Figure 9 - uploaded by Mario Di Martino
Content may be subject to copyright.
Scanner aerial picture of the lake Cheko and river Figure 10: Aerial photo of part of the Southern swamp (1999), Kimchu (1999), spectral band: 0.61-0.69 µm. where is located the point usually called "Fast Epicentre".
Source publication
A multidisciplinary investigation of the Tunguska site (Central Siberia) devastated in 1908 by the explosion of a cosmic body has been carried out in July 14-29, 1999 by the Tunguska99 expedition (see http://www-th.bo.infn.it/tunguska/). In this framework, the remote sensing of a 300-km2 territory has been performed in collaboration with the Russia...
Context in source publication
Context 1
... we can see in Fig. 7 that the Northern surroundings of the lake Cheko, at the time of Kulik's survey (1938), was covered by the grown forest thus masking some details appearing in Fig. 8 (1999) shot after the site was cleared by a local forest fire (1995). Further information can be obtained from the scanner images in different spectral bands (Fig. 9). The comparison between the 1938 and 1999 aerophotosurveys can shed light on the location of possible secondary "epicenters" of the explosion. Fig. 10 shows a portion of the western part of the Southern swamp photographed on 26 July 1999. Kulik's analysis 6-7 has indicated in this area the possible presence of two secondary centers of ...
Citations
... Original expeditions by Kulik (1933Kulik ( , 1939Kulik ( , 1976, beginning 20 years after the event, discovered radially felled trees but found no meteorite material. Later surveys conducted and documented by Florenskiy, Fast, and Longo better mapped the tree damage regions, which formed a butterfly shaped blast footprint covering over 2000 km 2 , extending 15-35 km radially outward around an epicenter where 'telegraph pole' trees remained standing with broken and burned branches (Florenskiy, 1963;Fast 1967;Fast et al., 1967Fast et al., , 1983Longo and di Martino, 2003). Vasilyev (1998), Trayner (1997), and Florenskiy (1963) provide good summaries of the expeditions and forest damage data, and Longo et al. (2005) consolidated the most comprehensive damage map from multiple survey data sources. ...
The Tunguska meteor airburst has been extensively studied and modeled in attempts to deduce its size, properties, and impact characteristics. However, most of the existing modeling and simulation studies have investigated a small subset of cases based on assumptions of representative densities, velocities, or other properties. In this study, we use a probabilistic asteroid impact risk model to assess the entry, burst, and ground damage from 50 million Tunguska-scale asteroid impacts, covering a full range of potential impactor properties. The impact cases are sampled from probabilistic distributions representing our current knowledge of asteroid properties, entry trajectories, and size frequencies. The results provide a broader characterization of the range and relative likelihood of asteroid properties that could yield Tunguska-scale impacts. Results show that Tunguska-like events can be produced by a broad range of impact scenarios, and prevailing size and energy estimates of 50–80 m or 10–20 Mt remain within the relatively likely property ranges. However, our results suggest that objects with slightly larger initial energies of 20–30 Mt and diameters 70–80 m are more likely to cause Tunguska-scale damage areas than objects on the smaller end of the potential size range. Even when relative size frequencies are accounted for, the greater damage potential of larger objects outweighs their rarity, while the low damage potential of small objects counteracts their frequency.
... The effects of the explosion of the Tunguska Cosmic Body (TCB) on the forest have been described by many authors, and their data gave information on the released energy and the height of the explosion. The map of the fallen trees was recently reconstructed and enlarged by Longo et al. (2005) using aerial surveys and data collected during investigations performed on the ground , 2003. This map suggests an azimuth trajectory a = 110 ± 5°. ...
