U/Th dating of carbonate deposits from Chafarinas Islands, Spain

U/Th dating of carbonate deposits from Chafarinas Islands, Spain A. Martı ´nez-Aguirre, J. M. Alcaraz-Pelegrina Abstract The Chafarinas Islands are a group of volcanic outcrops off the north coast of the African continent. Quaternary deposits and calcrete formations cover most of the islands’ planed surface, mainly on Rey Francisco I Island. Uranium-series ages are presented for a geological section with 18 m of detrital deposits and calcrete layers. Calcretes consist of impure carbonate; the leach/leach (L/L) method was therefore used to obtain 234 U/ 238 U and 230 Th/ 234 U activity ratios in the pure authigenic carbonate fraction to obtain ages of the samples. Resulting ages range from 14 to 80 ka. Except for one sample at the bottom of the age sequence, with the youngest age, all others have ages that agree with their stratigraphic location. Ages obtained using the L/L method were also compared with the method of Allegre and Condomines (Earth Planet Sci Lett 28:395–406, 1976) and with a more recent method (Ludwig and Paces Geochim Cosmochim Acta 66:487–506, 2002, Sharp et al. Quat Res 59:139–150, 2003, Haase-Schramm et al. Geochim Cosmochim Acta 68:985–1005, 2004). The latter two methods were also used to estimate isotopic composition of the detrital component in the impure carbonate and model a more realistic age for the anomalous sample. Minimal uranium isotope activity ratio differences imply that environmental conditions between 25 and 60 ka remained relatively uniform and that a major change may have occurred *80 ka ago. Keywords U-series dating Dirty carbonates  Quaternary deposits Karst landform Introduction Isotopes from the 238 U decay chain have been widely used to evaluate the time when certain geological systems were initially formed and to study their evolution over time. In particular, isotopic disequilibrium between 230 Th and 234 U has been used to date carbonate precipitated from natural water. The solubility of U combined with the insolubility of Th in natural waters permit U to be incorporated in precipitates from aqueous solutions, whereas initial Th is not present in pure authigenic cements. Any 230 Th activity found later in such precipitates is thus derived from the radioactive decay of 238 U and 234 U. The systematic increase in 230 Th toward secular equilibrium with U isotopes accurately reflects the age of the sample, if the system has remained close to isotope exchange since its formation. However, many carbonates formed in nearsurface environments, particularly those precipitated from continental natural waters, are impure and contain variable amounts of detrital material, consisting mainly of clays. When impure carbonate is dissolved for U and Th analysis, the detrital component contributes significant and usually unknown quantities of isotopes from the 238 U and 232 Th decay chains. It is thus necessary to apply a correction method to evaluate isotope concentrations in the pure carbonate fraction of the impure carbonate, in order to obtain an accurate age for the samples. Correction for the presence of initial 230 Th, 234 U and 238 U from the detrital component is usually made by using the 232 Th concentration as an index of detritus content. Direct measurement of the U–Th isotopic composition of the A. Martı ´nez-Aguirre (&) Departamento de Fı ´sica Aplicada I, ETSIA, Universidad de Sevilla, Ctra. Utrera, km 1, 41013 Sevilla, Spain e-mail: [email protected] J. M. Alcaraz-Pelegrina Departamento de Fı ´sica, Facultad de Ciencias, Universidad de Co ´rdoba, Campus de Rabanales. Ctra. de Madrid N-IV-a, km 396, 14071 Co ´rdoba, Spain detrital component may be difficult if the detrital particles are too fine to separate mechanically or if there is substantial variation in the U–Th isotopic composition of the detrital component. At the time of precipitation, the pure authigenic carbonate component contains negligible amounts of 232 Th and 230 Th. This means that any 232 Th found in a leachate from the