3.71 ± 0.01 3.83 0.00653 9.33e−5 1.37e−7 242949

3.55Flaugergues – – – – – –Galle – – – – – –Kepler 3.92 ± 0.02 4.03 0.0221 3.16e−4 4.65e−7 60865

3.81Lyot 3.40 ± 0.05 3.62 0.0022 3.16e−5 4.64e−8 331437

2.25Secchi – – – – –

Basin (larger than 250 km)Antoniadi 3.79 ± 0.01 3.90 0.0101 1.44e−4 2.11e−7 160152

3.66Cassini 4.03 ± 0.01 4.14 0.0453 6.47e−4 9.51e−7 265082

3.92Copernicus 4.00 ± 0.05 4.10 0.0371 5.31e−4 7.80e−7 56724

3.90de Vaucouleurs 3.95 ± 0.01 4.06 0.0268 3.83e−4 5.63e−7 55865

3.84Herschel 3.95 ± 0.01 4.06 0.0268 3.83e−4 5.63e−7 43889

3.84Huygens 3.98 ± 0.02 4.08 0.0326 4.66e−4 6.84e−7 106490

3.87Koval’sky 3.96 ± 0.01 4.06 0.0286 4.09e−4 6.01e−7 86005

3.85Newcomb 4.00 ± 0.05 4.10 0.0371 5.39e−4 7.80e−7 45663

3.89Newton 4.11 ± 0.05 4.21 0.0774 1.07e−4 1.63e−6 52516

3.99Schiaparelli 3.92 ± 0.05 4.03 0.0221 3.16e−4 4.65e−7 55378

3.81Schroeter 3.92 ± 0.01 4.03 0.0221 3.16e−4 4.65e−7 43294

3.81Tikhonravov 4.10 ± 0.03 4.20 0.0724 1.03e−3 1.52e−6 139185

4.00

PlanitiaeArgyre 3.83 ± 0.01 3.94 0.0127 1.82e−4 2.67e−7 316282

3.70Hellas 3.99 ± 0.01 4.09 0.0348 4.97e−4 7.31e−7 237757

3.88Isidis 3.96 ± 0.01 4.06 0.0286 4.09e−4 6.01e−7 319580

3.85

Acidaliaa (∼3.7) – – – – 460903Utopiaa (∼3.8) – – – – 543370

The listed errors are calculated by fitting the upper and lower limit of the statistical error of the measurements and the least-square fitting error (column 2).Uncertainties in the chronology model (Hartmann and Neukum, 2001) are argued to be of a factor of two. This implies a possible systematic shift towards youngeror older ages, respectively. Here, the minimum and maximum absolute ages are calculated by halving and doubling the cumulative crater frequency at the referencediameter of the resulting best-fit isochron. While ages below about 3 Ga translate linearly due to a roughly constant cratering rate, the decaying flux dominatesthe period discussed here. Considering an uncertainty of a factor of two in the crater frequency, the absolute age variation is much smaller. Column 3 includes themaximum and minimum systematically shifted age for each measurement. For comparison with other investigations, the cumulative numbers per square kilometerfor 1, 10, and 200 km of the fitted isochron and the area size are given.

a Ages are derived from crater counts of visible craters and ghost craters in the area of giant polygonal terrain (Werner, 2005).

cellarum basin on the near-side of the Moon (Wilhelms, 1987).The South Pole-Aitkin basin is saturated by superposed impactstructures, while the Procellarum basin is modified by mare vol-canism.

In this study, lunar basin formation ages were compiled bymaking use of the crater-frequency measurements obtained byNeukum (1983) and Wilhelms (1987). Their numbers weretranslated to absolute ages (Table 2) applying different com-


Fig. 4. The frequency distribution of the ages of lunar (open and hatched) and martian (filled) impact basins given here. The dashed-hatched area covering the agerange between 4.15 and older indicates the time period for the 10 basins where no crater frequency measurements are available. The frequency distribution of agesof lunar impact basins are given as a homogeneous set (open bars) as published by Wilhelms (1987) and as a compilation (hatched bars) from Neukum (1983) andWilhelms (1987). All crater frequencies were translated by the same blend of chronology model crater size–frequency distribution.

binations of crater size–frequency descriptions and chronologymodels (Neukum, 1983; Ivanov, 2001; Neukum et al., 2001).

For the 10 basins that have no crater frequency attributed,several interpretations of the age distributions are possible:(1) the South Pole-Aitkin basin age is approximately the sameas the average lunar highland age of about 4.35 Ga (e.g.Lugmair and Carlson, 1978), providing an upper limit. As alower limit, Al-Khwarizmi/King is used, which is stratigraph-ically the youngest in the sequence of old basins (Wilhelms,1987). It has been dated in terms of superposed crater fre-quency to form about 4.1 Ga ago (crater counts by Wilhelms,1987, Table 2). While no information is available for these re-maining basins, they are distributed between these two bound-aries. (2) Alternatively, the age distribution could peak at oneor the other boundary. Nevertheless, the correlation betweenradiometric ages of lunar samples and crater frequencies forthis age range, and thus the cratering-chronology models, isstill strongly debated (e.g. Tera et al., 1973; Ryder, 1990;Stöffler and Ryder, 2001; Hartmann and Neukum, 2001; Gomeset al., 2005; Chapman et al., 2007). Therefore, for the remain-ing “uncounted” ten basins any age remains highly specula-tive. When comparing the ages resulting from different blendsof crater size–frequency distribution and cratering chronologymodels, systematic differences are obvious. For example, theearlier version of crater distribution and chronology model(Neukum, 1983) tend to result in ages systematically older thanthe most-up-to-date combination (Ivanov, 2001; Neukum et al.,2001). When comparing crater frequencies found by Wilhelms(1987) and Neukum (1983), the latter’s numbers are higher onaverage. Admittedly, these differences cannot be judged, be-cause no detailed distribution and unit boundary informationare obtainable.