Dendrochronological research was carried out on 23 trees samples (Larix sibirica and Picea obovata) sampled during the 1999 expedition in two locations, close to the epicentre zone and near Cheko lake (N 60°57′, E 101°51′). Basal Area Increment (BAI) analysis has shown a general long growth suppression before 1908, the year of Tunguska event (TE), followed by a sudden growth increase due to diminished competition of trees that died due to the event. In one group of the trees, we detected growth decrease for several years (due to damage to the trunk, branches and crown), followed by growth increase during the following 4–14 years. We show that trees that germinated after the TE, and living in close proximity of Cheko lake (Cheko lake trees) had different behaviour patterns when compared to those trees living further from Cheko lake, inside the forest (Forest trees). Cheko lake trees have shown a vigorous continuous growth increase. Forest trees have shown a vigorous growth during the first 10–30 years of age, followed by a period of suppressed growth. We interpret the suppressed growth by the re-established competition with the surroundings trees. Cheko lake pattern, however, is consistent with the formation of the lake at the time of TE. This observation supports the hypothesis that Cheko lake formation is due to a fragment originating during TE, creating a small impact crater into the permafrost and soft alluvial deposits of Kimku River plain. This is further supported by the fact that Cheko lake has an elliptical shape elongated towards the epicentre of TE.
... Neither is there evidence of any ejected material; in fact, aerial photos of the lake from 1938 and 1999 show mature trees that pre-date 1908 lining the rim of the lake (Longo and Di Martino, 2003). It is hard to imagine how a violent impact event could excavate a 300-m-wide hole without affecting trees so close. ...
In a provocative paper Gasperini et al. (2007) suggest that Lake Cheko, a ∼300-m-wide lake situated a few kilometres downrange from the assumed epicentre of the 1908 Tunguska event, is an impact crater. In this response, we present several lines of observational evidence that contradicts the impact hypothesis for the lake’s origin: un-crater-like aspects of the lake morphology, the lack of impactor material in and around the lake, and the presence of apparently unaffected mature trees close to the lake. We also show that a tensile strength of 10–40 MPa is required for an asteroid fragment to traverse the Earth’s atmosphere and reach the surface intact and with sufficient velocity to excavate a crater the size of Lake Cheko. Inferred tensile strengths of large stony meteorites during atmospheric disruption are 10–100 times lower. We therefore conclude that Lake Cheko is highly unlikely to be an impact crater.
... To analyse the 1938 aerophotosurvey (APS) and to link its photos to the ground, the Tunguska99 expedition carried out a new APS. The 1999 APS78 covered a ~300 km 2 surface between the latitudes 60° 50' 00" N and 60° 58' 30" N and between the longitudes 101°45' 00" E and 102° 05' 00" E, corresponding to an area a little larger than that of the 1938 APS (dashed line inFig.1). Finally, we carried complementary on-site measurements in July 1999 and 2002 to obtain the coordinates of different reference points in the same area. ...
... Evidence that Lake Cheko is not an impact crater • G. S. Collins et al. (Longo and Di Martino, 2003). It is hard to imagine how a violent impact event could excavate a 300-m-wide hole without affecting trees so close. ...
In a provocative paper Gasperini et al. (2007) suggest that Lake Cheko, a 300-m-wide lake situated a few kilometres down- range from the assumed epicentre of the 1908 Tunguska event, is an impact crater. In this response, we present several lines of observational evidence that contradicts the impact hypothesis for the lakes origin: un-crater-like aspects of the lake mor- phology, the lack of impactor material in and around the lake, and the presence of apparently unaffected mature trees close to the lake. We also show that a tensile strength of 10-40 MPa is required for an asteroid fragment to traverse the Earths atmosphere and reach the surface intact and with sufficient velocity to excavate a crater the size of Lake Cheko. Inferred tensile strengths of large stony meteorites during atmospheric disruption are 10-100 times lower. We therefore conclude that Lake Cheko is highly unlikely to be an impact crater.
The detection and mapping of crops are important for estimating potential harvest as well as for agricultural field management and land cover change. Optical and Synthetic Aperture RADAR (SAR) remote sensing are among the most attractive options because it offers vegetation indices and some data have been distributed free of charge. Few years ago, the European Union (EU) started an ambitious program, Copernicus, that includes the launch of a new generation of earth observation satellites known as Sentinel. Sentinel-1 carries a Synthetic Aperture RADAR (SAR) that operates on the C-band. This platform offers SAR data day- and-night and in all-weather conditions. The Interferometric Wide (IW) mode was chosen since it is the main and default operational mode over land of the Sentinel-1 satellite.