carbonate must come from the detrital part of the impure carbonate. If the detrital component has a single, uniform U–Th isotopic composition at the time of precipitation of the impure carbonate (i.e. constant U/Th and 230 Th/ 232 Th activity ratio), mathematical techniques that assume simple two-component mixing between a 232 Th-free component (authigenic cement) and a 232 Th-bearing component (detritus) are often employed [9,14,5,8]. Allegre and Condomines [4] developed a method for dating volcanic rocks. However, Bischoff and Fitzpatrick [5] proposed the use of their method for dating impure carbonate, as the principles of radioactive decay are the same for impure carbonates and for mineral solidifying from a melt. They [4] showed that if U isotopes are in equilibrium and the initial 230 Th/ 232 Th activity ratio in all fractions leached from an impure carbonate is equal to that in the detrital material, the time evolution of this activity ratio is given by the equation: 230Th=232Th  L¼230Th=232Th  Do ekot þ238U=232Th  L1ekot  ;ð1Þ where L means the activity ratio in the leachate, D o the activity ratio in the detritus at time of deposition and k o the decay constant for 230 Th. At t=0, the time of precipitation, U and Th measurement of several leachates from the same homogenized impure carbonate result in a horizontal line when plotting 230 Th/ 232 Th versus 238 U/ 232 Th activity ratios. After time t, the 230 Th activity concentration grows, due to the decay of U isotopes in the pure carbonate; thus in the same plot the activity ratios in the several leachates will fall on a line with a slope 1 ekot. As time increases to values above 1/k o , for each aliquot 230 Th/ 232 Th = 238 U/ 232 Th and the data points will fall along the so called equiline: having a unit slope and passing through the origin. Isochrons evolving at different values of t rotate around a point called the equipoint which defines the 230 Th/ 232 Th activity ratio in the detritus at the time of precipitation. If U isotopes are in disequilibrium, the 230 Th/ 232 Th activity ratios in leachates are described by the equation: 230Th=232Th  L¼230Th=232Th  Do ekot þ234U=232Th  L1ekot  :ð2Þ A plot of 230 Th/ 232 Th versus 234 U/ 232 Th ratios in several leachates will allow estimation of the 230 Th/ 234 U activity ratio of the pure carbonate sample as well as the 230 Th/ 232 Th ratio in the detrital material at the time of analysis. Even considering that the 230 Th/ 232 Th activity ratio in the detritus is probably different from the activity ratio in the detritus part leached from the impure carbonate, the above equation can also be used, though the 230Th=232Th  Do activity ratio must be changed by a factor r 0 /r 2 , with r 0 and r 2 being the fractions of 230 Th and 232 Th respectively extracted from the detritus into the solution during the leaching process. Hence: r0=r2230Th=232Th  Do instead of 230Th=232Th  Do : An isochron approach involving total dissolution of several presumably cogenetic samples is often used [5,13], considering that the detritus is different in each cogenetic sample. However, the fundamental requirement of a cogenetic sample suite is often difficult or impossible to verify before analysis of the total sample. Other correction methods involving isochrons arising from the analysis of several leachings of different aliquots from the same impure carbonate have been also used [17]. They proposed that an isochron can be constructed from leachates of the various aliquots from a homogenized sample (the L/L method) even though different fractions are extracted from the detrital material for each isotope. They showed that the leachates will obey the linear relation: 230Th=232Th  L¼230Th=234U  C 234U=232Th  L þr0=r2230Th=232Th  Dr4=r2230Th=234U  C 234U=232Th  D ð3aÞ 234U=232Th  L¼234U=238U  C 238U=232Th  L þr4=r2234U=232Th  Dr8=r2234U=238U  C 238U=232Th  D ð3bÞ r 0 ,r 2 ,r 8 and r 4 being the fractions of 230 Th, 232 Th, 238 U and 234 U isotopes leached from the detritus into the solution. If relative fractions