In summary, the lunar basins could range between 3.85 Gaand about 4.35 Ga, while the most homogeneous set of ages,also in comparison with the martian results, defined by crater

counts [counts: Wilhelms (1987); crater size–frequency distri-bution: Ivanov (2001); cratering chronology model: Neukum etal. (2001)] range between about 3.7 and 4.1 Ga.

4. The basin formation period on Mars and Moon:A comparison

The martian surface observed today appears to be no olderthan about 4.15 Ga, which follows from the counts presentedhere and agrees with the results of Frey (2006). As an impor-tant reference locality, Noachis Terra (the type region for theoldest stratigraphic unit) provides an age of 4.02 Ga (Werner,2005) based on crater counts, in accordance with the oldestbasin ages. Almost all martian basins are 3.7 to 4.0 Ga old oryounger, while datable lunar basins give ages between 3.7 and4.1 Ga or 3.85 and 4.25 Ga, depending which subset of countsis considered (Fig. 4). Hence, on average, lunar basins appearolder (3.95 Ga) than the martian ones (3.92 Ga) with the ma-jority of occurrences prior to 3.95 Ga. The oldest surface unitson the Moon are considered to be 4.35 Ga old (Lugmair andCarlson, 1978; Wilhelms, 1987). The lunar distribution shows amaximum 3.95 Ga ago, while the martian data have a maximumbasin occurrence at 4.0 Ga ago. Generally, crater statistic-basedages indicate that the basin formation occurred during a periodlasting about 400 Ma, as opposed to a cataclysmic interpretation(less than 100 Ma). Certainly, one can consider this as result ofthe input model (cratering chronology model), even though thismodel application results in a coherent sequence of events whenapplied to understanding lunar or martian geologic evolutionaryhistories.

In order to better compare the lunar and martian basin pop-ulation, I plotted the frequency of basins versus the formationage (Fig. 4). Mars, for the time span 3.7–4.0 Ga, shows a totalof 15 basins, whereas the Moon has 23 and 15 basins, respec-tively, for the same period. These numbers relate to the martian

52 S.C. Werner / Icarus 195 (2008) 45–60

Table 2List of the lunar impact basins, starting from the oldest towards the youngest (Wilhelms, 1987)

Name Diameterin km

Densitya

N (D � 20 km)pro 106 km2

Areab in106 km2

Agec

in GaN83

Aged

in GaN&I01

Densitye

N (D � 1 km)pro km2

Agef

in GaN&I01

Pre-Nectarian basins (stratigraphically related)Procellarum 3200South Pole-Aitken 2500 ∼4.35Tsiolkovskiy–Stark 700Grissom–White 600Insularum 600Marginis 580Flamsteed–Billy 570Balmer–Kapteyn 550Werner–Airy 500Pingré–Hausen 300Al–Khwarizmi/King 590 197 0.320 4.17 4.04

Pre-Nectarian basins (ages based on crater frequencies)Fecunditatis 990 – – – –Australe 880 (>212) (0.608) 4.18 4.05Tranquillitatis 800 – – – –Mutus–Vlacq 700 225 0.336 4.19 4.06Nubium 690 – – – –Lomonosov–Fleming 620 177 0.356 4.15 4.03Ingenii 650 162 0.228 4.14 4.01Poincaré 340 190 (0.168) 4.16 4.04Keeler–Heaviside 780 186 0.371 4.16 4.03 3.82 × 10−1 4.27Coulomb–Sarton 530 (145) (0.296) 4.12 4.00Smythii 840 166 0.445 4.14 4.02 2.93 × 10−1 4.23Lorentz 360 159 0.208 4.14 4.01 2.63 × 10−1 4.21Amundsen–Ganswindt 355 (108) (0.102) 4.08 3.95Schiller–Zucchius 325 (112) (0.143) 4.09 3.96Planck 325 (110) 0.082 4.08 3.95Birkhoff 330 127 0.401 4.11 3.97 2.46 × 10−1 4.20Freundlich–Sharanov 600 129 0.629 4.11 3.98 1.57 × 10−1 4.14Grimaldi 430 (97) (0.154) 4.06 3.93Apollo 505 119 0.480 4.10 3.97 1.38 × 10−1 4.12

Nectarian basins (ages based on crater frequencies)Nectaris 860 79 1.286 4.03 3.90 1.00 × 10−1 4.07Mendel–Rydberg 630 (73) (0.247) 4.02 3.89 9.76 × 10−2 4.07Moscoviense 445 87 0.609 4.05 3.92 7.29 × 10−2 4.02Korolev 440 79 1.113 4.03 3.90 8.12 × 10−2 4.04Mendeleev 330 63 0.569 4.00 3.87 8.37 × 10−2 4.04Humboldtianum 700 62 0.515 4.00 3.86 8.13 × 10−2 4.04Humorum 820 56 0.428 3.98 3.85 5.95 × 10−2 3.99Crisium 1060 53 0.843 3.97 3.94 5.70 × 10−2 3.99Serenitatis 740 (83) 0.108 4.04 3.91Hertzsprung 570 58 0.883 3.99 3.85 5.68 × 10−2 3.98Sikorsky–Rittenhouse 310 (27) (0.075) 3.87 3.72Bailly 300 (31) (0.096) 3.89 3.74

Imbrium 3.71 × 10−2 3.92

Orientale 2.24 × 10−2 3.84

Crater frequencies after Wilhelms (1987) and Neukum (1983), translated into ages using different crater size–frequency distributions (Neukum, 1983; Neukum etal., 2001) and a chronology model (Neukum, 1983; Ivanov, 2001).

a Crater densities after Wilhelms (1987).b Reference area for the crater densities given by Wilhelms (1987).c Ages derived from crater densities given by Wilhelms (1987) using the crater size–frequency distribution and chronology model from Neukum (1983).d Ages derived from crater densities given by Wilhelms (1987) using the most up-to-date crater size–frequency distribution from Ivanov (2001) and chronology

model from Neukum et al. (2001).e Crater densities after Neukum (1983).f Ages derived from crater densities given by Neukum (1983) using the most up-to-date crater size–frequency distribution from Ivanov (2001) and chronology

model from Neukum et al. (2001).


