We study the evolution of galaxy structure since z ~ 1 to the present. From a GOODS-S multi-band catalog we define (blue) luminosity- and mass-weighted samples, limited by M_B <= -20 and M_star >= 10^10 M_Sun, comprising 1122 and 987 galaxies, respectively. We extract early-type (E/S0/Sa) and late-type (Sb-Irr) subsamples by their position in the concentration-asymmetry plane, in which galaxies exhibit a clear bimodality. We find that the early-type fraction, f_ET, rises with cosmic time, with a corresponding decrease in the late-type fraction, f_LT, in both luminosity- and mass-selected samples. However, the evolution of the comoving number density is very different: the decrease in the total number density of M_B <= -20 galaxies since z = 1 is due to the decrease in the late-type population, which accounts for ~75% of the total star-formation rate in the range under study, while the increase in the total number density of M_star >= 10^10 M_Sun galaxies in the same redshift range is due to the evolution of early types. This suggests that we need a structural transformation between late-type galaxies that form stars actively and early-type galaxies in which the stellar mass is located. Comparing the observed evolution with the gas-rich major merger rate in GOODS-S, we infer that only ~20% of the new early-type galaxies with M_star >= 10^10 M_Sun appeared since z ~ 1 can be explained by this kind of mergers, suggesting that minor mergers and secular processes may be the driving mechanisms of the structural evolution of intermediate-mass (M_star ~ 4x10^10 M_Sun) galaxies since z ~ 1.

Aims: We study the major merger fraction in a SPITZER/IRAC-selected catalogue in the GOODS-S field up to z ~ 1 for luminosity- and mass-limited samples. Methods: We select disc-disc merger remnants on the basis of morphological asymmetries, and address three main sources of systematic errors: (i) we explicitly apply morphological K-corrections, (ii) we measure asymmetries in galaxies artificially redshifted to z_d = 1.0 to deal with loss of morphological information with redshift, and (iii) we take into account the observational errors in z and A, which tend to overestimate the merger fraction, though use of maximum likelihood techniques. Results: We obtain morphological merger fractions (f_m) below 0.06 up to z ~ 1. Parameterizing the merger fraction evolution with redshift as f_m(z) = f_m(0) (1+z)^m, we find that m = 1.8 +/- 0.5 for M_B <= -20 galaxies, while m = 5.4 +/- 0.4 for M_star >= 10^10 M_Sun galaxies. When we translate our merger fractions to merger rates (R_m), their evolution, parameterized as R_m(z) = R_m(0) (1+z)^n, is quite similar in both cases: n = 3.3 +/- 0.8 and n = 3.5 +/- 0.4, respectively. Conclusions: Our results imply that only ~8% of today's M_star >= 10^10 M_Sun galaxies have undergone a disc-disc major merger since z ~ 1. In addition, ~21% of this mass galaxies at z ~ 1 have undergone one of these mergers since z ~ 1.5. This suggests that disc-disc major mergers are not the dominant process in the evolution of M_star >= 10^10 M_Sun galaxies since z ~ 1, but may be an important process at z > 1.

(Abridged) We measure the fraction of galaxies undergoing disk-disk major mergers (f_m) at intermediate redshifts (0.35 <= z < 0.85) by studying the asymmetry index A of galaxy images. Results are provided for B- and Ks-band absolute magnitude selected samples from the Groth strip in the GOYA photometric survey. Three sources of systematic error are carefully addressed: (i) we avoid morphological K-corrections, (ii) we measure asymmetries in artificially redshifted to z_d = 0.75 galaxies to lead with loss of morphological information with redshift, and (iii) we take into account the observational errors in z and A, that tend to overestimate the merger fraction, by maximum likelihood techniques. We find: (i) our data allow for a robust merger fraction to be provided for a single redshift bin centered at z=0.6. (ii) Merger fractions have low values: f_m = 0.045 for M_B <= -20 galaxies, and f_m = 0.031 for M_Ks <= -23.5 galaxies. And, (iii) failure to address the effects of the observational errors leads to overestimating f_m by factors of 10%-60%. Combining our results with those on literature, and parameterizing the merger fraction evolution as f_m(z) = f_m(0)(1+z)^m, we obtain that m = 2.9 +- 0.8, and f_m(0) = 0.012 +- 0.004$. Assuming a Ks-band mass-to-light ratio not varying with luminosity, we infer that the merger rate of galaxies with stellar mass M >= 3.5x10^10 M_Sun is R_m = 1.6x10^-4 Mpc^-3 Gyr^-1. When we compare with previous studies at similar redshifts, we find that the merger rate decreases when mass increases.

The determination of galaxy merger fraction of field galaxies using automatic morphological indices and photometric redshifts is affected by several biases if observational errors are not properly treated. Here, we correct these biases using maximum likelihood techniques. The method takes into account the observational errors to statistically recover the real shape of the bidimensional distribution of galaxies in redshift - asymmetry space, needed to infer the redshift evolution of galaxy merger fraction. We test the method with synthetic catalogs and show its applicability limits. The accuracy of the method depends on catalog characteristics such as the number of sources or the experimental error sizes. We show that the maximum likelihood method recovers the real distribution of galaxies in redshift and asymmetry space even when binning is such that bin sizes approach the size of the observational errors. We provide a step-by-step guide to applying maximum likelihood techniques to recover any one- or bidimensional distribution subject to observational errors.