The dual VV-VH polarizations were selected in line with the results of previous studies that found these polarizations useful in crop classification. Finally, a new product consisting of the ratio between the pixel values of the intensity band in VH and VV has been calculated for each image. Sentinel-2(A-B), which are equipped with a multispectral sensor (MSI) with blue, green, red and near-infrared-1 bands at 10 m; red edge 1 to 3, near-infrared-2 and shortwave infrared 1 and 2 at 20 m; and 3 atmospheric bands (Band 1, Band 9 and Band 10) at 60 m. Sentinel 2 also offer some vegetation indices calculated to assess vegetation status, a selection of 7 of the most important vegetation indices used for remote sensing of agriculture has been chosen for this study.
Furthermore, information about the morphology of the terrain available trough a Digital Elevation Model, provided by NASA, has been involved in the model.
The geo-referenced data about the crop type and other related variables used as predictors, such as irrigation,surface cover technique and fence presence, have been provided by Italian Ministry of Agriculture (MiPAAF), in particular micro data from 2016 AGRIT project has been used. The selected study area is the center-north part of Tuscany.
The main present crop(Grape, Olive), two seasonally product(Sunflower, Winter cereal) with different prevalence in the sample(0.03 $\%$ and 0.16$\%$ respectively), and finally a class called "other" that contain all the unit with prevalence lower than 0.05 $\%$, have been selected as 5 response classes in the models.
The main objective of this study has been to compare the classification performance obtained after performing several ensemble methods based on classification trees and a parametric model. More precisely, the non parametric models that have been tested are: Bagging, Boosting and Random forest; while, the tested parametric model is Penalized Logistic Regression. The results have shown the best accuracies and robustness with Random Forest (0.83).
The best classifier has been chosen, in order to test it in four different scenario: i) Sentinel-1, ii) Sentinel-2, iii) Sentinel-2 and vegetation indices, iv) Sentinel-1, Sentinel-2, DEM and vegetation indices. Moreover, the classification performance of Random Forest classifier has been tested for different sample size of the train set. The classifications have been performed using a pixel approach. Furthermore, from the robustness point of view, the classification performance of Random Forest has been tested for different sample size of the train set. The results of the learning curve for the Random Forest are stable(minimum 0.79 and maximum 0.83) with respect to variation of the sample size(from 0.6 to 0.8 percentage of the entire sample). The maximum accuracy is obtained when using Sentinel-1, Sentinel-2, DEM and vegetation indices.
This final result shows the great advantage of using combined multi-sensor and multitemporal sentinel data, in a pixel based approach, for mapping agricultural land use.
Longo G.: "The Tunguska event" . Chapter 18, pp. 303-330 in the book: "Comet/Asteroid Impacts and Human Society, An Interdisciplinary Approach, Bobrowsky, Peter T.; Rickman, Hans (Eds.)." , 546 p., © Springer-Verlag, Berlin Heidelberg New York, 2007
In the early morning of 30th June 1908, a powerful explosion over the basin of the Podkamennaya Tunguska River (Central Siberia), devastated 2 150 ± 50 km2 of Siberian taiga. Eighty millions trees were flattened, a great number of trees and bushes were burnt in a large part of the explosion area. Eyewitnesses described the flight of a fire ball, bright as the sun. Seismic and pressure waves were recorded in many observatories throughout the world. Bright nights were observed over much of Eurasia. These different phenomena, initially considered non-correlated, were subsequently linked together as different aspects of the Tunguska event (TE).
The forest damage (fallen trees) by a typhoon or a strong wind is frequently generated in recent years. For example, typhoon SONGDA (2004.09) brought about destructive strong winds to almost western half of Hokkaido. According to a temporary report issued by Japan Meteorological Agency, in Sapporo, 50.2 m/s was recorded as a maximum instantaneous wind speed in the history of observation. This typhoon also caused serious damage to the forest of Hokkaido. ASTER (advanced spaceborne thermal emission and reflection radiometer) aboard NASA's satellite Terra is a high resolution multispectral radiometer of 14 bands. The spatial resolution is 15 m in visible and near infrared, 30 m in short wave infrared and 90 m in thermal infrared spectra respectively. Data from ASTER were used to produce the along of fallen trees extraction for the south western coastal area of Hokkaido. We evaluated the damage of forestry based on the spectrum characteristics of fallen trees. As a result the extraction of almost all fallen trees data has been succeeded.