Fig. 5. The sequence of events as discussed in the text. On the left the martian epochs and cumulative crater frequencies are plotted, on the right translated absoluteages are shown. As time indicator martian meteorite crystallization ages (Nyquist et al., 2001) and basin formation ages are given, and the histogram for the martianbasin formation ages is added.

highland surface that contains basins and is almost two timeslarger than the entire lunar surface, the reference surface forthe considered lunar basins. The current/modern cratering-rateratio between Mars and Moon ranges between 0.6 to 1.2, de-pending on the steepness of the crater distribution, and basedon an impact-rate ratio of 2.04 for an asteroid of the samesize (Ivanov, 2001). A preliminary investigation by Werner andNeukum (2003) concluded that the number of basins per ageperiod per unit area is the same for both Moon and Mars. Here,it is shown that the number of basins formed per age periodper unit area for the Moon is higher by a factor of two to threewhen the original crater frequencies are translated by a coherentset of crater size–frequency distribution and cratering chronol-ogy model. Considering modern cratering rates (Ivanov, 2001)the trend is reversed. As pointed out above, the crater/projectilesize–frequency distribution is considered to be time-invariantwith the exception of the largest basin-forming projectile popu-lation. The latter impacts only during the “heavy bombardmentperiod.” The crater distribution superposed to the basins (diam-eters less than about 300 km) apparently is similar on Moon andMars and also over time. The basin-forming population disap-pears latest 3.7 Ga ago.

In summary, the oldest martian crustal structures morpho-logically observed today are no more than 4.15 Ga old whereasthe lunar surface record probably dates back to 4.3–4.4 Ga.Using the global lunar basin record, as listed by Wilhelms(1987), an average surface age of 4.2 Ga is received. The globalmartian basin record of diameters and location cataloged byBarlow (1988) supports this result, yielding an average surfaceage of 4.1 Ga for the highland surface. On Mars, endogenic

and surface erosional processes have probably erased the ear-lier record. The basin-forming projectile population is mostlikely different from the general impactor population, com-monly linked to the asteroidal projectile class, with referenceto the inner Solar System.

5. The early evolution of Mars

The basin chronology can be used for timing two key geo-physical processes on early Mars: the existence and cessationof a magnetic field and the formation of the crustal thicknessdichotomy on Mars. During the Mars Global Surveyor mission,vector magnetic field observations of the martian crust wereacquired. The location of observed magnetic field sources ofmultiple scales, strength, and geometry correlates remarkablywell with the ancient cratered terrain of the martian highlands(Acuña et al., 1999). On the other hand, these remanent magne-tization sources appear to be absent in the lowland plains, nearlarge impact basins such as Hellas and Argyre, and in most ofthe volcanic regions. The Mars Global Surveyor mission notonly obtained a detailed topography through the MOLA mea-surements, but also mapped the gravity field through two-wayDoppler tracking of the spacecraft (Smith et al., 1999a, 1999b).Based on the topography and the gravity fields, a map of thecrustal thickness distribution was derived (Zuber et al., 2000;Neumann et al., 2004). Martian basin formation ages may alsogive a time frame for the thermodynamical evolution of Mars(Nimmo and Tanaka, 2005).

The sequence of events discussed here, is illustrated inFig. 5. The visible crustal age in the southern highland unit,

54 S.C. Werner / Icarus 195 (2008) 45–60

carrying the strongest remanent magnetization, appears to haveformed before 4.1 to 4.2 Ga, whilst the oldest known mar-tian meteorites indicate a crustal age of about 4.5 Ga (Nyquistet al., 2001). Whether the martian crust, as observable today,formed earlier than 4.2 Ga, is beyond the limit of crater count-ing methods, because saturation may have been reached andnet accumulation may be indetermined, due to the presumablyhigh impact flux. Most of the southern highland areas, showingweaker magnetization, are subsequently resurfaced. The mar-tian dynamo was active before 3.9 Ga ago. The cessation ofthe magnetic field seems to be correlated in time with the for-mation of the large impact basins, Hellas, Argyre, Isidis andUtopia, which are devoid of long-wavelength magnetic anom-alies. At least Hellas and Isidis formed about 4.0 Ga ago, whileArgyre and probably also Utopia are somewhat younger. Al-though, the period of cooling of a melt pool formed by sucha large impact event and the temperature drop below the Curietemperature (acquisition of a remanent magnetization) is poorlyunderstood, ages found for basins on Earth such as the Vrede-fort (South Africa) suggest a cooling within a few millions ofyears since U-Pb zircon ages (Kamo et al., 1996) and 40Ar–39Ar mineral ages (Reimold et al., 1996) statistically overlap ataround 2025 Ma. For Mars models also suggest rapid cooling(Reese and Solomatov, 2006), while others produce sustainingmelt pools over tens to hundreds (unlikely) of millions of years(Reese et al., 2002; Ivanov, 2006). This leaves a range of about100 million years for the actual timing of dynamo cessation,and the basin formation ages provide only a minimum age. Dat-ing the cessation of the martian dynamo also provide importantconstraint on understanding the thermal evolution and internaldynamics, i.e. core-freezing or heat flux across the core-mantleboundary became too low to sustain a geodynamo.

Basin ages can also be used as time-markers for the forma-tion of the crustal dichotomy, as reflected in the North–Southvarying crustal thickness. The observed crustal thickness distri-bution was established early (before 4.0 Ga) and was modifiedthrough these impacts, punching holes, such as Isidis, Hellas,Argyre and supposedly Utopia. Compared to the Moon, endo-genic and surface erosional processes have modified the crater-ing record. To modify the record of large (100’s and 1000’s km)craters the resurfacing should be effective within a surface layera few kilometers thick, but the remanent magnetization andcrustal distribution was maintained as it formed very early inmartian history. This implies a higher intensity of resurfacingprocesses, before the cessation of the martian dynamo and es-tablishment of the crustal dichotomy.