r i /r j are constant for coeval samples [3]a plot of 230 Th/ 232 Th versus 234 U/ 232 Th activity ratios in leachates will define a line whose slope is the 230 Th/ 234 U ratio of the pure carbonate. Furthermore, a plot of 234 U/ 232 Th versus 238 U/ 232 Th in leachates will yield a slope of 234 U/ 238 U in the pure carbonate. Using the Bateman disequilibrium equation, the age of the pure carbonate sample is obtained. Over the last several years the analysis of several leachates has become unusual in the dating of impure carbonates, mainly due to differential fractionation in the leachates and to the chemical complexity of the analysis. If differential fractionation occurs, activity ratio values from different leachates would not define a line and activity ratios in the pure carbonate will therefore not be obtained. The L/L method equations can nevertheless also be used in total dissolution of different aliquots of cogenetic samples. These samples must have different activity ratios in the detritus, which is difficult to ascertain prior to the analysis of the sample. More recently, several approaches have been developed to date carbonates regardless of the amount of respective detrital material content [11,18,6]. Those carbonate samples with sufficiently high ratios of authigenic 230 Th to allogenic 230 Th present within the detrital component require small corrections for the detrital U and Th isotopes. So they assume that even approximate assumptions about the U–Th isotopic composition of the detritus enable useful precision in their calculated 230 Th/U age. They considered that analysis of a single complete sample is equivalent to determine a two-point isochron age, except that the second point on the isochron is defined by estimated isotope ratios with realistic errors assigned to the estimate ( 234 U/ 238 U= 230 Th/ 238 U=1.0 ±0.1 and 232 Th/ 238 U= 1.21 ±0.60 in [11,18] and 232 Th/ 238 U=0.271 ±0.027 in [6]). Although the latter correction method was supposedly developed for samples with high 230 Th/ 232 Th activity ratios, i.e. nearly pure carbonates, if the assumed activity ratios are those in the real detritus it may be used even in very dirty carbonates. Samples from the Chafarinas Islands present high detrital material content. Thus, in this case an estimate of activity ratios in the detritus would probably (except in the case where the estimate is actually the real activity ratios in the detritus) give wrong ages and very different corrected ages for every subsample from the same impure carbonate. However, assuming that the pure detritus activity ratios must lie on the same isochron line as the several leachates from an impure carbonate [6,3], Osmond diagrams have been used to estimate activity ratios in detrital material from the Chafarinas Islands samples. The purpose of this study was to evaluate the above different correction methods for U–Th dating on carbonate deposits from the Chafarinas Islands. We will also use these correction methods to obtain and discuss some interesting activity ratios in the detritus of such impure carbonates, particularly 230 Th/ 232 Th activity ratio. Samples and radiochemical methods The Chafarinas Islands are located off the north coast of Morocco about 50 km east of Melilla and 4 km north of Cape Agua (35.1826°N latitude, -2.4240°longitude; Fig. 1). The small islands are formed from volcanic rocks from the Upper Miocene to Pliocene ages. Quaternary surface deposits including calcrete cover most of the surface area, especially on Rey Francisco I, the easternmost island. A geological description of the islands can be found in Rodrı ´guez-Vidal et al. [15]. Five samples of calcrete were collected from an 18-m-thick section of surface deposits constituting the most complete stratigraphic sequence in the northern part of the island. Sample CHR-9 was taken from the base of the sequence, sample CHR-8 from its middle and samples CHR-27, CHR-28 and CHR-29 from the uppermost calcrete layer, which appears to have lateral evolution (NESW) with older samples