About Visakhapatnam Water Intelligence
An open-source water intelligence platform for Indian cities, built to make public data accessible and actionable.
Sections are collapsed by default. Click any heading to expand it.
What we track for Visakhapatnam
Visakhapatnam is governed by Greater Visakhapatnam Municipal Corporation, Water Supply wing with civic services under Greater Visakhapatnam Municipal Corporation (98 wards).
Reservoirs the city draws on: 4 with a daily state reading, 2 without
All six supply the city's drinking water. Five are shared with irrigation, industry or other users; only GVMC's own Mudasarlova serves the city alone. Capacities are live storage, the water above each dam's lowest outlet. Mudasarlova's is left blank because GVMC's own figures for it disagree (see Data quality).
Reading this dashboard
Why there is no days-of-water-left number
The dashboard does not count down the days the stored water would last, and that is deliberate. Five of the six reservoirs behind the city are shared: Yeleru also feeds irrigation and the steel plant, Raiwada and Thatipudi are irrigation dams with a city share carved out of them, and Meghadrigedda and Gambheeram are dams the Water Resources Department operates and maintains, with other users beside the city. Only Mudasarlova serves the city alone, and it gives about 4.5 of some 380 MLD (CGWB 2025).
Two lifts also top the system up from the Godavari: VIWSCO's Kateru pump house sends water through a 56 km pipeline into the Yeleru canal (since December 2004), and the Purushottapatnam lift pumps it through the Polavaram Left Main Canal into Yeleru reservoir (since July 2018). Storage divided by the city's demand would count water that belongs to farms and industry and leave out water that arrives by pump, so a runway worked out that way would overstate what is the city's to drink. We have also not found a measured figure for how much the city uses each day: the 490 MLD ‘demand’ in CGWB's reports is a 2020 row carried forward unchanged.
Reading the dashboard, card by card
The dashboard follows the water from its sources to the tap.
- The supply headline. It leads with the distance: the Godavari and Yeleru system, the largest source, which CGWB lists at 190 km from the city. That is CGWB's distance to the source, not the length of the journey. The water comes by the Kateru pipeline and the Yeleru canal, which the draft Green City Action Plan puts at about 150 km (GVMC's 2021 supply sheet prints 156 km against Yeleru reservoir). The tiles beside it are GVMC's water treatment capacity (11 water treatment plants, 489 MLD, from the draft plan), that distance, and the people served: GVMC's own estimate of about 23 lakh in its NGT filings, as there has been no census since 2011.
- The supply overview. The chain from the Godavari lifts to the tap, and supply by source: about 380 MLD available from nine sources, 182 MLD of it from the Godavari and Yeleru (CGWB gives 182 to 204; 182 is the figure that makes its 380 total add up). These are GVMC's figures; CGWB's tables carry the same quantities unchanged since its June 2022 report, which said GVMC was able to draw 345.5 MLD of the 380. The newest present-supply figure we have found is 334.03 MLD, in the Chief Secretary's report to the NGT Principal Bench in OA 606/2018 (10.07.2024, Annexure-VI, p33). None of these is a daily measurement.
- Reservoir cards and history. Daily storage for Yeleru, Raiwada, Thatipudi and Meghadrigedda from APWRIMS, as live storage. Yeleru's figure includes the Godavari water the Purushottapatnam lift pumps into it; the Kateru water goes into the canal, not the reservoir, so no card holds it. Yeleru can read zero in dry spells, when its water is below the lowest outlet. Mudasarlova and Gambheeram are not in the feed and carry no reading. The history chart opens on three years; the archive runs from 2012 for Yeleru.
- Rainfall. IMD's gridded record for the cell over the city, 1970 to 2025, with the months after IMD's last published month filled from Open-Meteo and marked provisional.
What each page shows for Visakhapatnam
Dashboard - where the water comes from
Visakhapatnam has no large perennial river of its own. Its largest source is the Godavari and Yeleru system, which CGWB lists at 190 km from the city; that is CGWB's distance to the source, not the length of the water's journey. The water reaches the city through the Yeleru canal, a conveyance first built for the steel plant, about 150 km long in the draft Green City Action Plan. Two lifts bring Godavari water into the system: VIWSCO's Kateru pump house, through a 56 km pipeline to the Yeleru canal (since December 2004), and the Purushottapatnam lift, through the Polavaram Left Main Canal into Yeleru reservoir (since July 2018). That is why the dashboard opens on the supply chain rather than on days of water left.
About 380 MLD is available from nine sources, 182 to 204 MLD of it from the Godavari and Yeleru (GVMC's figures; CGWB's tables carry the same quantities unchanged since its June 2022 report). That report said GVMC was able to draw 345.5 MLD of the 380, because of shortages at Raiwada and the Gosthani. We have not found a figure that splits the system's share between the two rivers. The next largest source, Raiwada, gives 72.5 MLD. The newest present-supply figure we have found is 334.03 MLD, in the Chief Secretary's report to the NGT Principal Bench in OA 606/2018 (10.07.2024, Annexure-VI, p33).
The canal is shared
On 11 March 2021, the only day for which we have found GVMC's split, about 46 percent of the canal's tail-end water went to the city (41.641 million gallons) and 39 percent to the steel plant (35.640), with the rest to APIIC's industrial parks, NTPC, the port and rural water supply. The sheet prints no unit for the split; it is read as million gallons a day, like its other columns.
What we have not found
A measured non-revenue water figure for GVMC: the draft plan puts losses at nearly 30 percent on one page and 35 percent on another, without saying how either was measured, so no losses figure is shown. A measured figure for the groundwater the city draws: GVMC's 2021 sheet lists 2.75 million gallons a day from bores as a set figure, the draft plan says 30 percent of the city's needs, and CGWB's 2025 plan 17.37 percent of supply.
Reservoirs - daily from the state, none of them the city's alone
The reservoir cards carry the four dams in APWRIMS, the Water Resources Department's portal: Yeleru, Raiwada, Thatipudi and Meghadrigedda. APWRIMS publishes level and storage for each every morning through a public interface with no login, with a daily archive from 2012 for Yeleru, so the history can be read back rather than starting on the day we began collecting. Storage is shown as live storage: the published gross storage less the design dead storage below the lowest outlet. Up to 15 September 2026 the portal's figures for Meghadrigedda were already live storage, reaching the design live capacity of about 1.04 TMC at full level, so they are used as published.
None of them is the city's alone. Yeleru also feeds irrigation and the steel plant; Raiwada (completed 1981-82) and Thatipudi (1968) are irrigation dams with a city share; Meghadrigedda is a dam the Water Resources Department operates and maintains. GVMC's own Mudasarlova reservoir, the town's first water works (handed over in May 1903), and the Gambhiramgedda reservoir, which GVMC draws on from March to June, are not in the feed, so their cards carry no reading. Dividing the stored water by the city's demand would count water that belongs to farms and industry and leave out the water the lifts bring, so no days-left runway is computed.
Readings are checked on every run
Each dam's design values (full reservoir level, gross and dead storage) are checked against the portal's own attributes on every run, and the reader stops if they change. Each reading is dated by its own timestamp, and days whose level or storage does not pass a basic check, including days whose storage does not fit their level, are left out. Yeleru reads zero in dry spells, when its water is below the lowest outlet.
