Evidence date: 15 October 2026. Prepared by Amaravati Property Desk. Read the article.
The four waterway maps are original geographic context, not flood maps. Blue is selected mapped water, not an observed flood classification. No protected area, flood depth, tunnel route or complete catchment has been inferred. Map scales differ. Each map uses a local UTM projection, a projected scale bar and grid north.
The local map uses OSM water features and an APCRDA capital planning outline captured on 4 October 2026. The live APCRDA boundary service could not be revalidated on 15 October; the dated capture is disclosed. The outline is not a property boundary or a flood-protection boundary. The OSM name “Kondaveedu Vaagu” is retained as a database label, not treated as a surveyed reconstruction of the entire Kondaveeti/Pala network.
International maps use selected named water features captured on 15 October 2026. Approximate infrastructure markers identify geographic context. The Showa/Ryukyukan marker uses a nearby bus-stop point, not a surveyed pump-building coordinate. Official PUB, Rijkswaterstaat and MLIT documents establish the infrastructure roles. The Japan facility is at Kasukabe, Saitama; it does not protect all Tokyo.
Map data © OpenStreetMap contributors, available under the Open Database Licence 1.0. These retained extracts remain available under the ODbL. Original cartography does not turn community map data into an official hazard map.
Only closed water areas were filled. Open or clipped fragments remain lines; no closing edges were invented. Areas without blue must not be interpreted as dry in a flood or protected. The small named-feature capture was cached, serial and slower than one request per second; it is not a live website geocoder.
The analysis uses the retained Copernicus GLO-30 tile Copernicus_DSM_COG_10_N16_00_E080_00_DEM, accessed through Microsoft Planetary Computer. Data were read onto a 30 m EPSG:32644 grid with nearest-neighbour resampling and a cell-centre mask of the planning outline. All 241,738 selected cells were valid. Boundary area is about 217.6 km². Calculation receipt.
Sampled DSM heights are 16.5 m at the 5th percentile, 21.9 m at the median and 41.9 m at the 95th percentile, relative to EGM2008. These are surface-model statistics, not minimum/maximum surveyed ground levels. Water, buildings, vegetation and hills are included. Underlying TanDEM-X observations principally date from 2011–2015; this is not a current ground survey. No flood probability, water depth or safety clearance was calculated. Local engineering reduced levels and river gauges must not be subtracted from this dataset without reconciling vertical references.
produced using Copernicus WorldDEM-30 © DLR e.V. 2010-2014 and © Airbus Defence and Space GmbH 2014-2018 provided under COPERNICUS by the European Union and ESA; all rights reserved
The organisations in charge of the Copernicus programme by law or by delegation do not incur any liability for any use of the Copernicus WorldDEM-30.
Copernicus product, specifications and licence · Dataset DOI.
The article cites NRSC/ISRO’s RISAT-1A rapid product acquired at 06:00 IST on 3 September 2024. Its pages 3–4 show different riverbank patterns and local southern-area water coverage. The preliminary product was not ground-verified and water can include low-ground/paddy standing water. Its pre-event images are from January–February, a seasonal mismatch with September. No polygon-fitted flood percentage or building damage estimate was derived. The original source is linked; its “For Official Use” pages are not reproduced as publication artwork.
The 2018 EIA’s account of the capital site escaping the 2009 Krishna flood is a retrospective project report. Its description of annual local Kondaveeti Vagu inundation is a historical baseline, not a census through 2026. The World Bank’s 2024 statement concerns its program area, not a certificate for every part of the planned city. Government accounts of operation or work commencement are labelled as such. A contract award or finance order is not a commissioning record.
A discovery sample of six political/media items did not establish the exact universal wording “Amaravati floods every monsoon” in those items, or deliberate fabrication by a named outlet. It was not an exhaustive internet census. No individual video was conclusively geolocated and found fabricated. The source/date/location of each specific claim still matters.
The reviewed evidence does not establish a bank-wide safety advantage for Vijayawada over Amaravati. Nor does it establish the reverse or equal risk. Left and right mean looking downstream; they are not height or protection categories. Compare named sites in the same hydraulic reach, including their position upstream or downstream of Prakasam Barrage. USACE HEC-RAS convention and cross-section inputs.
The NRSC Budameru event model hosted by APSAC uses a 2.5 m CartoDEM, rainfall/runoff modelling and a satellite-extent comparison for the 2024 event. This is evidence of a studied left-side hazard, not a normalized future comparison with Amaravati. Reported agreement of flood extent is not validation of every modelled depth. Direct local downloads of this PDF timed out on 15 October; its text was inspected through the web research service and the source archive link is retained. No new simulation was executed.
The Krishna district irrigation account documents a left-margin protection-wall project at Ramalingeswara Nagar. Its historical progress percentage is not treated as current commissioning evidence. On the right side, official studies describe Kondaveeti/Pala runoff, flat low tracts and constrained discharge when the Krishna is high. River walls or embankments do not by themselves dispose of internal rainfall.
A matched assessment needs current surveyed ground/floor/access levels in compatible vertical references; common Krishna, tributary and local-rain scenarios; comparable urbanisation and climate horizons; and the actual condition and operating reliability of walls, drains, storage, outlets, pumps and power. It must assess overtopping, breaches, blockage and relevant equipment failure, then report depth, duration, access and uncertainty. A city DSM median or the 2024 event alone cannot establish that ranking. For a building, finished-floor elevation minus the scenario water-surface elevation is a useful screening margin only after datum reconciliation and the required allowance; other hazards and access still require checking.
For an unchanged annual exceedance probability p and independent years, P(at least one exceedance in n years) = 1 − (1 − p)^n. At p = 0.01 and n = 30, this is about 26%. This is a generic design-event illustration, not measured Amaravati flooding, pump-failure or household-loss probability. Climate, land use and other changes can invalidate stationarity assumptions. The five-year local drain standard and 100-year major-system design descriptions do not directly equal household flood probabilities.
Official records, operator statements, calculations and editorial inference remain distinct. No independent hydraulic certification, property inspection, matched international risk ranking or complete capital-site alternatives appraisal was performed. Sources and their dates are listed with the article. Correction policy.