India–Asia section A–A′
SELECTED TIME
Present 0 Ma
Modern World
Present
−30 °C 0 20 40 °C
Relief shading No shading True relief Relief ×5 Relief ×20 Sea level +0 m
This view colours the PALEOMAP PaleoDEM 6-minute (0.1°) elevation and bathymetry grids (Scotese & Wright 2018, CC BY 4.0). The source data is on Zenodo ↗ .
The coastline is the grid's 0 m contour and the colours are height and depth. Shading lights the grid's slopes from the upper left; ×5 and ×20 exaggerate the slopes. Before land plants spread, 470 to 385 Ma, land is drawn in bare rock tones. Before 540 Ma the timeline continues with masks segmented from the PaleoAtlas (2016) maps, which carry no heights. Nothing here computes plate motion, climate or sea level: it draws the published reconstruction grids as released.
Showing 3D terrain. Zoomed in close, the ground is raised by the grid's heights and the view tilts toward the horizon. Heights are exaggerated 25 times, and the sea is kept flat at sea level. Close up, the grid cells can show.
Showing ice. The present-day glaciers and ice sheets (white) and the Antarctic ice shelves (pale blue) are Natural Earth 10 m data, public domain. Past ice is this project's segmentation of the ice the PALEOMAP PaleoAtlas (2016) maps paint white; each grid borrows the nearest map within 5 Myr. Maps 5 to 10 Myr apart cannot show glacial cycles, and sea ice the maps paint white appears too. White on those maps is also the colour of the highest ground, so plateaus inside 45° of latitude, Tibet among them, are left out, and mountain glaciers appear only at the present. There is no data before 540 Ma. The West Antarctic ice sheet rests on bedrock below sea level, so ice sits where the grid draws ocean.
The ice at this stop is a modelled limit, not a drawn outline. The PaleoAtlas (2016) paints none here, but the land-ice estimate of van der Meer et al. (2022) exceeds 5 million km³ (the shaded span on the strip), so a cap is drawn poleward of the same paper's ice-latitude limit.
Rivers are shown. These lines are neither observations nor a published reconstruction: they are potential drainage this project computed by routing water over each stop's PaleoDEM grid. The grid is resampled to 2048 × 1024, every pit is filled to the level at which it spills, and each cell sends its water to the steepest of its eight neighbours, summing the area upstream. Cells draining more than 10,000 km² are drawn, thicker the larger the area. Rain is taken as even and evaporation as nil, so this is drained area, not discharge; and the grids are smooth, so gorges narrower than a cell are closed and basins that drain through one (the Congo, Sichuan) spill over just as basins closed for real (Chad, Tarim) do. Between stops the two networks are mixed. Lowering the sea-level slider mixes in a field routed again with the sea at the stop's lowest level, so the rivers run on across the exposed shelf. Grid spacing is not reconstruction accuracy. The 2048×1024 calculation grid spans about 20 km per cell at the equator; it does not validate individual valleys, gorges, or actual ancient river courses. The glacial river fields show computed pools with ice and crustal depression included (from about 2.5 m of computed depth). An approximate rule excludes a pool if at least half its footprint also pools in the bare grid. These are not observations or validated lake reconstructions. On the present grid the water is routed along today's rivers (Natural Earth), so it leaves basins the grid closes because their gorge is narrower than a cell (the Danube at the Iron Gates), and today's lakes (HydroLAKES) are painted at their mean depth; the Caspian stays a lake whatever the sea-level slider says. The other grids have no such data, so every basin spills over as described above. On the present grid a lowered sea switches to fields routed over the ice of the glacial cycle at that sea level: the ice thickness and crustal depression of PaleoMIST 1.0 (Gowan et al. 2021, CC BY 4.0) laid on today's grid every 2,500 years, so meltwater runs along the ice margins and a lake dammed by ice drains where it spills. Nothing is drawn on the ice itself, and between two steps the fields are mixed by sea level.