Allocations - one canal, many claimants
The allocations page sets out each documented claim on the city's raw water that we have found, with the instrument that sets it and the one dated record of what was delivered: the city's share of the Polavaram Left Main Canal, GVMC's allocations from the Yeleru and from the Godavari at Kateru, its share of each reservoir, and the claims of the steel plant, NTPC, APIIC's industrial parks, Gangavaram Port and the HPCL refinery.
Most delivered figures come from one day, 11 March 2021, the only day for which we have found GVMC's split. The recycled water delivered to HPCL's refinery, 25 MLD, is from the state's February 2026 affidavit in NGT Principal Bench OA 606/2018. Three official figures for the city's Polavaram share are shown side by side: 23.44 TMC a year (Polavaram Project Authority), 24 TMC (a 2004 order cited in GVMC's 2019 assessment for the Asian Development Bank, which we have not found) and 12 TMC (CGWB 2025). The page also lists four claims still to come. Polavaram Left Main Canal water for the city and industry, where CAG reported 76.05% physical progress on the canal works in March 2023, the latest official figure we have found. The AM/NS steel plant at Anakapalli, whose 2026 environmental clearance says about 104 MLD will be obtained from the same canal, subject to permission; we have not found a Water Resources Department allocation. APIIC's data-centre park at Madhurawada, whose 2024 clearance allows up to 1.43 MLD of fresh water, to come from GVMC; we have not found an allotment order. And three data-centre campuses whose clearances list 0.69 MLD of fresh water.
Orders cited but not found
The 2004 order for 24 TMC; the VIWSCO concession agreement and the steel plant's bulk water supply agreement; an order setting priority between the city and industry on the Yeleru canal in a dry year; an overall entitlement for APIIC's parks; and a plant-level entitlement for NTPC's Simhadri plant. Where a row has no entitlement we have found, its note says so.
Commitments - mostly about sewage
The commitments register tracks 18 dated promises about the city's water and sewage, from GVMC's and the state's own notes, affidavits and progress reports to the National Green Tribunal and the High Court's committee, two AP High Court orders (12 November 2025 and 11 March 2026) and MoHUA's AMRUT 2.0 lists. Eleven of them are about sewers and sewage treatment: how much of the city is connected (60 percent of people, by the state's February 2026 affidavit), how much is treated, and when the rest will be.
Seven are past their latest stated date without a dated record of completion that we have found; four have had their dates moved by the institution itself; for seven we can cite the promise but not yet its current status. A status changes only against a dated document, and history is kept, never overwritten. APPCB's reports and the draft Green City Action Plan are cited only for dated status and context: a draft is not a promise. Press reports are not used.
Groundwater - by mandal, with one mandal to watch
The groundwater map colours the eleven mandals of Visakhapatnam district by the national assessment (IN-GRES), in three editions from 2023-2024 to 2025-2026. The measure is the stage of extraction: how much of the groundwater that can be drawn each year is drawn. In 2024-2025, the edition CGWB printed in its December 2025 report, Visakhapatnam Rural mandal drew 95.06 percent and was the district's only critical unit, while the district as a whole stood at 43.85 percent, which IN-GRES classes safe.
The map colours the 2025-2026 edition, newer than CGWB's December 2025 report. In it Gajuwaka reads 89.39 percent (semi-critical) and Visakhapatnam Rural 47.87 percent (safe). Both moves come from revised inputs, not measured change: Visakhapatnam Rural gains a new recharge entry from surface irrigation that nearly doubles its extractable resource, and Gajuwaka's estimated industrial draw rises from 5.4 to 97.0 hectare-metres. Every mandal's rainfall in that edition repeats its 2024-2025 figure. In the seven urban mandals, 94 percent of the groundwater drawn in 2024-2025 went to homes; in Anandapuram, Padmanabham and Pendurthi more than 95 percent is irrigation.
Why the series starts at 2023-2024
Andhra Pradesh divided the district in April 2022 and IN-GRES kept one district id across the change. Until 2022-2023 the whole city was assessed as one 513.8 km2 GVMC unit, which became seven urban mandals in 2023-24. A series joined across those redraws would show jumps that are boundary changes, so no trend is drawn across either.
The dots are wells inside the eleven mandals: automatic recorders on the state's APWRIMS portal, read every day and shown as monthly means, and AP Ground Water Department wells read by hand once a month, the oldest from 2002, from the National Water Data Portal. Where a recorder and a hand-read well stand at the same site, they are kept as separate series. Each well is named by the state mandal its coordinates fall in, not by the district it is tagged to.
Why there is no ward-by-ward depth map
The wells are too few across eleven mandals for a depth surface, and we have not found ward geometry that can be published. IN-GRES does not carry depth to water at all; its figures are a balance of recharge and draw.
Rivers - a gedda, a river and a canal
CPCB's 2019-2024 tables list no river-type station in Visakhapatnam. The rivers page shows three channels as OpenStreetMap maps them, the Meghadri Gedda, the Gosthani and the Yeleru canal, with the readings CPCB takes at the reservoirs, the Naval Dockyard creek and the canal on them. OpenStreetMap names other geddas in the city that are not drawn.
There are four stations: the Meghadrigedda reservoir and a tidal creek at the Sardar Vallabhai Patel bridge by the Naval Dockyard, close to where the Meghadri Gedda reaches the port's inner harbour; Thatipudi reservoir on the Gosthani, upstream in Vizianagaram district; and the Yeleru canal near Pharma City at Parawada. At the dockyard creek CPCB's 2024 samples recorded dissolved oxygen as low as 2.4 mg/L, after 1.0 mg/L in 2023, below the 4.0 mg/L CPCB prints as its criterion for bathing-class coastal water. CPCB's stations named ‘Gostani’ are on a different river, in West Godavari district, and are not used.
Each channel also carries dated events from official records: GVMC stopping releases from the Meghadrigedda reservoir after the LG Polymers styrene release in May 2020, and the two sets of reservoir samples that followed; APPCB's 2021 reports to the NGT on the Meghadrigedda and Thatipudi reservoirs; a joint committee's 2023 list of industrial discharges into the Meghadrigedda surplus course; and APPCB's record of the Yeleru canal as the pharma city's water source. Two industrial points are mapped: the marine outfall point recommended for the steel plant, and the JN Pharma City common effluent treatment plant, whose treated effluent goes to the sea through a marine outfall, not into the canal.
Quoted but not mapped
APPCB samples sewage channels and geddas where they meet the sea every month, but we have not found published coordinates for those stations, so their figures are quoted in the notes and never placed on the map. We have not found a published coordinate for the Atchutapuram effluent treatment plant either.
Flood risk - cyclones over a hilly city
Visakhapatnam floods from the sky and the sea, not from a large river. The flood-risk page carries a dated record of the cyclones, depressions and heavy rain that hit the city, each event credited to its own source: IMD's after-event reports for the storms (Hudhud in 2014, the deep depression of October 2020, Gulab in 2021, Michaung in 2023 and Montha in 2025), with the 24-hour rainfall at IMD's Visakhapatnam stations, and the Government of Andhra Pradesh's flood memorandum for the six days of rain from 29 August 2024 (129 mm across the district). The heaviest day in the list is Gulab's: 28 cm in the 24 hours to 8:30 am on 27 September 2021. When Hudhud crossed the coast over the city, the surge at the city's tide gauge was 1.4 m above the predicted tide.