Conclusively, the early bombardment history reflects a sim-ilar history for the Moon and Mars with respect to the time de-pendence of impact rate, the source of impactors, and the shapeof the crater size–frequency distribution, with the exception ofthe basin-forming projectile population. Here, no straightfor-ward interpretation is possible following the idea that the basinswere formed by the same projectile group, with respect to theimpact rate ratios valid for the current flux and the shape ofthe crater size–frequency distribution observed in the diameterrange above 300 km in diameter. The idea of a marker-horizonas first suggested by Wetherill (1975) as a dynamical explana-

tion and elaborated by, e.g., Morbidelli et al. (2001), Levisonet al. (2001), Chambers and Lissauer (2002), and Gomes et al.(2005) is supported by this work. Nevertheless, to determinewhether the early bombardment appears as a cataclysmic eventor decaying flux is not firmly established. To establish the ap-plied cratering chronology model the flux prior to 4.0 Ga agoneeds further support from a sample return and the determina-tion of radiometric ages. Samples from one of the oldest lunarbasins, such as South Pole-Aitkin on the Moon would be ofcrucial importance. One or two samples of previously crater-based dated units could provide evidence, for the validity of thechronology transfer between Moon and Mars, and to confirmthe simultaneous bombardment history.

Acknowledgments

The author thanks Alexander T. Basilevsky and GerhardNeukum and the two referees John E. Chappelow and KennethL. Tanaka for helpful discussion and suggestions, and acknowl-edges project funding by the Deutsche Forschungsgemeinschaft(DFG).

Appendix A

Tables 3–22: Crater size–frequency measurements for all 19martian basins and the Noachis Terra area as discussed in thetext.

Table 3Argyre

Diameter [km] N (cum/km2) Error

1.3 6.86E−04 4.66E−051.5 6.80E−04 4.64E−051.7 6.70E−04 4.60E−052 6.39E−04 4.49E−052.5 5.53E−04 4.18E−053 5.09E−04 4.01E−053.5 4.68E−04 3.85E−054 4.11E−04 3.60E−054.5 3.79E−04 3.46E−055 3.48E−04 3.32E−056 3.10E−04 3.13E−057 2.78E−04 2.97E−058 2.59E−04 2.86E−059 2.40E−04 2.76E−05

10 2.15E−04 2.61E−0511 1.93E−04 2.47E−0512 1.83E−04 2.41E−0513 1.61E−04 2.26E−0514 1.52E−04 2.19E−0515 1.33E−04 2.05E−0517 1.20E−04 1.95E−0520 7.90E−05 1.58E−0525 3.79E−05 1.10E−0530 2.85E−05 9.49E−0635 2.53E−05 8.94E−0640 1.58E−05 7.07E−0650 9.49E−06 5.48E−06


Table 4Antonadi


0.9 1.09E−03 8.26E−051 1.08E−03 8.21E−051.1 1.07E−03 8.19E−051.2 1.04E−03 8.07E−051.3 1.02E−03 8.00E−051.4 9.93E−04 7.87E−051.5 9.74E−04 7.80E−051.7 8.80E−04 7.41E−052 7.99E−04 7.06E−052.5 6.68E−04 6.46E−053 5.56E−04 5.89E−053.5 4.81E−04 5.48E−054 4.00E−04 5.00E−054.5 3.62E−04 4.76E−055 3.31E−04 4.55E−056 3.06E−04 4.37E−057 2.56E−04 4.00E−058 2.25E−04 3.75E−059 1.87E−04 3.42E−05

10 1.56E−04 3.12E−0511 1.31E−04 2.86E−0512 1.12E−04 2.65E−0513 9.99E−05 2.50E−0514 8.74E−05 2.34E−0515 8.12E−05 2.25E−0517 6.87E−05 2.07E−0520 5.00E−05 1.77E−0525 3.75E−05 1.53E−0535 3.12E−05 1.40E−0570 1.87E−05 1.08E−05

170 6.24E−06 6.24E−06

Table 5Cassini


0.6 1.89E−03 8.45E−050.9 1.89E−03 8.44E−051 1.88E−03 8.43E−051.1 1.85E−03 8.36E−051.2 1.84E−03 8.34E−051.3 1.80E−03 8.24E−051.4 1.77E−03 8.17E−051.5 1.75E−03 8.12E−051.7 1.61E−03 7.79E−052 1.43E−03 7.35E−052.5 1.18E−03 6.67E−053 1.03E−03 6.23E−053.5 8.75E−04 5.75E−054 7.77E−04 5.41E−054.5 7.21E−04 5.21E−055 6.79E−04 5.06E−056 5.62E−04 4.60E−057 5.24E−04 4.45E−058 4.60E−04 4.17E−059 4.15E−04 3.96E−05

10 3.85E−04 3.81E−0511 3.51E−04 3.64E−0512 3.28E−04 3.52E−0513 3.09E−04 3.42E−0514 2.90E−04 3.31E−0515 2.53E−04 3.09E−0517 2.26E−04 2.92E−05

Table 5 (continued)


20 1.89E−04 2.67E−0525 1.32E−04 2.23E−0530 1.06E−04 2.00E−0535 7.54E−05 1.69E−0540 6.41E−05 1.56E−0545 4.90E−05 1.36E−0550 3.40E−05 1.13E−0560 2.26E−05 9.24E−0670 1.13E−05 6.53E−0680 7.54E−06 5.34E−06

110 3.77E−06 3.77E−06

Table 6Copernicus


0.7 1.59E−03 1.67E−040.9 1.57E−03 1.66E−041 1.55E−03 1.65E−041.1 1.53E−03 1.64E−041.2 1.52E−03 1.63E−041.3 1.50E−03 1.63E−041.5 1.43E−03 1.59E−041.7 1.36E−03 1.55E−042 1.18E−03 1.44E−042.5 9.87E−04 1.32E−043 9.17E−04 1.27E−043.5 8.29E−04 1.21E−044 7.58E−04 1.16E−044.5 7.05E−04 1.11E−045 6.52E−04 1.07E−046 4.94E−04 9.33E−057 4.41E−04 8.81E−058 4.05E−04 8.45E−059 3.70E−04 8.08E−05