situated to the NE. All samples of calcrete contain substantial amounts of detrital material, greater than 20 % by weight, with the highest percentages in samples CHR-8 and CHR-9, of around 40 %. Bulk samples were crushed and homogenized and then separated in several aliquots [2]. Aliquots from the same bulk sample would therefore share the same detritus and have the same activity ratios for the detrital component. The aliquots were dissolved in different nitric acid concentrations (2–8 M) in order to totally dissolve the carbonate and to extract different isotopic concentrations from the detrital material. These leachates were separated as quickly as possible from the residue and stored for U and Th analysis. Fig. 1 Map of the Chafarinas Islands off the north coast of Africa. Locations of samples collected at Rey Francisco I Island are also shown. Circled numbers from 0 to 14 are the stratigraphic levels at Rey Francisco I Island, as described by Rodrı ´guez-Vidal et al. [15] The residue remaining from one aliquot was also analyzed for U and Th isotopes. It was digested alternatively in concentrated nitric acid and aqua regia and in each step the solution was taken to dryness at high temperature (it can last from 7 to 10 days). The very small final white or clear grey residue was separated by centrifuging and filtration and discarded and the solution was added to the initial leachate for U and Th analysis. These samples were considered as totally dissolved samples. Additional residue samples were also analyzed in the same way. Iron carrier (FeCl 3 ) and known amounts of highly purified 232 U and 229 Th isotope tracers, for yield determinations, were added to the acid solutions prepared from weak or strong acid leaching. Iron hydroxides, along with dissolved U and Th, were precipitated and then separated from the solutions. A solvent extraction method was used to separate U from Th isotopes by first dissolving the iron precipitate in 8 M HNO 3 , and then adding tributylphosphate (TBP) to back-extract both uranium and thorium from solution into the TBP phase. Once the TBP phase was separated from the aqueous phase, 20 ml of Xilene was mixed with the TBP phase. Th was extracted from the organic phase using 1.5 M HCl as the inorganic phase. Finally, uranium was extracted from the organic phase Table 1 U and Th isotopes activity concentrations (mBq/g) and some activity ratios of samples colleted at Rey Francisco I Island (Chafarinas Islands) Sample 238 U 234 U 230 Th 232 Th 234 U/ 238 U 230 Th/ 234 U 230 Th/ 232 Th T (ky) CHR8: level 5 CHR8-1 11.7 ±0.6 16.2 ±0.8 14.0 ±1.2 13.9 ±1.2 1.385 ±0.028 0.861 ±0.084 1.105 ±0.020 182 ±43 CHR8-2 7.62 ±0.30 11.7 ±0.5 8.73 ±0.71 5.69 ±0.47 1.533 ±0.020 0.747 ±0.067 1.533 ±0.032 133 ±22 CHR8-R 28.6 ±1.0 28.4 ±1.0 40.3 ±1.5 57.1 ±2.0 0.990 ±0.032 1.421 ±0.070 0.705 ±0.018 CHR9: level 1 CHR9-1 5.02 ±0.34 7.12 ±0.47 9.81 ±0.87 11.3 ±1.0 1.420 ±0.051 1.378 ±0.153 0.867 ±0.015 [350 CHR9-2 5.39 ±0.15 9.76 ±0.24 4.17 ±0.14 4.04 ±0.14 1.809 ±0.050 0.427 ±0.018 1.032 ±0.035 57.8 ±3.1 CHR9-3 5.33 ±0.19 9.97 ±0.30 4.47 ±0.16 3.87 ±0.14 1.872 ±0.066 0.448 ±0.021 1.156 ±0.042 61.3 ±3.6 CHR9-4 5.27 ±0.17 9.87 ±0.26 4.52 ±0.16 4.08 ±0.15 1.874 ±0.058 0.458 ±0.020 1.107 ±0.038 63.1 ±3.6 CHR9-5 2.65 ±0.09 4.47 ±0.13 3.63 ±0.14 4.02 ±0.15 1.688 ±0.046 0.812 ±0.038 0.903 ±0.026 152.4 ±13.9 CHR9-R 7.28 ±0.39 7.92 ±0.41 24.8 ±1.1 30.0 ±1.2 1.088 ±0.071 3.136 ±0.211 0.828 ±0.031 CHR9-T 5.08 ±0.21 6.28 ±0.22 14.7 ±0.6 17.6 ±0.6 1.236 ±0.067 2.344 ±0.134 0.835 ±0.044 [350 CHR27: level 14, NE CHR27-1 4.16 ±0.10 5.43 ±0.12 3.80 ±0.11 3.30 ±0.10 1.304 ±0.029 0.700 ±0.026 1.152 ±0.032 122 ±8 CHR27-2 4.36 ±0.13 5.68 ±0.16 3.98 ±0.11 3.84 ±0.11 1.303 ±0.037 0.699 ±0.028 1.036 ±0.026 122 ±8 CHR27-3 4.42 ±0.10 5.46 ±0.11 3.71 ±0.15 3.94 ±0.15 1.237 ±0.025 0.730 ±0.024 1.034 ±0.022 133 ±8 CHR27-4 4.55 ±0.09 5.85 ±0.11 4.39 ±0.17 3.82 ±0.15 1.286 ±0.022 0.745 ±0.024 1.104 ±0.023 137 ±8 CHR27-5 4.58 ±0.14 5.89 ±0.17 4.34 ±0.13 3.46 ±0.11 1.285 ±0.039 0.736 ±0.030 1.251 ±0.035 134 ±10 CHR27-6 