We have not found a published release threshold for the Meghadrigedda reservoir, inside the city, or a GVMC drainage design standard, so the page has no release threshold or rainfall trigger. It draws the three channels as a sketch of how the city drains, not an engineering map, and lists the official channels to follow during an event, among them the Meghadrigedda reservoir's daily level on APWRIMS.
What we have not found
A GVMC flood map, return-period zones, a current list of waterlogging spots, a drainage master plan, GVMC drain network data or a gedda register, and city-only damage figures for Gulab, Michaung or Montha. A state GIS layer maps the drains of 2019 without names or capacities; the page mentions it and does not draw it. GVMC's older lists of inundation-prone areas use ward numbers no longer in use, so they are not used. The only modelled flood maps we have found that cover the city are regional, drawn at region scale, and do not show street waterlogging.
Shoreline - measured from satellite
The shoreline page measures change from Appikonda to Pedda Nagamayyapalem, just north of Bheemunipatnam, from Landsat and Sentinel-2 images in ten epochs from 1990 to 2026, every 100 m along the coast. The bar at the top counts reliable transects only: 147 of the 421 are retreating, 244 are stable and 30 are advancing. Most of the city's coast is holding steady; the beaches from Mangamaripeta to Pedda Nagamayyapalem are eroding.
Our rates are set beside the official record: NCCR's national shoreline assessment, and the Ministry of Earth Sciences' 2024 reply to the Lok Sabha that draws on it, which classes four stretches here as Low Erosion (Yarada, RK Beach, ‘Mangamari Peta to Bheemili’ and ‘Bheemunipatnam to Pedanagamayyapalem’). Its list does not name Lawson's Bay or Rushikonda. RK Beach is a managed beach: the port moves sand from south of its harbour onto it (sand bypassing, in the coastal erosion guidelines NIO prepared for CWC in 2020), and a sand-trap dredging for beach nourishment was inaugurated in February 2023. So a steady line or a gain there is partly the added sand, not only natural change; 30 of its 37 reliable transects read stable.
Low-confidence transects
55 transects are drawn as low confidence and left out of the counts, each with its reason: they sit on the Gangavaram port's built shore, cross a breakwater, sit at the Dolphin's Nose headland or beside river and stream mouths, show large jumps in their measured waterline, or, for seven of them, have recent positions that scatter or jump. The two port basins are not measured. No tidal correction is applied; the spring tide range here is 1.43 m.
Water bodies - from OpenStreetMap, as we have not found a lake register
The water-bodies map draws 1,088 ponds, tanks, lakes and reservoirs from OpenStreetMap, 212 of them named, in a box covering GVMC and about 5 km around it, as retrieved on 14 October 2026. It is not an official register: we have not found a gazetted lake register for the city. Drains and wastewater polygons are left out.
Five water bodies carry sourced histories: GVMC's Mudasarlova reservoir, the Meghadrigedda and Gambheeram reservoirs, the Kanithi balancing reservoir at the tail of the Yeleru canal, and Kondakarla Ava, a natural freshwater lake south-west of the city. Thatipudi and Raiwada are left out of that set: they are irrigation dams 56 to 60 km away that the dashboard tracks as supply sources, not city lakes.
A ground-height layer shows FABDEM's bare-earth elevation in seven bands, from 0-2 m to 100 m and above. FABDEM stores the sea as 0 m, so the sea is removed rather than drawn as the lowest band, and as we have not found published GVMC boundary geometry, the bands cover land just outside the city too.
No list of lost tanks, and no catchment view yet
No list of lost tanks is shown: we have not found a published official one. The lake catchment view has not been built for Visakhapatnam yet.
Lake restoration - the AMRUT 2.0 tanks
The lake-restoration page carries the city's dated, budgeted lake work: the water-body component of AMRUT 2.0, 11 GVMC tanks worth Rs 23.31 crore. MoHUA's status list of 5 March 2026 marks 10 completed and one, Thungalam, being implemented, with completion scheduled for September 2026. MoHUA's live list, checked on 14 October 2026, has no Thungalam row. In its place is a Rs 1.27 crore ‘Thimmaraju Tank’ project, which appears to be the same project under a new name; MoHUA's live list has no project IDs and does not say so. The live list also shows a 12th project, Rajupalem tank in GVMC's Anakapalli division (Rs 2.00 crore, at DPR stage), which is not in the 5 March list and is not counted.
Beside the programme, the page shows the five water bodies with sourced histories, with conflicting figures side by side. Proposals without a sanction, such as the ponds and sponge parks in the draft Green City Action Plan, are left out: a draft is not a programme.
No court orders yet
We have not found a court or tribunal order about a Visakhapatnam lake or tank to the page's standard: court, case number, date, and the order or a primary report of it. The NGT's public judgment search offers only a date range and a captcha, so this is a gap in what could be searched, not a finding that no case exists.
Facts and rainfall
The facts page is a set of quotable cards, each with its source and date, built from the same data files as the pages they summarise rather than typed in a second time. Some cards are gaps or cautions: that CPCB samples the city's channels at reservoirs, a creek and a canal but at no river-type station, and that the water-bodies map is not an official register. One card reports a peer-reviewed study and says it is research, not official monitoring.
Rainfall comes from IMD's 0.25-degree gridded record, 1970 to 2025, at the cell 17.75 N, 83.25 E, inside GVMC and the grid point nearest the city centre. Three cells nearer the coast are sea and have no value in any year from 1991 to 2025, so picking one would write zeros. We checked the choice: over 1991-2020 this cell gives 1,053.4 mm a year, against 1,054.5 mm at IMD's Visakhapatnam airport observatory and 1,113.1 mm at its other Visakhapatnam observatory. Its 1970-2020 average is about 1,043 mm. Months after IMD's last published month are filled from Open-Meteo's reanalysis and marked provisional until IMD catches up.
Origins - a harbour with no large river
The origins page is the long read, in eight chapters from the trading posts of the 1600s to the data-centre clearances of 2026: the inner harbour dug at the mouth of the Meghadri Gedda between 1927 and 1933, the first piped water from a dammed hill stream (handed over in May 1903, in the reservoir known today as Mudasarlova), the wartime Gosthani and Meghadrigedda works, the reservoirs planned for farms and the steel plant from which the city's share was carved, the Godavari lifts, Cyclone Hudhud in 2014, and what the ground and the sea are saying now.
It sets different official readings side by side rather than forcing them into one number, and it lists what we looked for and did not find in an official record, such as a count of the days the city went without piped water after Hudhud.
What is not shown, and why
Each absence below is a decision with a reason, recorded in the repository so that an omission is never mistaken for a bug.
No ward pages
We have not found ward geometry that can be published. The 98-ward boundaries in circulation are an Esri India tracing of a GVMC CAD drawing. We have not found a licence anywhere in the chain, and their zone field disagrees with GVMC's own ward list for 94 of 98 wards. The 98 wards are also superseded: GVMC notified 120 wards in 2026, and we have not found a published map of them. This closes when GVMC publishes official ward boundaries.
No tanker page
The only tanker figures from GVMC that we have found are for a single day: 615 trips on 11 March 2021, in an NGT filing. GVMC runs a Dial-a-Tanker booking line, but we have not found a series by day, zone or ward. One day is not a tanker record.