Table 7De Vaucouleurs


0.9 2.29E−03 2.03E−041 2.27E−03 2.02E−041.1 2.24E−03 2.00E−041.3 2.22E−03 1.99E−041.4 2.18E−03 1.98E−041.5 2.06E−03 1.92E−041.7 1.88E−03 1.83E−042 1.70E−03 1.74E−042.5 1.31E−03 1.53E−043 1.11E−03 1.41E−043.5 1.04E−03 1.36E−04

(continued on next page)

56 S.C. Werner / Icarus 195 (2008) 45–60

Table 7 (continued)


4 9.67E−04 1.32E−044.5 8.77E−04 1.25E−045 8.41E−04 1.23E−046 7.16E−04 1.13E−047 6.27E−04 1.06E−048 5.37E−04 9.80E−059 4.12E−04 8.58E−05

10 3.58E−04 8.01E−0511 2.86E−04 7.16E−0513 2.69E−04 6.93E−0514 2.51E−04 6.70E−0515 2.33E−04 6.45E−0517 1.97E−04 5.94E−0520 1.43E−04 5.06E−0525 8.95E−05 4.00E−0530 7.16E−05 3.58E−0550 3.58E−05 2.53E−05

Table 8Gusev


1.1 2.10E−03 2.26E−041.2 2.08E−03 2.25E−041.3 2.03E−03 2.22E−041.4 2.00E−03 2.21E−041.5 1.98E−03 2.20E−041.7 1.86E−03 2.13E−042 1.71E−03 2.04E−042.5 1.42E−03 1.86E−043 1.25E−03 1.74E−044 1.17E−03 1.69E−044.5 1.15E−03 1.67E−045 1.05E−03 1.60E−046 9.52E−04 1.52E−047 7.81E−04 1.38E−048 7.08E−04 1.31E−049 6.84E−04 1.29E−04


Table 9Hellas


0.9 1.06E−03 6.69E−051 1.06E−03 6.68E−051.1 1.05E−03 6.65E−051.2 1.04E−03 6.61E−051.3 1.03E−03 6.60E−051.4 1.02E−03 6.54E−051.5 1.01E−03 6.52E−051.7 9.97E−04 6.48E−052 9.63E−04 6.36E−05

Table 9 (continued)


2.5 8.54E−04 5.99E−053 7.70E−04 5.69E−053.5 6.90E−04 5.39E−054 6.56E−04 5.25E−054.5 6.39E−04 5.19E−055 6.10E−04 5.06E−056 5.43E−04 4.78E−057 4.84E−04 4.51E−058 4.58E−04 4.39E−059 4.25E−04 4.23E−05

10 3.83E−04 4.01E−0511 3.66E−04 3.92E−0512 3.45E−04 3.81E−0513 3.15E−04 3.64E−0514 2.82E−04 3.44E−0515 2.52E−04 3.26E−0517 2.19E−04 3.03E−0520 1.72E−04 2.69E−0525 1.39E−04 2.42E−0530 1.01E−04 2.06E−0535 7.15E−05 1.73E−0540 5.05E−05 1.46E−0545 4.21E−05 1.33E−0550 2.94E−05 1.11E−0560 8.41E−06 5.95E−0670 4.21E−06 4.21E−06

Table 10Herschel


0.6 1.96E−03 2.11E−041 1.89E−03 2.08E−041.3 1.85E−03 2.05E−041.4 1.75E−03 2.00E−041.5 1.69E−03 1.96E−041.7 1.50E−03 1.85E−042 1.30E−03 1.72E−042.5 9.34E−04 1.46E−043 7.97E−04 1.35E−043.5 7.75E−04 1.33E−044 7.06E−04 1.27E−044.5 6.15E−04 1.18E−045 5.24E−04 1.09E−046 4.56E−04 1.02E−047 4.10E−04 9.67E−058 3.87E−04 9.39E−059 3.42E−04 8.82E−05

12 2.96E−04 8.22E−0513 2.73E−04 7.89E−0515 2.28E−04 7.21E−0517 2.05E−04 6.84E−0520 1.37E−04 5.58E−0525 9.11E−05 4.56E−0530 6.84E−05 3.95E−0570 2.28E−05 2.28E−05

Table 11Huygens


1.1 1.39E−03 1.14E−041.2 1.37E−03 1.13E−041.3 1.35E−03 1.13E−04


Table 11 (continued)



10 2.72E−04 5.06E−0511 2.54E−04 4.88E−0512 2.44E−04 4.79E−0514 2.07E−04 4.40E−0515 1.78E−04 4.09E−0517 1.50E−04 3.76E−0520 1.13E−04 3.25E−0525 9.39E−05 2.97E−0530 7.51E−05 2.66E−0535 6.57E−05 2.48E−0540 5.63E−05 2.30E−0550 4.70E−05 2.10E−0560 2.82E−05 1.63E−05

Table 12Isidis


0.6 1.81E−03 7.52E−050.8 1.81E−03 7.52E−050.9 1.80E−03 7.51E−051 1.80E−03 7.50E−051.1 1.79E−03 7.48E−051.2 1.78E−03 7.46E−051.3 1.75E−03 7.40E−051.4 1.69E−03 7.27E−051.5 1.63E−03 7.15E−051.7 1.55E−03 6.96E−052 1.37E−03 6.56E−052.5 1.16E−03 6.02E−053 1.04E−03 5.71E−053.5 9.20E−04 5.37E−054 8.42E−04 5.13E−054.5 8.07E−04 5.03E−055 7.38E−04 4.81E−056 6.35E−04 4.46E−057 5.82E−04 4.27E−058 5.32E−04 4.08E−059 4.72E−04 3.85E−05