Showing a sea-level offset: a what-if view that shifts the grid's heights from that stop's own sea level and recuts the coast; not a reconstruction. 'Apply curve departure' moves it only by how far the long-term curve of van der Meer et al. 2022 departs from the datum of the two bracketing grids. The grids already carry the sea level of their time, so the curve's value is never added outright. With the ice layer on, the sheets move too: about 0.4 million km³ of land ice per metre of sea level (the same paper's ratio) and area as the 0.8 power of volume, each sheet growing or shrinking evenly from its edge, an approximation. At the present stop the ice follows a reconstruction of the last deglaciation: the margins of NADI-1 (Dalton et al. 2023, CC BY 4.0) for North America and DATED-1 (Hughes et al. 2016, CC BY 3.0) for Eurasia, laid over today's ice every thousand years, each at its age's sea level from the Spratt & Lisiecki (2016) stack, and the slider mixes between them (−130 m is 24 ka). Antarctica, Greenland, Iceland and the mountain glaciers keep their present extent. Where there is no ice none appears.
Showing surface air temperature: blue at −30 °C through pale at 0 °C to red at 40 °C. These are the 1° temperature maps of Scotese (2021) , climate-model output (Valdes et al. 2021 ) nudged to isotope and rock indicators. A compromise between model and proxies, not an observation; between stops it is interpolated. The global mean is the area-weighted mean of each map.
Showing the continent mask: land and ocean as two flat colours. The elevation grids are cut at 0 m; the PaleoAtlas maps before 540 Ma are an interpretation segmented from colour and carry no heights.
Choose a plate rotation model from the boundaries dropdown in the menu (☰) to reconstruct the continents at the same time and draw them over the map. There are 5 models; the default is no boundaries. With a model chosen, its plate boundaries are drawn over the land mask.
The pink lines are a different dataset: the continents at the same time reconstructed from a plate rotation model, with a different source and method from the segmented Scotese maps. Where the two disagree, the two models disagree.
Click the map to drop a pin: one is enough, up to three. Click a pin again, or its ×, to remove it. A pin rides the plate its land sits on as you move through time; with two or more, the distances between them are listed.
A pin's position is computed from the PALEOMAP rotation model (Scotese 2016, CC BY 4.0), not observed. Plates are treated as rigid, so two pins on one plate never change their distance. Other models put the same place elsewhere, and before 300 Ma distances differ between models by up to a factor of two.
The orange lines are PaleoCoastlines: the PALEOMAP elevation model's coastlines moved to the maximum transgression that marine fossils in the Paleobiology Database indicate. Where a line runs inside the mask's edge, the fossils say that ground was sea. They share the PALEOMAP frame with the mask, so they are drawn without rotation.
Kocsis, A. T., and Scotese, C. R., 2021. Mapping paleocoastlines and continental flooding during the Phanerozoic. Earth-Science Reviews, 213, 103463. Data: doi:10.5281/zenodo.4297693 · Creative Commons Attribution 4.0 International
Sea level over the last 800,000 years, metres relative to present (Spratt & Lisiecki 2016 ). Older to the left, the horizontal line is present sea level, glacial minima about −130 m. Variation no 5 Myr grid can show. The present (0 Ma) is shown as a satellite image: NASA Blue Marble (shaded relief and bathymetry, a composite of about 2004), fetched from NASA EOSDIS GIBS and served from this server. It is not a photograph of today but a cloud-free composite. The ice is the ice in the photograph, and its coastline can sit a few cells off the elevation grid's. Turn off 'Satellite image' in the menu (☰) to return to the elevation colours.
The present-day wind, currents and clouds are not live. Wind and clouds are the day's 12 UTC analysis, fetched once a day (around 17:10 UTC); the time is written under the globe and in the legend. The currents are a mean over many years. Turn them on and off in the menu (☰).