No climate-risk page
Buildable but not built. The sub-basin risk layer comes from HydroBASINS, which covers the north Andhra coast, so this is a backlog item rather than a refusal.
Not in Telugu yet
The city's name is shown in Telugu, but the pages have not been translated. A native speaker has to translate them rather than a machine, so they stay in English until then.
Intelligence & AI narratives
Visakhapatnam has no AI layer yet. Its pages carry no daily briefing, no AI-written narrative and no ward profiles. What you read on them is fixed text written against the cited documents, and numbers read from the data files listed under Data sources.
- No daily briefing. The briefing summarises current storage, its 7-day change and the days of water left, and Visakhapatnam has no days-left figure.
- No AI narrative. The longer weekly narrative is not switched on for Visakhapatnam.
- No ward profiles. There are no ward pages to attach them to; see ‘What is not shown, and why’ above.
Data sources for Visakhapatnam
All operational data is collected by the Python pipeline and supporting scripts that power the dashboard. Raw source data and Earth Engine summaries are upserted into Supabase (PostgreSQL) and then exposed as small, readable product signals.
Reservoirs and supply
Daily level, storage, inflow and outflow for Yeleru, Raiwada, Thatipudi and Meghadrigedda through a public interface with no login: the dashboard's reservoir cards and history, and the facts page's reservoir card. The archive runs from 2012 for Yeleru, from October 2014 for Thatipudi and Raiwada (which also has most of 2012) and from November 2015 for Meghadrigedda. Each dam's design values are checked against the portal's own attributes on every run. GVMC's Mudasarlova and the Water Resources Department's Gambhiramgedda reservoir are not in it. We have not found a licence or terms of use on the portal, only a disclaimer and a footer reading 'All Rights Reserved', so it is carried as restricted and cited with attribution.
The dashboard's supply-by-source table, which CGWB gives 'as per the data provided by GVMC': about 380 MLD available from nine sources, 182 to 204 MLD of it from the Godavari and Yeleru, which it lists at 190 km from the city. The same nine quantities and the 380 total are in CGWB's June 2022 report (AAP 2020-21), which said GVMC was able to draw 345.5 MLD of the 380, and in its November 2023 report, so they are shown as GVMC's figures, not a current measurement. Also read for each source's year and distance and the groundwater share of supply. Government of India publication, cited with attribution.
GVMC's daily supply sheet of 11 March 2021, annexed to its action-taken report in NGT OA 14/2021 (SZ), is the only source-by-source sheet from GVMC itself that we have found: each source's allocation, and the split of the Yeleru canal's water between the city, the steel plant, other industry and rural water supply. The Chief Secretary's report to the NGT Principal Bench in OA 606/2018 (10.07.2024, Annexure-VI, p33) gives GVMC's present supply as 334.03 MLD, and the state's affidavit in the same case (February 2026) gives the sewage and recycled-water figures. Also read: APPCB's reports in OA 14/2021 (SZ) and OA 141/2023 (SZ), GVMC's affidavit in OA 40/2024 (SZ), and the filings behind the commitments register, among them the state's monthly progress report for February 2024 in NGT Principal Bench OA 673/2018. The origins page also cites a joint committee's April 2021 report in NGT Principal Bench OA 175/2018 on the Purushottapatnam lift. Government of India tribunal record, cited with attribution.
Carries the draft Visakhapatnam Green City Action Plan (prepared by IFC for GVMC, dated September 2024), which GVMC posted for public comment on 19 October 2024. Read for water treatment capacity (11 water treatment plants, 489 MLD), the 105 to 110 litres a person a day estimate, and the targets it proposes; those are proposals in a draft, not GVMC commitments, and some lines on its supply page do not agree with each other, so they are not repeated. Its 2026 notice of 120 wards is read for the ward gap. GVMC's documents for the Asian Development Bank, its disaster plan and its ward records are listed under their own groups below. GVMC publication, cited with attribution.
The project page's 23.44 TMC a year for 'in and around Visakhapatnam city including Vizag Steel Plant and drinking water facilities', shown on the allocations page beside the two other official figures: 24 TMC, from a 2004 order cited in GVMC's 2019 assessment for the Asian Development Bank, which we have not found, and 12 TMC (CGWB 2025). Government of India publication, cited with attribution.
Each clearance states the project's water quantity and source. Read for two of the allocations page's claims still to come and the facts card on data centres: three data-centre campuses cleared in April and June 2026 (239, 207 and 246 kilolitres a day of fresh water, 0.69 MLD together) and APIIC's data-centre park at Madhurawada (December 2024). Only the project, date, quantity and source are used; the clearances also name the proponent's contact person, which is not. Government of India publication, cited with attribution.
The company's About and Project Infrastructure pages: the 56 km pipeline from the Kateru pump house to the Yeleru canal, completed in December 2004, an allocated share of 7 TMC from Yeleru, and the GVMC and APIIC shareholding. Read for the supply chain on the dashboard, the allocations page and the rivers page, as the company's own description. We have not found a licence or terms of use on the site, which reads 'Copyright © 2026 VIWSCO'; cited with attribution.
A one-page note of August 2009: VIWSCO's drawals of 5 TMC from the Godavari and 5 TMC from Yeleru, its commissioning in December 2004, and its restructuring with GVMC and APIIC as promoters. Read for VIWSCO on the allocations page. APIIC publication, cited with attribution.
The register's Andhra Pradesh rows for the dams behind the city: Meghadrigedda (as 'Mehadrigedda', completed 1977, operated and maintained by the Water Resources Department), Gambhiramgedda (completed 1957, and the point used for its card), and Yeleru, Raiwada and Thatipudi, whose capacities agree with APWRIMS. Read for dam years and operators on the allocations page, the supply overview and the water-body cards, and on the origins page. Government of India publication, cited with attribution.
The department's old project pages, read from archived copies because the live site no longer resolves. The Yeleru page lists 5.00 TMC for industry 'for G.V.M.C or VIWSCo', the figure on GVMC's 2021 sheet, for the allocations page; the three pages give the dam years and the city's shares for the origins page. Andhra Pradesh government publication, cited with attribution.
The audit of the Polavaram project: the Purushottapatnam lift, which sends Godavari water into Yeleru reservoir and was completed by July 2018, and the Left Main Canal works' physical progress of 76.05 per cent in March 2023, the latest official figure we have found. It is also one of three documents giving 23.44 TMC for the city. Read for the dashboard's supply chain, the allocations page and the facts card on the Polavaram share. Government of India publication, cited with attribution.
The ministry's project note, which gives 23.44 TMC of drinking water supply 'to a population of 28.50 lakh in enroute 611 villages'. One of the three documents behind the 23.44 TMC on the allocations page and the facts card; it describes who the water serves differently from the Polavaram Project Authority and CAG, and both readings are shown. Government of India publication, cited with attribution.
Orders, clearances and courts
The central clearance of 20 March 2026: of the plant's water need, 1,03,968 cubic metres a day of fresh water (about 104 MLD) 'will be obtained from the Polavaram Left Main Canal', and the company 'shall obtain necessary permission from the Competent Authority'. Read for a claim still to come on the allocations page; we have not found a Water Resources Department allocation for it. Only the project, date, quantity and source are used. Government of India publication, cited with attribution.
The state's order on the AM/NS plant: industrial water at Rs 50 a kilolitre for 10 years, against the company's request for 6,600 cubic metres an hour from Polavaram. It sets a price, not a quantity, and the allocations page shows it that way. The state's government orders register allows reproduction free of charge, with the source acknowledged.