4.44 ±0.11 5.70 ±0.13 4.30 ±0.13 3.57 ±0.11 1.284 ±0.029 0.755 ±0.028 1.204 ±0.033 141 ±10 CHR27-R 12.9 ±0.4 12.2 ±0.4 20.5 ±0.6 34.4 ±1.0 0.943 ±0.031 1.682 ±0.074 0.596 ±0.013 CHR27-T 5.34 ±0.11 6.34 ±0.12 6.05 ±0.12 7.50 ±0.17 1.187 ±0.033 0.954 ±0.026 0.807 ±0.024 265 ±28 CHR28: level 14 Intermediate CHR28-1 5.56 ±0.13 7.11 ±0.16 3.95 ±0.13 5.02 ±0.15 1.278 ±0.026 0.555 ±0.022 0.787 ±0.021 84.8 ±4.8 CHR28-2 4.92 ±0.11 6.01 ±0.13 3.73 ±0.12 4.88 ±0.15 1.221 ±0.022 0.620 ±0.024 0.764 ±0.020 101 ±6 CHR28-3 5.53 ±0.13 7.01 ±0.16 4.53 ±0.14 6.05 ±0.18 1.268 ±0.025 0.647 ±0.025 0.749 ±0.018 107 ±7 CHR28-R 11.6 ±0.5 11.1 ±0.5 13.9 ±0.5 25.8 ±0.8 0.955 ±0.045 1.255 ±0.068 0.540 ±0.014 CHR28-T 7.07 ±0.16 8.11 ±0.17 6.44 ±0.16 10.2 ±0.2 1.147 ±0.035 0.794 ±0.026 0.632 ±0.021 161.6 ±11.7 CHR29: level 14 SW CHR29-1 4.06 ±0.10 5.24 ±0.12 3.05 ±0.10 4.30 ±0.14 1.290 ±0.027 0.582 ±0.024 0.709 ±0.020 90.7 ±5.6 CHR29-2 3.09 ±0.08 3.94 ±0.10 2.71 ±0.09 4.26 ±0.13 1.277 ±0.032 0.686 ±0.028 0.635 ±0.017 119 ±8 CHR29-3 4.07 ±0.09 5.23 ±0.12 3.19 ±0.10 4.60 ±0.14 1.284 ±0.023 0.610 ±0.023 0.693 ±0.016 95.3 ±7.5 CHR29-4 4.62 ±0.13 6.09 ±0.16 3.41 ±0.11 5.09 ±0.15 1.316 ±0.036 0.560 ±0.023 0.670 ±0.019 85.6 ±5.1 CHR29-R 11.2 ±0.5 11.3 ±0.5 14.2 ±0.5 26.7 ±0.8 1.012 ±0.046 1.253 ±0.065 0.532 ±0.013 The sample number followed by L means a leachate, R a residue of a leachate, while T means total samples, obtained by the addition of dissolved residue to the leachate. In column 9 nominal ages obtained by directly using the Bateman disequilibrium equation are also shown. Errors are 1 -r using distilled water. The uranium fraction is ready for electroplating after solvent extraction, while the thorium fraction needs further purification as some traces of uranium are also extracted with the thorium. Thorium purification used anion exchange resin (AG1 98 hydrochloric form) conditioned in HCl 8 M. A second Th purification was necessary using the same resin conditioned in HNO 3 7M. Purified U and Th solutions were transferred onto stainless steel disks after adding 1 ml of 0.3 M NaSO 4 and electroplating at 1.2 A for 1 h. Radioactivity of 229 Th, 230 Th, 232 Th, 232 U, 234 U and 238 U were measured by decay counting in an alpha spectrometer system equipped with PIPS detectors [19]. Moderate-to-high recovery yields and well resolved spectra provide analytically reliable results. Results and discussion Table 1shows the U and Th isotope activity concentrations as well as several activity ratios in each leachate (L), in the analyzed residue (R) and in the total dissolved sample (T) obtained by adding the dissolved residue to the leachate. The data’s 1 -rerror is also included. Uncorrected ages calculated from measured 230 Th/ 234 U and 234 U/ 238 U activity ratios using the U-series disequilibrium age equation (Appendix A of [7]) are given for each sample separately. The 238 U activity concentrations range from 2.65 to 11.7 mBq/g in leachates, with the highest concentration in sample CHR-8, whereas most of the others are in the 4–5 mBq/g range. Residues tend to have higher 238 U activity concentrations than leachates from the same bulk (a) 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 CHR-8 equiline 234U/232Th 238U/232 Th (b) 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 0.6 0.8 1.0 1.2 1.4 1.6 1.8 CHR-8 equiline (t>350 ky) t = 0 230Th/232Th 234U/232 Th (c) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 CHR-8 232Th/238U)D=1.97 234U/238U 232Th/238U (d) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 0.9 1.0 1.1 1.2 1.3 1.4 1.5 CHR-8 230Th/238U)D= 1.40 232Th/238U)D=1.97 230 Th/238U 232Th/238U Fig. 2 Results for sample CHR-8. a 234 U/ 232 Th versus 238 U/ 232 Th data points. The equiline at which both uranium isotopes are in equilibrium is also shown. b 230 Th/ 232 Th versus 234 U/ 232 Th data points. The equiline or line at which 230 Th is in equilibrium with 234 U and the line at which 230 Th/ 232 Th remains constant and the same as its value at the time of precipitation are also shown. c,dOsmond diagrams