اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدLunar impact craters identification and age estimation with ChangE data by deep and transfer learning
373
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Impact craters, as lunar fossils, are the most dominant lunar surface features and occupy most of the Moons surface. Their formation and evolution record the history of the Solar System. Sixty years of triumphs in the lunar exploration projects accumulated a large amount of lunar data. Currently, there are 9137 existing recognized craters. However, only 1675 of them have been determined age, which is obviously not satisfactory to reveal the evolution of the Moon. Identifying craters is a challenging task due to their enormous difference in size, large variations in shape and vast presence. Furthermore, estimating the age of craters is extraordinarily difficult due to their complex and different morphologies. Here, in order to effectively identify craters and estimate their age, we convert the crater identification problem into a target detection task and crater age estimation into a taxonomy structure. From an initial small number of available craters, we progressively identify craters and estimate their age from ChangE data by transfer learning (TL) using deep neural networks. For comprehensive identification of multi-scale craters, a two-stage craters detection approach is developed. Thus 117240 unrecognized lunar craters that range in diameter from 532 km to 1 km are identified. Then, a two-stage classification approach is developed to estimate the age of craters by simultaneously extracting their morphological features and stratigraphic information. The age of 79243 craters larger than 3 km in diameter is estimated. These identified and aged craters throughout the mid and low-latitude regions of the Moon are crucial for reconstructing the dynamic evolution process of the Solar System.
قيم البحث
اقرأ أيضاً
Crater counting on the Moon and other bodies is crucial to constrain the dynamical history of the Solar System. This has traditionally been done by visual inspection of images, thus limiting the scope, efficiency, and/or accuracy of retrieval. In thi
s paper we demonstrate the viability of using convolutional neural networks (CNNs) to determine the positions and sizes of craters from Lunar digital elevation maps (DEMs). We recover 92% of craters from the human-generated test set and almost double the total number of crater detections. Of these new craters, 15% are smaller in diameter than the minimum crater size in the ground-truth dataset. Our median fractional longitude, latitude and radius errors are 11% or less, representing good agreement with the human-generated datasets. From a manual inspection of 361 new craters we estimate the false positive rate of new craters to be 11%. Moreover, our Moon-trained CNN performs well when tested on DEM images of Mercury, detecting a large fraction of craters in each map. Our results suggest that deep learning will be a useful tool for rapidly and automatically extracting craters on various Solar System bodies. We make our code and data publicly available at https://github.com/silburt/DeepMoon.git and https://doi.org/10.5281/zenodo.1133969 .
We present a study on the relationship between the ratio of the depth of a crater to its diameter and the diameter for lunar craters both on the maria and on the highlands. We consider craters younger than 1.1 billion years in age, i.e. of Copernican
period. The aim of this work is to improve our understanding of such relationships based on our new estimates of the craterss depth and diameter. Previous studies considered similar relationships for much older craters (up to 3.2 billion years). We calculated the depths of craters with diameters from 10 to 100 km based on the altitude profiles derived from data obtained by the Lunar Orbiter Laser Altimeter (LOLA) onboard the Lunar Reconnaissance Orbiter (LRO). The ratio h/D of the depth h of a crater to its diameter D can diverge by up to a factor of two for craters with almost the same diameters. The linear and power approximations (regressions) of the dependence of h/D on D were made for simple and complex Copernican craters selected from the data from Mazrouei et al. (2019) and Losiak et al. (2015). For the separation of highland craters into two groups based only on their dependences of h/D on D, at D<18 km these are mostly simple craters, although some complex craters can have diameters D>16 km. Depths of mare craters with D<14 km are greater than 0.15D. Following Pikes (1981) classification, we group mare craters of D<15 km as simple craters. Mare craters with 15<D<18 km fit both approximation curves for simple and complex craters. Depths of mare craters with D>18 km are in a better agreement with the approximation curve of h/D vs. D for complex craters than for simple craters. At the same diameter, mare craters are deeper than highland craters at a diameter smaller than 30-40 km. For greater diameters, highland craters are deeper.