The Rs 1,638 crore sanction for the Purushottapatnam lift, which lifts 3,500 cusecs from the Godavari into the Polavaram Left Main Canal and names 23.44 TMC for Visakhapatnam city. Read with CAG's report for the lift on the allocations page. The state's government orders register allows reproduction free of charge, with the source acknowledged.
In a drought, water released to industry from a reservoir is cut pro rata to what the reservoir holds. The policy does not set priority between the city and industry on the Yeleru canal, and the allocations page lists that order among those we have not found. The state's government orders register allows reproduction free of charge, with the source acknowledged.
A special water rate of Rs 11.80 a kilolitre for 2012-13, rising 5% a year, for RINL, NTPC and GVMC under VIWSCO's concession agreement, which we have not found. On the allocations page's timeline. The state's government orders register allows reproduction free of charge, with the source acknowledged.
0.518 TMC from Yeleru reservoir to the Pampa reservoir project between 16 January and 15 May 2026, for a rabi crop. On the allocations page's timeline, as a draw on Yeleru beside the city's. The state's government orders register allows reproduction free of charge, with the source acknowledged.
Two orders in the commitments register, each read from a filing to the National Green Tribunal in OA 240/2025 (SZ): 12 November 2025, recording the Advocate General's timeline for a second sewage treatment plant at Appughar, and 11 March 2026, recording a Rs 272 crore contract and a last date fixed for March 2027. The court's joint committee's 2023 list of discharges into the Meghadrigedda surplus course is among the rivers page's events. Andhra Pradesh High Court judicial record, cited with attribution.
Weather and rainfall
Monthly rainfall at the grid cell 17.75 N, 83.25 E, inside GVMC on the Gopalapatnam and Simhachalam side, 1970 to 2025: the dashboard's rainfall chart and the facts page's normal, about 1,043 mm a year over 1970-2020. It is the grid point nearest the city centre, about 8 km away, and a land cell with a value in every year. The coastal cells 17.5 N 83.0 E, 17.5 N 83.25 E and 17.75 N 83.5 E are sea, with no value in any IMD year from 1991 to 2025, so a seaward pick would write zeros. Checked against IMD's observatories: over 1991-2020 this cell gives 1,053.4 mm, against a normal of 1,054.5 mm at Visakhapatnam airport (station 43149) and 1,113.1 mm at the Visakhapatnam observatory (station 43150); the AP planning department's district normal is 1,117.7 mm. IMD's months also take over from the provisional fill below as they are published. IMD data policy; cited with attribution.
The 1991-2020 normal rainfall at IMD's two Visakhapatnam observatories: 1,054.5 mm a year at the airport (station 43149) and 1,113.1 mm at the other (station 43150). The grid cell above is checked against them, and the facts page's rainfall card quotes both. We have not found a licence or terms of use on the table pages; Government of India publication, cited with attribution.
Daily rainfall from Open-Meteo's ERA5-family reanalysis fills the months after IMD's last published month. Those months are marked provisional and replaced by IMD's values as its refresh catches up. For Visakhapatnam it is used for rainfall only, not for temperature or other weather. CC BY 4.0: weather data by Open-Meteo.com.
Groundwater
The national groundwater assessment by mandal, for the groundwater map and facts: three editions on today's eleven units, 2023-2024 to 2025-2026, drawn on IN-GRES's own unit outlines. The measure is the stage of extraction: how much of the groundwater that can be drawn each year is drawn. The 2024-2025 edition matches CGWB's Dynamic Ground Water Resources of Andhra Pradesh 2025 (December 2025) for these mandals; the map colours the 2025-2026 edition, newer than CGWB's December 2025 report. The district was split in April 2022 and the single GVMC unit was split into seven urban mandals in 2023-24, so no trend is drawn across either redraw. We have not found terms of use, a disclaimer or a licence statement on the portal; a Government of India assessment, cited with attribution.
Automatic recorders read every day, shown on the groundwater map as monthly means of at least five good days: an older piezometer network, mostly from September 2022, and a network installed from December 2025 to January 2026. The archive carries some values forward and fills gaps with straight lines, and both are removed before averaging. Each well is named by the mandal its coordinates fall in on the portal's own mandal layer. We have not found a licence or terms of use on the portal, only a disclaimer and a footer reading 'All Rights Reserved', so it is carried as restricted and cited with attribution.
The AP Ground Water Department's hand-read monthly well record from May 2002, the long record behind the APWRIMS recorders; its telemetry is read only to check which recorder is which. Where a recorder and a hand-read well stand at the same site they are kept as separate series, not joined. District tags follow the pre-2022 lines, so wells are placed by their coordinates. Its copyright policy allows reproduction free of charge without specific permission, reproduced accurately and with the source acknowledged.
Water bodies and restoration
Every pond, tank, lake and reservoir polygon on the water-bodies map, in a box covering GVMC and about 5 km around it (1,088 polygons, 212 of them named, retrieved 14 October 2026), and the three channels on the rivers and flood pages: the Meghadri Gedda, the Gosthani and the Yeleru Left Main Canal. OpenStreetMap is a map, not a register; we have not found a gazetted lake register for the city, and small or seasonal tanks may be missing. Drains and wastewater polygons are left out. Open Database License (ODbL) 1.0.
The state's layer of 38,443 tank polygons, read only to confirm a water body's identity, name and location: Kondakarla Ava is tank MI012625, and the tank MI012862 near Gambheeram is a smaller tank 4 km east of the dam, not the reservoir. Its capacity field carries no unit and its fill field no date, so neither is used. We have not found a licence or terms of use on the portal, only a disclaimer and a footer reading 'All Rights Reserved', so it is carried as restricted and cited with attribution.
The status list as on 5 March 2026: 44 GVMC rows worth Rs 825.55 crore, among them the 11 water-body projects on the lake-restoration page (Rs 23.31 crore, 10 completed) and the water and sewerage projects in the commitments register. MoHUA's live list, checked on 14 October 2026, has no Thungalam row and lists instead a Rs 1.27 crore 'Thimmaraju Tank' project, which appears to be the same project under a new name; MoHUA's live list has no project IDs and does not say so. A new edition is a new file name. Government of India publication, cited with attribution.
The initial environmental examination for a 3 MWp floating solar plant on the reservoir: constructed 1977, commissioned 1979, and a reservoir surface it describes as owned by GVMC, set beside CWC's register, which lists the Water Resources Department as operator. Read for the Meghadrigedda card, the allocations page, the supply overview and the origins page. GVMC publication, cited with attribution.
The two documents for the Mudasarlova lake development project, of September and October 2018. The examination gives 1.5 million cubic metres as the original capacity and 0.75 million at full tank level after siltation, with no survey or date behind either, so its capacity is left blank. Read for the Mudasarlova card. GVMC publication, cited with attribution.
Its Table 4 holds the reservoir drawal figures APPCB repeated to the NGT in 2021, which is why the rivers page does not date them to 2021, and it quotes a 2004 order for 24 TMC of Polavaram water, which we have not found. Read for the rivers page's events, the allocations page and the water-body cards; the origins page also cites its March 2016 draft. GVMC publication, cited with attribution.