for CHR-8 sample. The assumed [6] isotope ratios in the detrital ( 234 U/ 238 U= 230 Th/ 238 U=1±0.1 and 232 Th/ 238 U= 1.21 ±0.60) have been included as an example. Assuming secular equilibrium between both uranium isotopes in the detritus, the lines were used to obtain the 232 Th/ 238 U(c) and 230 Th/ 238 U(d) activity ratios in the detritus sample, with the highest concentration again from sample CHR-8. This difference between leachates and residues is more evident in the case of Th isotopes. Leachates have 230 Th and 232 Th activity concentrations that are lower than in the residues, below unity, show values below those in the leachates. As expected, 234 U/ 238 U activity ratios in residues are lower than those in the leachates. 234 U/ 238 U activity ratios from the leachate represents preferentially dissolved carbonates which might have 234 U/ 238 U[1as inherited from the water they precipitated from. For leachates, measured uranium isotope activity ratios range 1.1–1.3 for samples CHR-27, CHR-28 and CHR-29, and higher for samples CHR-8, 1.4–1.5, and CHR-9, 1.5–1.8. As expected, residue samples have 234 U/ 238 U activity ratios either close to the secular equilibrium value of 1.0. For the 230 Th/ 232 Th activity ratios, leachates have values ranging from 0.7 to 1.5 whereas residues have lower values, consistent with the higher Th/U values present in the detrital component compared to pure carbonate. (a) 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 CHR-9 equiline 234U/232Th 238U/232Th (b) 0.0 0.5 1.0 1.5 2.0 2.5 0.8 0.9 1.0 1.1 1.2 CHR-9 t = 0 equiline (t > 350 ka) 230Th/232Th 234U/232Th (c) 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 1.0 1.2 1.4 1.6 1.8 2.0 CHR-9 232Th/238U)D = 4.44 234U/238U 232Th/238U (d) 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 CHR-9 230Th/238U)D= 3.64 232Th/238U)D= 4.44 230Th/238U 232Th/238U Fig. 3 Results for sample CHR-9. a 234 U/ 232 Th versus 238 U/ 232 Th data points. The equiline at which both uranium isotopes are in equilibrium is also shown. b 230 Th/ 232 Th versus 234 U/ 232 Th data points. The equiline or line at which 230 Th is in equilibrium with 234 U and the line at which 230 Th/ 232 Th remains constant and the same as its value at the time of precipitation are also shown. c,dOsmond diagrams for CHR-9 sample. The assumed [6] isotope ratios in the detrital ( 234 U/ 238 U= 230 Th/ 238 U=1±0.1 and 232 Th/ 238 U= 1.21 ±0.60) have been included as an example. Assuming secular equilibrium between both uranium isotopes in the detritus, the lines were used to obtain the 232 Th/ 238 U(c) and 230 Th/ 238 U(d) activity ratios in the detritus or similar to those for U, however residues have Th isotope activity concentrations well above those for U in the same samples. Again the highest Th activity concentrations are found in sample CHR-8. A high percentage of detrital materials in these samples as well as higher Th concentration in the residues gives high 230 Th/ 234 U activity ratios, which results in overestimation of ages even though uncorrected ages for different leachates of the same sample are similar. These overestimated ages are corroborated by the values of the 230 Th/ 232 Th activity ratios in every aliquot (see Table 1). The results of the 230 Th/ 232 Th activity ratio For illustration purposes only, 2D Rosholt-type diagrams for each sample are shown (a, b) in Figs. 2,3,4,5 and 6. Activity ratios in the 232 Th-free, pure carbonate end member were calculated using a 3D-isochron fit with Isoplot [10,12], along with the age, initial 234 U/ 238 U activity ratio and regression statistics (MSWD and probability of fit; Table 2). Generally, the ages range from 14.1 ±1.5 ka for sample CHR-9 to 76 ±11 ka for sample CHR-8. As commented above, samples CHR-27, CHR-28 and CHR-29 were taken at the top of the series in a calcrete layer that seems to have lateral evolution with older samples situated at NE positions [15]. The ages obtained support this model because sample CHR-29 was situated in the SW position and it has the youngest age, 