40 4.07E−05 1.13E−0545 2.19E−05 8.28E−0650 1.88E−05 7.66E−0660 9.39E−06 5.42E−06

Table 13Kepler


1.7 6.90E−04 1.06E−042 6.41E−04 1.03E−042.5 5.91E−04 9.86E−053 5.75E−04 9.72E−054 5.59E−04 9.58E−055 5.42E−04 9.44E−056 5.26E−04 9.29E−057 4.44E−04 8.54E−058 3.94E−04 8.05E−059 3.45E−04 7.53E−05

10 3.29E−04 7.35E−0512 2.96E−04 6.97E−0514 2.63E−04 6.57E−0515 2.30E−04 6.15E−0517 1.97E−04 5.69E−0520 1.15E−04 4.35E−0525 9.86E−05 4.02E−0530 3.29E−05 2.32E−0570 1.64E−05 1.64E−05

Table 14Kovalsky


0.7 1.62E−03 1.37E−040.8 1.60E−03 1.37E−041 1.59E−03 1.36E−041.1 1.53E−03 1.34E−041.2 1.48E−03 1.31E−041.3 1.45E−03 1.30E−041.4 1.37E−03 1.26E−041.5 1.31E−03 1.24E−041.7 1.22E−03 1.19E−042 1.10E−03 1.13E−042.5 9.42E−04 1.05E−043 8.02E−04 9.66E−053.5 7.67E−04 9.45E−054 7.21E−04 9.16E−054.5 6.63E−04 8.78E−055 6.05E−04 8.38E−056 5.58E−04 8.06E−057 5.00E−04 7.62E−058 4.53E−04 7.26E−059 3.60E−04 6.47E−05

10 3.26E−04 6.15E−0511 2.79E−04 5.70E−0512 2.67E−04 5.58E−0513 2.56E−04 5.45E−0514 2.44E−04 5.33E−0515 2.33E−04 5.20E−0517 1.98E−04 4.79E−0520 1.51E−04 4.19E−0525 1.05E−04 3.49E−0530 6.98E−05 2.85E−0540 4.65E−05 2.33E−0570 2.33E−05 1.64E−05

58 S.C. Werner / Icarus 195 (2008) 45–60

Table 15Lowell


1.7 1.56E−04 2.54E−052 1.52E−04 2.50E−052.5 1.48E−04 2.47E−053 1.40E−04 2.40E−054 1.32E−04 2.33E−054.5 1.23E−04 2.25E−055 1.11E−04 2.14E−056 9.88E−05 2.02E−057 7.82E−05 1.79E−058 6.17E−05 1.59E−059 5.76E−05 1.54E−05


Table 16Lyot



10 3.02E−05 9.54E−0611 2.41E−05 8.53E−0614 2.11E−05 7.98E−0615 1.81E−05 7.39E−0620 9.05E−06 5.23E−06

Table 17Newcomb


1.1 1.16E−03 1.59E−041.5 1.14E−03 1.58E−041.7 1.09E−03 1.55E−042 9.85E−04 1.47E−042.5 9.20E−04 1.42E−043 8.10E−04 1.33E−043.5 7.23E−04 1.26E−044 6.57E−04 1.20E−044.5 5.91E−04 1.14E−045 5.47E−04 1.09E−046 4.16E−04 9.55E−057 3.94E−04 9.29E−059 3.72E−04 9.03E−05

10 3.28E−04 8.48E−05




120 2.19E−05 2.19E−05

Table 18Newton


0.45 2.86E−03 2.33E−040.5 2.84E−03 2.32E−040.6 2.82E−03 2.32E−040.7 2.78E−03 2.30E−040.8 2.65E−03 2.25E−040.9 2.49E−03 2.18E−041 2.40E−03 2.14E−041.1 2.29E−03 2.09E−041.2 2.11E−03 2.01E−041.3 1.94E−03 1.92E−041.4 1.89E−03 1.89E−041.5 1.73E−03 1.82E−041.7 1.60E−03 1.75E−042 1.47E−03 1.67E−042.5 1.24E−03 1.54E−043 1.03E−03 1.40E−043.5 8.95E−04 1.31E−044 8.19E−04 1.25E−044.5 7.81E−04 1.22E−045 7.24E−04 1.17E−046 6.86E−04 1.14E−047 6.28E−04 1.09E−048 5.33E−04 1.01E−049 5.14E−04 9.89E−05

10 4.95E−04 9.71E−0511 4.57E−04 9.33E−0512 4.38E−04 9.13E−0513 3.81E−04 8.52E−0514 3.05E−04 7.62E−0515 2.86E−04 7.37E−0517 2.67E−04 7.12E−0520 2.48E−04 6.87E−0525 1.90E−04 6.02E−0530 1.33E−04 5.04E−0535 9.52E−05 4.26E−0545 7.62E−05 3.81E−0550 5.71E−05 3.30E−0570 3.81E−05 2.69E−0590 1.90E−05 1.90E−05

Table 19Npl1 (Terra Noachis)


0.6 2.28E−03 6.62E−050.7 2.27E−03 6.61E−050.8 2.26E−03 6.59E−050.9 2.21E−03 6.53E−051 2.16E−03 6.44E−05




1.1 2.10E−03 6.35E−051.2 1.98E−03 6.17E−051.3 1.86E−03 5.98E−051.4 1.72E−03 5.75E−051.5 1.60E−03 5.55E−051.7 1.41E−03 5.21E−052 1.18E−03 4.76E−052.5 9.31E−04 4.23E−053 8.02E−04 3.93E−053.5 7.00E−04 3.67E−054 6.33E−04 3.49E−054.5 5.87E−04 3.36E−055 5.44E−04 3.24E−056 5.06E−04 3.12E−057 4.66E−04 2.99E−058 4.27E−04 2.87E−059 4.04E−04 2.79E−05