The portal's lists of Ramsar sites and wetlands of international significance, checked on 14 October 2026 for the Kondakarla Ava card: the lake is on neither list. Its copyright policy allows reproduction free of charge, with the source acknowledged.
Rivers and pollution
The readings on the rivers page and the water-body cards. CPCB's 2019-2024 tables list no river-type station on any Visakhapatnam or Anakapalli river. On the city's three channels it samples the Meghadrigedda reservoir and a tidal creek by the Naval Dockyard (Meghadri Gedda), Thatipudi reservoir (Gosthani) and the Yeleru canal near Pharma City at Parawada; its other stations here are one lake (Kondakarla Ava), the sea and groundwater wells. CPCB's 'Gostani' stations and polluted stretch are in West Godavari district and are not used. Also read: CPCB's polluted river stretch lists of 2018, 2022 and 2025, none of whose Andhra Pradesh stretches is in Visakhapatnam. CPCB's website policy allows download and printing without prior permission and makes any other use subject to CPCB's approval.
Monthly samples where sewage channels and geddas meet the sea, with annual averages from 2010-11 to 2024-25, quoted in the rivers page's notes and on the facts page: for example 0.7 mg/L of dissolved oxygen on average in 2024-25 where the 'lavender canal' sewage meets the sea at the harbour. We have not found published coordinates for these stations, so they are quoted, never mapped. APPCB publication; its site footer reads 'All Rights Reserved'. Cited with attribution.
APPCB's own regulatory documents for the two common effluent treatment plants: the Visakha Pharmacity plant's consent order of 7 April 2025, Directions Order 692 of 14 December 2023 for the Atchutapuram plant, and committee minutes from 2025. Read for the pharma city plant's marker on the rivers page and the two plants' compliance record. APPCB publication; its site footer reads 'All Rights Reserved'. Cited with attribution.
Form-V statements and six-monthly compliance reports: the steel plant's marine outfall marker and effluent figures on the rivers page, and the claims of the steel plant, HPCL and Gangavaram Port on the allocations page. Also NTPC Simhadri's 2021 compliance report for its Yeleru draw on the rivers page, and RINL's 2018 filings with MoEFCC for the Kanithi balancing reservoir on the water-body cards and the origins page. Cover letters say 'within norms' while the attached tables show exceedances; both are carried. Statutory filings the operators post publicly, cited with attribution.
Read for the rivers page's events on the Meghadri Gedda and the origins page: GVMC stopping releases from the Meghadrigedda reservoir on 8 May 2020, and two sets of reservoir samples, one detecting styrene and one not, both shown. Read from a copy on a third-party site (safteng.net). Government of Andhra Pradesh committee report, cited with attribution.
Flood and civic infrastructure
After-event reports for the flood-risk page's storms (Hudhud 2014, the deep depression of October 2020, Gulab 2021, Michaung 2023 and Montha 2025): date, track, the 24-hour rainfall at IMD's Visakhapatnam stations, and storm surge. The 2024 heavy-rain row is not IMD's: it is the state government's district figure from its flood memorandum, and the page credits it so. IMD's reports may not be reproduced in part or full without permission, so only event figures are cited and no report is reproduced.
The source of the flood page's 2024 row: 129 mm across Visakhapatnam district over the six days from 29 August, a district average rather than a station reading, and the district's damage figures. We have not found a licence or terms of use on the APSDMA site, whose footer reads 'All rights reserved'; cited with attribution.
Listed on the flood page for the state's alerts and post-event memoranda. Its disaster history page gives Hudhud's death toll as 61, which the flood page sets beside the 46 in the state's table in IMD's report, and it hosts the 2018 district disaster plan and the state GIS layer of drains described below. We have not found a licence or terms of use on the site, whose footer reads 'All rights reserved'; cited with attribution.
The only modelled flood maps we have found that cover the city: the state planning society's storm-surge and river-flood maps, reproduced at region scale (PDF pp47-48), with the 11 villages in the Meghadrigedda reservoir's floodable zone and GVMC's 2,714.5 km of storm-water drains. Also read for the Kondakarla Ava card. A draft, so its proposals are not shown as commitments. Andhra Pradesh government plan document, cited with attribution.
Listed on the flood page for ocean-state and storm-surge forecasts during a cyclone. No figure on any page comes from it. Its disclaimer does not permit reproducing its material for commercial purposes without permission, so nothing is reproduced.
The link opens Volume I, whose Table 5-2 lists inundation-prone areas in 14 of GVMC's earlier 72 wards. Volume II, a separate file on GVMC's site (cdmp2.pdf), maps storm-water drainage in 14 zones at city scale, as a picture. The flood page names both and maps neither, because the ward numbers are no longer in use. GVMC publication, cited with attribution.
Its list of flood-prone areas, in an older division numbering, its count of 22 major geddas and about 1,400 km of storm-water drains, and its proposal of a drainage master plan, which we have not found published. Also read for source distances measured from the town service reservoir and for supply figures of about 2006 on the origins page. GVMC plan, cited with attribution.
River flooding seen from satellite, 2000 to 2020. The flood page cites it only to say that it states local and flash floods may not be captured. Government of India publication, cited with attribution.
The appraisal gives Hudhud's death toll as 61, shown on the flood page beside the state's 46, and describes the US$370 million recovery project on the origins page; the completion report counts the consumer accounts on the underground power network. World Bank project documents, cited with attribution.
We have not found a GVMC flood map, a current list of waterlogging spots, GVMC drain network data or a gedda register, so this group has no civic layer. A state GIS layer maps the drains of 2019 without names or capacities, with no licence statement we have found, so the flood page mentions it and does not draw it.
Satellite and remote sensing
With Sentinel-2, the images behind our own shoreline measurement: a water index traced on December-to-May median images in ten epochs from 1990 to 2026, measured every 100 m along transects from Appikonda to Pedda Nagamayyapalem, just north of Bheemunipatnam, with a weighted trend for each transect. USGS public domain, with courtesy attribution.
The recent epochs of the same shoreline measurement. No tidal correction is applied; the spring tide range here is 1.43 m. Copernicus free and open data, attribution required.
The official record the shoreline page sets our measurement beside: the 1990-2016 national assessment (Kankara, Ramana Murthy and Rajeevan, 2018), and the stretch classes in the Ministry of Earth Sciences' Lok Sabha reply of 11 December 2024, which draws on it. That reply classes four stretches here as Low Erosion: Yarada, RK Beach, 'Mangamari Peta to Bheemili' and 'Bheemunipatnam to Pedanagamayyapalem'. The other stretches are not on its erosion-prone list. Government of India (MoES and NCCR) publication, cited with attribution.
The design tide levels behind the shoreline page's 1.43 m spring tide range (from the Survey of India's 2017 tide tables), and the guidelines' account of the port moving sand onto RK Beach. Read from the Kerala Irrigation Department's copy. Government of India publication, cited with attribution.
A Rs 20 crore sand-trap dredging for beach nourishment, inaugurated on 12 February 2023, which is why the shoreline page treats RK Beach as a managed beach. Government of India publication, cited with attribution.
The ground-height bands on the water-bodies map: 30 m elevation with buildings and trees removed, grouped into seven bands from 0-2 m to 100 m and above. FABDEM stores the sea here as exactly 0 m, so those pixels are removed; we have not found published GVMC boundary geometry, so there is no city-boundary clip and the bands also cover land just outside GVMC. Bands, not spot heights. CC BY-NC-SA 4.0, non-commercial and share-alike; derived from Copernicus GLO-30, so its attribution to DLR and Airbus passes down.