25.3 ±2.9 ka (agreeing with the other 14 C date of 20.3 ±0.3 ka), whereas CHR-27 was situated in the NE position and has the oldest age, 62.9 ±2.9 ka, with CHR-28 situated in an intermediate position and having an intermediate age, 33.4 ±2.6 ka. The 234 U/ 238 U activity ratio at the time of deposition in this time interval is constant (1.538 ±0.038 as the mean value). This could indicate that during this time period (25–63 ka) climate conditions did not change and that the samples were precipitated in the same conditions and probably from similar waters. Sample CHR-8 was situated some meters below the surface and as expected is older than samples from the top layer, at 76 ±11 ka. Sample CHR-9 was taken at the bottom of the stratigraphic section and was expected to be the oldest sample. However, the age obtained was only 14.1 ±1.5 ka, which is younger rather than older than the overlying horizons. The 234 U/ 238 U initial activity ratios in both samples (CHR-8 and CHR-9) are higher than those of samples CHR-27 to (a) 0.4 0.6 0.8 1.0 1.2 1.4 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 CHR-27 equiline 234U/232Th 238U/232Th (b) 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 CHR-27 equiline (t>350 ky) t = 0 230Th/232Th 234U/232Th (c) 0.5 1.0 1.5 2.0 2.5 3.0 0.90 0.95 1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35 CHR-27 232Th/238U)D= 2.34 234U/238U 232Th/238U (d) 0.5 1.0 1.5 2.0 2.5 3.0 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 CHR-27 230Th/238U)D=1.47 232Th/238U)D=2.34 230Th/238U 232Th/238U Fig. 4 Results for sample CHR-27. a 234 U/ 232 Th versus 238 U/ 232 Th data points. The equiline at which both uranium isotopes are in equilibrium is also shown. b 230 Th/ 232 Th versus 234 U/ 232 Th data points. The equiline or line at which 230 Th is in equilibrium with 234 U and the line at which 230 Th/ 232 Th remains constant and the same as its value at the time of precipitation are also shown. c,dOsmond diagrams for CHR-27 sample. The assumed [6] isotope ratios in the detrital ( 234 U/ 238 U= 230 Th/ 238 U=1±0.1 and 232 Th/ 238 U= 1.21 ±0.60) have been included as an example. Assuming secular equilibrium between both uranium isotopes in the detritus, the lines were used to obtain the 232 Th/ 238 U(c) and 230 Th/ 238 U(d) activity ratios in the detritus Eq. 2and Rosholt diagrams (Figs. 2b, 3b, 4b, 5b, 6b) it is possible to obtain the 230 Th/ 232 Th activity ratios at present and at deposition time in the detritus. The results are shown in columns 2 and 3 of Table 3. Column 4 of this Table also shows the ages obtained by using the time evolution equation, considering that the 232 Th concentration remains constant during the period of time considered. The age errors were evaluated by error propagation. Even considering that the ages have a relatively high error, the results are similar to those obtained using the equations of the L/L method [17]. More interesting are the results obtained for the present day 230 Th/ 232 Th activity ratio in the detritus (see column 2 of Table 3). Except for sample CHR-9, all the others have similar values for this activity ratio with a mean (a) 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 0.4 0.6 0.8 1.0 1.2 1.4 CHR-28 equiline 234U/232Th 238U/232Th (b) 0.4 0.6 0.8 1.0 1.2 1.4 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 equiline (t>350 ky) t = 0 CHR-28 230Th/232Th 234U/232Th (c) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 0.85 0.90 0.95 1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35 CHR-28 232Th/238U)D=2.05 234U/238U 232Th/238U (d) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 0.7 0.8 0.9 1.0 1.1 1.2 1.3 CHR-28 230Th/238U)D= 1.135 232Th/238U)D= 2.05 230Th/238U 232Th/238U diagrams for CHR-28 sample. The assumed [6] isotope ratios in the detrital ( 234 U/ 238 U= 230 Th/ 238 U=1±0.1 and 232 Th/ 238 U= 1.21 ±0.60) have been included as an example. Assuming secular equilibrium between both uranium isotopes in the detritus, the lines were used to obtain the 232 Th/ 238 U(c) and 230 Th/ 238 