Wind is the NOAA GFS 0.25° analysis (10 m, 250 hPa); a particle's colour is its speed. Satellite clouds are the NOAA/NESDIS GMGSI geostationary infrared mosaic: empty beyond 72.7° latitude, and cold ground can look like cloud. Model clouds are the total cloud cover of the same GFS analysis.
The currents are the surface (5 m) velocity of the ECCO2 cube92 ocean model, one snapshot a month over 1992–2018, 324 in all, averaged. A model, not an observation; the averaging removes the eddies and leaves the large flows. The Baltic Sea is not in the model.
NOAA/NCEP GFS · NOAA/NESDIS GMGSI (public domain) · ECCO2 cube92 (NASA JPL·MIT), Menemenlis et al. 2008 · NASA Blue Marble · EOSDIS GIBS
Older than the oldest Scotese map. There is no map for this time, so only sea is painted and, if enabled, only the plate reconstruction is drawn. The globe's surface claims nothing.
This is an interpolated view between grids 5 Myr apart. Continents move by borrowing the pieces of the nearest PaleoAtlas map and the rotation of the PALEOMAP plate beneath them, and the two stops' coastline distance fields and heights are blended along that path. Deformation within a continent and sea-level change are approximated.
Method and data The 109 PaleoDEM 6-minute (0.1°) grids were resampled to 2048×1024 equirectangular textures. The coastline is the 0 m contour, the colours height and depth. Before 540 Ma three PaleoAtlas (2016) masks continue the timeline, 112 stops in all. Period names come from the age with ICS boundaries.
Scotese, C. R. & Wright, N. M. (2018). PALEOMAP Paleodigital Elevation Models (PaleoDEMs) for the Phanerozoic . Zenodo, CC BY 4.0.
Scotese, C. R. (2021). Global Mean Surface Temperatures for 100 Phanerozoic Time Intervals. Zenodo , CC BY 4.0. Scotese, Song, Mills & van der Meer (2021), Phanerozoic paleotemperatures , Earth-Science Reviews 215, 103503; Valdes, Scotese & Lunt (2021), Deep ocean temperatures through time , Climate of the Past 17, 1483.
Sea level: van der Meer, D. G. et al. (2022). Long-term Phanerozoic global mean sea level . Gondwana Research 111, 103–121, CC BY 4.0. Last 800,000 years: Spratt & Lisiecki (2016). A Late Pleistocene sea level stack . Climate of the Past 12, 1079–1092, data NOAA , CC BY 3.0.
Present-day ice: Natural Earth 10 m, public domain.
Potential rivers: Potential drainage computed by this project by routing water over the PaleoDEM grids above; not a published reconstruction. The present grid follows today's rivers and lakes: Natural Earth 10 m rivers and lake centerlines, public domain; Messager, M. L. et al. (2016). Estimating the volume and age of water stored in global lakes using a geo-statistical approach . Nature Communications 7, 13603 (HydroLAKES); data HydroSHEDS , CC BY 4.0.
Glacial ice: Gowan, E. J. et al. (2021). A new global ice sheet reconstruction for the past 80 000 years . Nature Communications 12, 1199; data PANGAEA , CC BY 4.0.
Compare: view with the PALEOMAP PaleoAtlas (2016) masks
Compare: view with the Scotese PALEOMAP web maps (2002) masks
Plate reconstruction · Merdith et al. 2021 (palaeomagnetic)Creative Commons Attribution 3.0 Unported
Plate reconstruction · Cao et al. 2024 (palaeomagnetic, extended to 1.8 Ga)Creative Commons Attribution 4.0 International
Plate reconstruction · Matthews et al. 2016 (hybrid mantle (GK07))Creative Commons Attribution 4.0 International
Plate reconstruction · Müller et al. 2022 (optimised mantle)Creative Commons Attribution 4.0 International
Plate reconstruction · Scotese 2016 PALEOMAP (PALEOMAP (plate 001 'Hot Spot to PMAG' is the identity))Creative Commons Attribution 4.0 International