History and research
The official record of the town's first water works, handed over in May 1903 with 25 million cubic feet of storage, in the reservoir known today as Mudasarlova. Read for the water-body cards and the origins page, with the Imperial Gazetteer for the 1891 and 1901 population. Public domain (published 1907 and 1908); read from the Internet Archive's scans and cited with attribution.
Populations and town amenities for the origins page, from the handbooks of 1951, 1971, 1981, 2001 and 2011. Population figures on the page are census figures only. Government of India publication, cited with attribution.
Answers to questions from 1986 to 1995 and committee reports from 1965 to 2010: the 1989 city scheme to draw from the Yeleru and Raiwada canals and the Yeleru water reaching the steel plant in 1991, on the allocations page, and the port, steel plant and Yeleru history on the origins page. Parliament of India record, cited with attribution.
A 1953 answer on the military's water works at the Gosthani and the Meghadri Gedda, and the 1964 budget speech on reservoirs for the steel plant, both quoted on the origins page. Assembly records, cited with attribution.
Dates from each institution's own pages: the inner harbour built at the mouth of the Meghadri Gedda in 1927-33, on the rivers and origins pages; GVMC's formation in 2005; the refinery, fertiliser plant and steel plant years; and the data-centre groundbreaking of 28 April 2026, whose releases state no cooling-water source. The same operators' statutory filings are listed under Rivers and pollution. Cited with attribution.
The announcement of Google's AI hub at Visakhapatnam, which the origins page notes gives no water figure. Government of India press release, cited with attribution.
How water supply was kept going after Hudhud, with generators at the head water works and power back within two to seven days in urban areas, and the death toll of 61, for the origins page. Government of India publication, cited with attribution.
Research, never presented as official monitoring: the PFAS study behind the facts page's research card, a Chennai shoreline study whose error figures weight our transects, and two groundwater studies on the origins page, read from their abstracts. Academic publications, cited with attribution.
Base geometry
The coastline the shoreline transects are laid along, and the breakwaters, piers and river mouths that mark a transect low confidence. The two port basins are bridged across rather than measured. The mandal outlines on the groundwater map are IN-GRES's own, listed above; we have not found ward geometry that can be published. Open Database License (ODbL) 1.0.
GVMC's 98-ward map is a CAD plot with no coordinates, and its ward list is what the zone field of the Esri India tracing in circulation disagrees with for 94 of 98 wards. With the 2026 notice of 120 wards, they are why no ward page is shown. GVMC publication; its site terms are a disclaimer, not a licence. Cited with attribution, and the map is not redrawn.
The order that set GVMC's 98 wards, on which the council elected in March 2021 sits; read for the ward count. The state's government orders register allows reproduction free of charge, with the source acknowledged.
Data quality & limitations
How we classify river health
CPCB publishes two parallel river-water-quality classification systems, and they don't always agree. Knowing which one we use - and why - matters for reading our river status badges honestly.
Designated Best-Use classes (A-E)
Computed from current dissolved-oxygen, BOD and coliform thresholds at each monitoring station. Updates every reading. Class A = drinking with disinfection only; Class B = outdoor bathing; Class C = drinking with conventional treatment; Class D = fisheries/wildlife; Class E = irrigation only. Below E = practically dead.
Polluted River Stretch (PRS) Priority I-V
A historical, multi-year stretch-level designation reflecting cumulative pollution. Slow to update; once a river stretch is on the Priority list it tends to stay there even if recent readings improve. Priority I = worst (BOD > 30 mg/L sustained); Priority V = least bad of the polluted stretches.
Our status badges ("dead", "severely degraded", "degraded", "stressed", "healthy") are computed from current readings via the Designated Best-Use thresholds - not from the PRS Priority list. We take the worst classification across a river's monitored stations and surface that as the river-level status. For Visakhapatnam, CPCB's 2019-2024 tables list no river-type station, so the badges are read from the stations CPCB does sample on the city's channels: two reservoirs, a creek and a canal.
For Visakhapatnam, CPCB's tables give no river stretch to set a badge against. Its 2019-2024 tables list no river-type station in the city, and none of the stretches on its polluted river lists of 2018, 2022 and 2025 is here (the ‘Gostani’ on those lists is in West Godavari district). The rivers page still colours each channel, using the thresholds above on the latest year at the stations along it. The Meghadri Gedda reads severely degraded because of the tidal creek by the Naval Dockyard (lowest dissolved oxygen 2.4 mg/L in 2024); the Meghadrigedda reservoir on the same channel would read stressed on its own. The Gosthani, read at Thatipudi reservoir upstream in Vizianagaram district, and the Yeleru canal, read near Pharma City at Parawada, read stressed. Each channel's CPCB class tile says ‘Not assessed (no river station in CPCB's 2019-2024 tables)’. Read the badges as the state of the water at those stations, not of a river.
Status thresholds follow CPCB's published Designated Best-Use criteria; readings are from CPCB's NWMP annual tables for reservoirs, canals and coastal water, 2019-2024.
Data quality for Visakhapatnam
What to keep in mind about the figures behind the Visakhapatnam pages, source by source.
GVMC's only supply sheet covers one day, and two of its figures are fixed ratios
GVMC's source-by-source sheet of 11 March 2021, annexed to its report in NGT OA 14/2021 (SZ), is the only one from GVMC itself that we have found. Its net supply is the total less a flat 15%, and its per-person figure divides the household share by an assumed 2.30 million people; neither is a measurement, and neither is used. The canal split prints no unit and is read as million gallons a day, like the sheet's other columns. Its bore rows give the same figure for normal and present supply, so they read as set figures, not readings.
The 380 MLD supply table has not changed since June 2022
CGWB's 2025 plan reports GVMC's supply by source as GVMC gave it: nine sources, about 380 MLD. The same quantities and total are in CGWB's June 2022 and November 2023 reports, so the dashboard shows them as GVMC's figures carried forward, not a current measurement. The rows add to 380 to 402 MLD depending on which end of the Godavari and Yeleru range (182 to 204 MLD) is taken; the printed 380 matches the low end. The ‘490 MLD demand against 391 MLD supply’ row is the 2022 report's figure for 2020.
APWRIMS groundwater readings carry values forward
The APWRIMS archive carried one value forward for 195 days at Visalakshinagar and fills gaps of up to 30 days with straight lines, and one station can carry different names in different lists. Repeated values, straight-line fills, spikes, readings more than 20 m from a well's usual level and stuck sensors are removed before monthly means are taken. Recorders whose identity we could not settle are flagged or left out.
Two redraws break the groundwater series
Andhra Pradesh divided Visakhapatnam district in April 2022, and IN-GRES kept one district id across the change: its district rows cover 8,547.7 km2 in 2016-2017 and 11,050.6 km2 in 2019-2020, then 1,001.7 km2 from 2021-2022. The single 513.8 km2 GVMC assessment unit became seven urban mandals in 2023-24. So the map starts at 2023-2024 and no trend is drawn across either redraw. In the 2025-2026 edition every mandal's rainfall repeats its 2024-2025 figure, and revised recharge and industrial inputs move four urban mandals by more than 15 points; each moved unit says so on the map.