10 3.73E−04 2.68E−0511 3.56E−04 2.62E−0512 3.27E−04 2.51E−0513 2.96E−04 2.39E−0514 2.77E−04 2.31E−0515 2.56E−04 2.22E−0517 2.39E−04 2.14E−0520 2.08E−04 2.00E−0525 1.52E−04 1.71E−0530 1.13E−04 1.48E−0535 8.27E−05 1.26E−0540 6.54E−05 1.12E−0545 4.81E−05 9.62E−0650 3.46E−05 8.16E−0660 2.31E−05 6.66E−0670 1.35E−05 5.09E−0680 7.69E−06 3.85E−0690 3.85E−06 2.72E−06

Table 20Schiaparelli


1.2 1.12E−03 1.42E−041.4 1.10E−03 1.41E−041.5 1.07E−03 1.39E−041.7 1.03E−03 1.36E−042 8.85E−04 1.26E−042.5 6.14E−04 1.05E−043 4.69E−04 9.21E−053.5 4.33E−04 8.85E−054 3.79E−04 8.28E−054.5 3.25E−04 7.66E−055 3.07E−04 7.45E−056 2.71E−04 6.99E−057 2.53E−04 6.76E−058 2.17E−04 6.26E−059 1.99E−04 5.99E−05

11 1.81E−04 5.71E−0513 1.44E−04 5.11E−0515 1.26E−04 4.78E−0520 1.08E−04 4.42E−0540 3.61E−05 2.55E−0550 1.81E−05 1.81E−05

Table 21Schroeter


0.8 1.99E−03 2.14E−040.9 1.94E−03 2.12E−041 1.89E−03 2.09E−041.1 1.85E−03 2.07E−041.2 1.78E−03 2.03E−041.3 1.66E−03 1.96E−041.4 1.52E−03 1.88E−041.5 1.43E−03 1.82E−041.7 1.18E−03 1.65E−042 1.04E−03 1.55E−042.5 8.55E−04 1.41E−043 7.85E−04 1.35E−043.5 7.39E−04 1.31E−044 6.24E−04 1.20E−045 6.01E−04 1.18E−046 5.08E−04 1.08E−047 4.85E−04 1.06E−048 4.16E−04 9.80E−059 3.23E−04 8.64E−05

10 3.00E−04 8.33E−0511 2.77E−04 8.00E−0512 2.54E−04 7.66E−0513 2.31E−04 7.30E−0514 2.08E−04 6.93E−0515 1.85E−04 6.53E−0517 1.62E−04 6.11E−0520 1.39E−04 5.66E−0530 6.93E−05 4.00E−05

100 2.31E−05 2.31E−05

Table 22Tikhonravov


0.8 1.58E−03 1.07E−041.2 1.57E−03 1.06E−041.3 1.55E−03 1.06E−041.4 1.54E−03 1.05E−041.5 1.52E−03 1.04E−041.7 1.46E−03 1.02E−042 1.30E−03 9.67E−052.5 1.14E−03 9.03E−053 9.91E−04 8.44E−053.5 8.48E−04 7.80E−054 7.54E−04 7.36E−054.5 6.75E−04 6.97E−055 6.25E−04 6.70E−056 5.46E−04 6.26E−057 5.03E−04 6.01E−058 4.31E−04 5.57E−059 3.81E−04 5.23E−05

10 3.52E−04 5.03E−0511 3.38E−04 4.93E−0512 3.23E−04 4.82E−0513 3.02E−04 4.66E−0514 2.95E−04 4.60E−0515 2.80E−04 4.49E−0517 2.66E−04 4.37E−0520 2.23E−04 4.00E−0525 1.87E−04 3.66E−0530 1.29E−04 3.05E−0535 1.15E−04 2.87E−0540 1.01E−04 2.69E−05

(continued on next page)

60 S.C. Werner / Icarus 195 (2008) 45–60



45 7.18E−05 2.27E−0550 5.75E−05 2.03E−0560 2.87E−05 1.44E−0590 1.44E−05 1.02E−05

100 7.18E−06 7.18E−06

References

Acuña, M.H., Connerney, J.E.P., Ness, N.F., Lin, R.P., Mitchell, D., Carlson,C.W., McFadden, J., Anderson, K.A., Reme, H., Mazelle, C., Vignes, D.,Wasilewski, P., Cloutier, P., 1999. Global distribution of crustal magne-tization discovered by the Mars Global Surveyor MAG/ER experiment.Science 284, 790–793.

Barlow, N.G., 1988. Crater size–frequency distributions and a revised martianrelative chronology. Icarus 75, 285–305.

Chambers, J.E., Lissauer, J., 2002. A new dynamical model for the lunar lateheavy bombardment. Lunar Planet. Sci. XXXIII. Abstract 1093.

Chapman, C.R., Cohen, B.A., Grinspoon, D.H., 2007. What are real constraintson the existence and magnitude of the late heavy bombardment. Icarus 189,233–245.

Frey, H.V., 2006. Impact constraints on the age and origin of the lowlands ofMars. Geophys. Res. Lett. 33 (8), doi:10.1029/2005GL024484.

Frey, H.V., Roark, J.H., Shockey, K.M., Frey, E.L., Sakimoto, S.E.H., 2002.Ancient lowlands on Mars. Geophys. Res. Lett. 29 (10), doi:10.1029/2001GL013832.

Gomes, R., Levison, H.F., Tsiganis, K., Morbidelli, A., 2005. Origin of thecataclysmic late heavy bombardment period of the terrestrial planets. Na-ture 435, 466–469.

Hartmann, W.K., Neukum, G., 2001. Cratering chronology and the evolution ofMars. Space Sci. Rev. 96, 165–194.

Ivanov, B.A., 2001. Mars/Moon cratering rate ratio estimates. Space Sci.Rev. 96, 87–104.

Ivanov, B.A., 2006. Giant martian Impact Basins—Numerical modeling. LunarPlanet. Sci. XXVII. Abstract 1263.