We use numerical modeling to investigate the combined effects of impact velocity and acoustic fluidization on lunar craters in the simple-to-complex transition regime. To investigate the full scope of the problem, we employed the two widely adopted B
lock-Model of acoustic fluidization scaling assumptions (scaling block size by impactor size and scaling by coupling parameter) and compared their outcomes. Impactor size and velocity were varied, such that large/slow and small/fast impactors would produce craters of the same diameter within a suite of simulations, ranging in diameter from 10-26 km, which straddles the simple-to-complex crater transition on Moon. Our study suggests that the transition from simple to complex structures is highly sensitive to the choice of the time decay and viscosity constants in the Block-Model of acoustic fluidization. Moreover, the combination of impactor size and velocity plays a greater role than previously thought in the morphology of craters in the simple-to-complex size range. We propose that scaling of block size by impactor size is an appropriate choice for modeling simple-to-complex craters on planetary surfaces, including both varying and constant impact velocities, as the modeling results are more consistent with the observed morphology of lunar craters. This scaling suggests that the simple-to-complex transition occurs at a larger crater size, if higher impact velocities are considered, and is consistent with the observation that the simple-to-complex transition occurs at larger sizes on Mercury than Mars.
Planetary impact events eject large volumes of surface material. Crater excavation processes are difficult to study, and in particular the details of individual ejecta fragments are not well understood. A related, enduring issue in planetary mapping
is whether a given crater resulted from a primary impact (asteroid or comet) or instead is a secondary crater created by an ejecta fragment. With mapping and statistical analyses of six lunar secondary crater fields (including Orientale, Copernicus, and Kepler) we provide three new constraints on these issues: 1) estimation of the maximum secondary crater size as a function of distance from a primary crater on the Moon, 2) estimation of the size and velocity of ejecta fragments that formed these secondaries, and 3) estimation of the fragment size ejected at escape velocity. Through this analysis, we confirmed and extended a suspected scale-dependent trend in ejecta size-velocity distributions. Maximum ejecta fragment sizes fall off much more steeply with increasing ejection velocity for larger primary impacts (compared to smaller primary impacts). Specifically, we characterize the maximum ejecta sizes for a given ejection velocity with a power law, and find the velocity exponent varies between approximately -0.3 and -3 for the range of primary craters investigated here (0.83-660 km in diameter). Data for the jovian moons Europa and Ganymede confirm similar trends for icy surfaces. This result is not predicted by analytical theories of formation of Grady-Kipp fragments or spalls during impacts, and suggests that further modeling investigations are warranted to explain this scale-dependent effect.
We compare the number of lunar craters larger than 15 km across and younger than 1.1 Ga to the estimates of the number of craters that could have been formed for 1.1 Ga if the number of near-Earth objects and their orbital elements during that time w
ere close to the corresponding current values. The comparison was performed for craters over the entire lunar surface and in the region of the Oceanus Procellarum and maria on the near side of the Moon. In these estimates, we used the values of collision probabilities of near-Earth objects with the Moon and the dependences of the crater diameters on the impactor sizes. According to the estimates made by different authors, the number density of known Copernican craters with diameters D>15 km in mare regions is at least double the corresponding number for the remaining lunar surface. Our estimates do not contradict the growth in the number of near-Earth objects after probable catastrophic fragmentations of large main-belt asteroids, which may have occurred over the recent 300 Ma; however, they do not prove this increase. Particularly, they do not conflict with the inference made by Mazrouei et al. (2019) that 290 Ma ago the frequency of collisions of near-Earth asteroids with the Moon increased by 2.6 times. For a probability of a collision of an Earth-crossing object (ECO) with the Earth in a year equaled to 10^-8, our estimates of the number of craters agree with the model, according to which the number densities of the 15-km Copernican craters for the whole lunar surface would have been the same as that for mare regions if the data by Losiak et al. (2015) for D<30 km were as complete as those for D>30 km. With this collision probability of ECOs with the Earth and for this model, the cratering rate may have been constant over the recent 1.1 Ga.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