IN-GRES's urban units also do not follow the state's mandal lines. Its Mulagada lies mostly in the mandal the state's own boundary map calls Pedagantyada, and its Peddagantyada is a strip around Gollalapalem and Gopalapatnam, so the two names look transposed. Figures are shown under IN-GRES's names, and each well names the state mandal its coordinates fall in. The unit outlines place the figures but are approximate: together they cover 973.8 km2 against the 1,001.7 km2 assessed.
Shoreline rates are our own measurement
We have not found a published rate for each point on this coast, so the rates are our satellite measurement, with no tidal correction (the spring tide range here is 1.43 m) and weighted with error figures borrowed from a published study of another coast, not re-estimated here. 55 transects are drawn as low confidence and left out of the counts: 48 that sit on the Gangavaram port's built shore, cross the Rushikonda breakwater, sit at the Dolphin's Nose headland or beside river and stream mouths, or show large jumps in their measured waterline, each with its reason; and 7 that the pipeline's own test marks low because their recent positions scatter or jump.
Ground height: the sea is zero and there is no city outline
FABDEM stores the sea here as exactly 0 m, so those pixels are removed rather than drawn as the lowest band. We have not found published GVMC boundary geometry, so the bands cover the box around the city, land just outside GVMC included. They are bands, not spot heights.
Thatipudi and Raiwada: one of two level records kept, late 2021 to March 2024
From late 2021 to 7 March 2024 the APWRIMS archive carries two level records for Thatipudi and Raiwada, and the published reading moves between them from day to day. Thatipudi's two sit about 4.4 m apart on most days; Raiwada's mostly agree but part by up to 3.5 m on some days in 2023. Since 8 March 2024 the feed has carried only one of them, and we keep that one from its first reading: 7 September 2021 for Thatipudi and 13 December 2021 for Raiwada. Days that carry only the other record are left out, so from those dates to 7 March 2024 Thatipudi has a reading on 169 of 913 days and Raiwada on 304 of 816, and the history chart has gaps there. Earlier readings are mostly from the other record. At Thatipudi that shows as a step: its live storage goes from 0.77 TMC on 6 September 2021 to 1.65 TMC the next day, where the record changes, so its years before then are not directly comparable with the years after. We have not found which record the department treats as correct; the one we keep is the one it publishes today. Yeleru's and Meghadrigedda's two records agree to within half a metre, and their readings are shown as published.
Known Limitations
- There is no days-of-water-left figure: five of the six reservoirs are shared with irrigation, industry or other users, two lifts add Godavari water, and we have not found a measured figure for the city's daily use.
- Supply figures come from documents, not meters: the 380 MLD table is GVMC's, carried unchanged in CGWB's reports since June 2022, and the canal split is one day, 11 March 2021. We have not found a measured non-revenue water figure for GVMC, so no losses figure is shown.
- Reservoir readings are daily and published by the state for four of the six reservoirs; Mudasarlova and Gambheeram are not in the feed.
- Groundwater is assessed by mandal, not by ward, and no trend crosses the 2022 district split or the 2023-24 unit split.
- CPCB's 2019-2024 tables list no river-type station in the city, so river badges describe two reservoirs, a creek and a canal.
- The flood record is a list of cyclones, depressions and one spell of heavy rain, each from its own source. We have not found a GVMC flood map, a current list of waterlogging spots or a drainage master plan.
- We have not found ward geometry that can be published, so nothing is shown by ward.
- Shoreline rates are our satellite measurement without tidal correction; the official record is NCCR's stretch classes.
- Ground-height bands come from FABDEM with the sea removed and no city-boundary clip. FABDEM's licence is non-commercial and share-alike.
- Utility-published feeds can be stale (weekends, holidays, portal outages). Each surface shows a freshness indicator.
- The pages are in English. Telugu is planned and has not been translated yet.
Where official figures disagree, all of them are shown
Official documents give different figures for the same thing. The pages show each figure as published, with its source, rather than choosing one or correcting it to make the sums agree.
- Distance to the sources. Raiwada is 60 km in CGWB's 2025 plan, 56 km in its 2022 and 2023 reports and 76 km on GVMC's 2021 sheet; Thatipudi is 56 km (CGWB 2025 and GVMC 2021) or 60 km (CGWB 2022 and 2023); Meghadrigedda is 5 km (CGWB 2025) or 20 km (GVMC 2021).
- The Godavari and Yeleru route. 190 km is CGWB's distance to the source. The canal is about 150 km in the draft Green City Action Plan; GVMC's 2021 sheet prints 156 km against Yeleru reservoir and 56 km against the Kateru pump house, which is the length of VIWSCO's pipeline from Kateru to the canal.
- Total supply. About 380 MLD available (CGWB 2025, Table 9.1), 391 MLD in another table of the same report, and about 395.5 MLD in the draft Green City Action Plan; 345.5 MLD that GVMC was able to draw (CGWB 2022); 334.03 MLD present supply (Chief Secretary's report, NGT Principal Bench OA 606/2018, 10.07.2024, Annexure-VI, p33).
- The city's Polavaram share. 23.44 TMC a year (Polavaram Project Authority), 24 TMC (a 2004 order cited in GVMC's 2019 assessment for the Asian Development Bank; we have not found the order) and 12 TMC (CGWB 2025).
- Meghadrigedda's dam. CWC's National Register of Large Dams (2019, as ‘Mehadrigedda’, AP01MH0042) gives completion in 1977 and lists it as operated and maintained by the Water Resources Department. GVMC's 2018 assessment says constructed 1977, commissioned 1979, and describes the reservoir surface as owned by GVMC. CGWB lists the source from 1976.
- Sewage generated. 248 MLD in the state's February 2026 affidavit to the NGT Principal Bench; 215.05 MLD in GVMC's own March 2024 affidavit in NGT OA 40/2024 (SZ), estimated as 85% of supply. The later state figure is used.
- Water losses. The draft Green City Action Plan gives nearly 30 percent on one page and 35 percent on another, without saying how either was measured, so no losses figure is shown.
- Groundwater in the supply. 2.75 million gallons a day from bores on GVMC's 2021 sheet, 30 percent of the city's needs in the draft plan, and 17.37 percent of supply in CGWB's 2025 plan. We have not found a measured figure.
- Mudasarlova's capacity. GVMC's 2018 assessment gives 1.5 million cubic metres, which it calls the original capacity, and 0.75 million cubic metres at full tank level after siltation, with no survey or date behind it, so the capacity is left blank.
About the project
Disclaimer
Not an official government tool. Neer Vazhvu is an independent, open-source project. It is not affiliated with, endorsed by, or connected to any water utility, municipal corporation, pollution control body, or other government agency whose published data it cites.
Informational purposes only. All data, estimates, and forecasts are provided “as is” for general awareness. Always refer to official advisories from your city's water utility and disaster-management authorities for critical decisions.
No personal data collected. Neer Vazhvu does not collect, store, or process any personal information. There are no user accounts, cookies, or analytics trackers.
Open Source
Neer Vazhvu is fully open source. The code, data pipeline, and methodology are transparent and available on GitHub. Contributions, bug reports, and data corrections are welcome.
View on GitHubSupport this project
Neer Vazhvu is free and open source. If you find it useful, consider supporting us on Patreon to help cover satellite data, hosting, and API costs.
Support on Patreon