Kamo, S.L., Reimold, W.U., Krogh, T.E., Colliston, W.P., 1996. A 2.023 Gaage for the Vredefort impact event and a first report of shock metamor-phosed zircons in pseudotachylitic breccias and Granophyre. Earth Planet.Sci. Lett. 144, 369–387.

Levison, H.F., Dones, L., Chapman, C.R., Stern, S.A., Duncan, M.J., Zahnle,K., 2001. Could the lunar “late heavy bombardment” have been triggers bythe formation of Uranus and Neptune? Icarus 151, 286–306.

Lugmair, G.W., Carlson, R.W., 1978. The Sm-Nd history of KREEP. Proc. Lu-nar Sci. Conf. 9, 689–704.

Moore, H.J., Hodges, C.A., Scott, D.H., 1974. Multiringed basins—Illustratedby Orientale and associated features. Proc. Lunar Sci. Conf. 5, 71–100.

Morbidelli, A., Petit, J.-M., Gladman, B., Chambers, J., 2001. A plausible causeof the late heavy bombardment. Meteorit. Planet. Sci. 36, 371–380.

Neukum, G., 1983. Meteoritenbombardement und Datierung planetarer Ober-flächen. Habilitation dissertation for faculty membership, Univ. of Munich,186 pp.

Neukum, G., Ivanov, B.A., Hartmann, W.K., 2001. Cratering records in theinner Solar System in relation to the lunar reference system. Space Sci.Rev. 96, 55–86.

Neumann, G.A., Zuber, M.T., Wieczorek, M.A., McGovern, P.J., Lemoine,F.G., Smith, D.E., 2004. Crustal structure of Mars from gravity and topog-raphy. J. Geophys. Res. 109, doi:10.1029/2004JE002262. E08002.

Nimmo, F., Tanaka, K., 2005. Early crustal evolution of Mars. Annu. Rev. EarthPlanet. Sci. 33, 133–161.

Nyquist, L.E., Bogard, D.D., Shih, C.-Y., Greshake, A., Stöffler, D., Eugster,O., 2001. Ages and geologic histories of martian meteorites. Space Sci.Rev. 96, 105–164.

Reese, C.C., Solomatov, V.S., Baumgardner, J.R., 2002. Survival of impact-induced thermal anomalies in the martian mantle. J. Geophys. Res. (Plan-ets) 107, 1–12.

Reese, C.C., Solomatov, V.S., 2006. Fluid dynamics of local martian magmaoceans. Icarus 184, 102–120.

Reimold, W.U., Gibson, R.L., Layer, P.W., 1996. Further 40Ar–39Ar stepheat-ing dating of fault rocks and metamorphic minerals from the VredefortDome and Witwatersrand Basin. Lunar Planet. Sci. XXVII. Abstract 1067.

Ryder, G., 1990. Lunar samples, lunar accretion and the early bombardment ofthe Moon. Eos Trans. AGU 71 (10), 313.

Smith, D.E., Sjogren, W.L., Tyler, L., Balmino, G., Lemoine, F.G., Konopliv,A.S., 1999a. The gravity field of Mars: Results from Mars Global Surveyor.Science 286, 94–97.

Smith, D.E., Zuber, M.T., Solomon, S.C., Phillips, R.J., Head, J.W., Garvin,J.B., Banerdt, W.B., Muhleman, D.O., Pettengill, G.H., Neumann, G.A.,Lemoine, F.G., Abshire, J.B., Aharonson, O., Brown, D.C., Hauck, S.A.,Ivanov, A.B., McGovern, P.J., Zwally, H.J., Duxbury, T.C., 1999b. Theglobal topography of Mars and implications for surface evolution. Sci-ence 284, 1495–1503.

Stöffler, D., Ryder, G., 2001. Stratigraphy and isotope ages of lunar geo-logic units: Chronological standard for the inner Solar System. Space Sci.Rev. 96, 9–54.

Tanaka, K.L., 1986. The stratigraphy of Mars. J. Geophys. Res. 91 (B13),E139–E158.

Tanaka, K.L., Scott, D.H., Greeley, R., 1992. Global stratigraphy. In: Kieffer,B., Jakosky, M., Snyder, C.W., Malthews, M.S. (Eds.), Mars. Univ. of Ari-zona Press, Tucson, pp. 345–382.

Tera, F., Papanastassiou, D.A., Wasserburg, G.J., 1973. A lunar cataclysm at3.95 AE and the structure of the lunar crust. Lunar Planet. Sci. IV. Abstract723.

Werner, S.C., 2005. Major aspects of the chronostratigraphy and geologic evo-lutionary history of Mars. Dissertation, Freie Universität Berlin, 252 pp.

Werner, S.C., Neukum, G., 2003. The end of the heavy bombardment as re-flected in the ages of martian impact basins. Lunar Planet. Sci. XXXIV.Abstract 1986.

Wetherill, G.W., 1975. Late heavy bombardment of the Moon and terrestrialplanets. Proc. Lunar Sci. Conf. 6, 1539–1561.

Wilhelms, D.E., 1987. The geologic history of the Moon. US Geol. Surv. Prof.Pap. 1348, 302 pp.

Zuber, M.T., Solomon, S.C., Phillips, R.J., Smith, D.E., Tyler, G.L., Aharonson,O., Balmino, G., Banerdt, W.B., Head, J.W., Johnson, C.L., Lemoine, F.G.,McGovern, P.J., Neumann, G.A., Rowlands, D.D., Zhong, S., 2000. Internalstructure and early thermal evolution of Mars from Mars Global Surveyortopography and gravity. Science 287, 1788–1793.

http://dx.doi.org/10.1029/2005GL024484

http://dx.doi.org/10.1029/2001GL013832

http://dx.doi.org/10.1029/2004JE002262

http://dx.doi.org/10.1029/2001GL013832

Documents

The early martian evolution—Constraints from basin ...rcoe/eart290C... · Icarus 195 (2008) 45–60 ... region is found at the edge of the inner, outer Solar System or in between,