U(d) activity ratios in the detritus Fig. 5 Results for sample CHR-28. a 234 U/ 232 Th versus 238 U/ 232 Th data points. The equiline at which both uranium isotopes are in equilibrium is also shown. b 230 Th/ 232 Th versus 234 U/ 232 Th data points. The equiline or line at which 230 Th is in equilibrium with 234 U and the line at which 230 Th/ 232 Th remains constant and the same as its value at the time of precipitation are also shown. c, d Osmond CHR-29 and quite similar to each other, with 2.056 ± 0.006 as the mean value. This fact may indicate that the precipitation conditions of both samples were different from the uppermost calcrete layer samples. This could eventually be associated to the rapid and sudden warm-humid event 5.1 at 79–80 ka that affected the northern part of Africa [1, 16]. Nevertheless, sample CHR9 at least seems to have behaved as an open system for U and/or Th isotopes since its formation, as discussed below. As can be observed in Rosholt diagrams (Figs. 2, 3, 4, 5, 6a, b), both residue and total samples follow the same isochron line as the leachates and they have thus been included in the age calculations. Following Allegre and Condomines [4] as 234 U and 238 U are clearly in disequilibrium, by using value of 0.429 ±0.017. Sample CHR-9 has a clearly higher value of the same activity ratio. It seems that either the original detrital material in this sample is different from that in the other samples or that an increase of 230 Th concentration in the detritus has been occurring since sample formation. Indeed, if the present-day value in the detritus is taken as the mean value in the other samples (0.429 ±0.017) and the initial value as that in Table 3 (0.890 ±0.043), then the time t since its formation results as being 79.2 ±6.8 ka, more in agreement with its expected age A different analytical approach for dating calcrete was used by Ludwig and Paces [11] and Sharp et al. [18] and for nearly pure samples of lacustrine aragonite by HaaseSchramm et al. [6]. These authors targeted small samples of relatively pure carbonate with high U/Th ratios that require only small corrections for initial 230 Th from a detrital component. They assumed that in those samples with sufficiently high ratios of authigenic 230 Th to allogenic 230 Th present within the carbonate, a small correction for the detrital U and Th isotopes is required. They assumed that the analysis of a single sample is equivalent to determining a two-point isochron age, except that the second point on the isochron is defined by a detrital component assumed to be in radioactive secular equilibrium (i.e., 234 U/ 238 U= 230 Th/ 238 U=1.0) with 232 Th/ 238 U activity ratio of *1.21 ±0.60 for silicate material [11,18] or 0.271 ±0.027 for detritus that included substantial (a) 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.4 0.6 0.8 1.0 1.2 equiline CHR-29 234U/232Th 238U/232Th (b) 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 0.50 0.55 0.60 0.65 0.70 0.75 R CHR-29 equiline (t>350 ky) t = 0 230Th/232Th 234U/232Th (c) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 0.9 1.0 1.1 1.2 1.3 1.4 CHR-29 232Th/238U)D=2.55 234U/238U 232Th/238U (d) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 CHR-29 230Th/238U)D=1.33 232Th/238U)D=2.55 230Th/238U 232Th/238U Fig. 6 Results for sample CHR-29. a 234 U/ 232 Th versus 238 U/ 232 Th data points. The equiline at which both uranium isotopes are in equilibrium is also shown. b 230 Th/ 232 Th versus 234 U/ 232 Th data points. The equiline or line at which 230 Th is in equilibrium with 234 U and the line at which 230 Th/ 232 Th remains constant and the same as its value at the time of precipitation are also shown. c,dOsmond diagrams for CHR-29 sample. The assumed [6] isotope ratios in the detrital ( 234 U/ 238 U= 230 Th/ 238 U=1±0.1 and 232 Th/ 238 U= 1.21 ±0.60) have been included as an example. Assuming secular equilibrium between both uranium isotopes in the detritus, the lines were used to obtain the 232 Th/ 238 U(c) and 230 Th/ 238 U(d) activity ratios in the detritus