Why Bursa in 2026 Is a Different Design Problem Than Istanbul or Ankara
Bursa draws on the same Marmara/Susurluk basin as Istanbul, where ISKI already imposes seasonal industrial cuts between July and September — and the Susurluk sub-basin carries the same drought profile (Ecologix, 2025). A 2026 Bursa site cannot assume unmetered municipal supply is permanent infrastructure, and a Turkish data center discharging 180,000–220,000 L/day of cooling-tower blowdown into a Marmara tributary will face the same seasonal supply pressure that has pushed ISKI toward multi-tier industrial tariffs.
The city's Organized Industrial Zones — NOSAB, BOSAB, GOSB — already operate shared WWTPs and carry an industrial discharge tariff history that the data center's permit will inherit, not a greenfield sewer negotiation. BUSKI (Bursa Water and Sewerage Administration) industrial discharge tariffs in 2025–2026 have moved toward multi-tier structures that penalize high-volume users, mirroring the ISKI trajectory (Ecologix, 2025). For a 20 MW site, that tariff trajectory alone is enough to push the math toward on-site reuse rather than a direct-sewer design.
The regulatory anchors a Bursa facility must design against are the SKKY (Su Kirliliği Kontrolü Yönetmeliği) Table 16 industry limits, BUSKI's connection and pretreatment rules, and the EWC-aligned waste-code family 11 01 09* that covers cooling blowdown in Turkey as in the EU. Verify against the latest consolidated SKKY text and the BUSKI tariff schedule in effect at the time of bid before any design freeze — both have moved within the last 12 months.
How Much Blowdown a Bursa Data Center Actually Produces
75–90% of global data centers still rely on water-based cooling (KETOS, 2025). At an industry-average WUE of 1.8 L/kWh, a 20 MW Bursa site needs roughly 720,000 L/day of makeup water; a 100 MW AI-ready campus at PUE 1.2 needs about 3,600,000 L/day (Ecologix, 2025). Because 70–80% of that makeup evaporates in an open cooling tower, only 20–30% exits as blowdown — but that fraction is the stream a Turkish engineer has to clarify, filter, meter, and either reuse or permit.
Blowdown volume follows B = E / (CoC − 1): at CoC 4 blowdown equals 25% of makeup, at CoC 6 it drops to 20% (Genesis Water Tech, 2025; Ecologix, 2025). Worked Bursa numbers: a 20 MW site at CoC 5–6 cycles 180,000–220,000 L/day of blowdown; a 100 MW campus cycles 900,000–1,100,000 L/day. Add ~0.02% of circulation as drift loss to the makeup figure (Ecologix, 2025).
Closed-loop liquid cooling reduces consumption to 5–10% of withdrawal, but most 2026 Bursa builds still specify hybrid air/liquid or open towers — blowdown is the design driver, not the cooling-architecture choice. AI-dense racks now exceed 50 kW per rack and pull PUE and WUE upward, so oversize equalization by 20–30% as cheap insurance against the next rack generation.
| Site size | Makeup (L/day) | Blowdown at CoC 5 (L/day) | Blowdown at CoC 6 (L/day) |
|---|---|---|---|
| 20 MW (WUE 1.8) | ~720,000 | ~180,000 | ~144,000 |
| 50 MW (PUE 1.3) | ~1,800,000 | ~450,000 | ~360,000 |
| 100 MW AI-ready (PUE 1.2) | ~3,600,000 | ~900,000 | ~720,000 |
What Is Actually in Bursa Cooling-Tower Blowdown
With Marmara municipal supply at roughly 250–500 ppm TDS and CoC 5–6, expected blowdown TDS is 1,500–3,000 ppm (Ecologix, 2025); pH 7.5–9.0, silica 20–80 mg/L as SiO₂, temperature 30–40°C. Residual biocides, phosphonates, and trace Cu/Zn from corrosion-inhibitor breakdown are typical — older chromate or high-phosphate programs can push metals and phosphorus above SKKY Table 16 limits (Ecologix, 2025).
Suspended solids from corrosion products, biofilm fragments, and airborne particulates run 10–50 mg/L even with basin filtration (Genesis Water Tech, 2025). Three species cap conventional RO recovery on this stream: silica, calcium carbonate, and calcium sulfate — the same scale-formers that limit brackish RO to 75–80% recovery before flux decays and cleaning frequency spikes (IDE Tech, 2025).
| Parameter | Expected range at CoC 5–6 | Driver / concern |
|---|---|---|
| TDS | 1,500–3,000 ppm | SKKY Table 16 sewer acceptance cap |
| pH | 7.5–9.0 | RO membrane compatibility, metals solubility |
| Silica (as SiO₂) | 20–80 mg/L | RO recovery ceiling |
| Temperature | 30–40°C | Receiving-water ΔT limit, RO flux |
| Suspended solids | 10–50 mg/L | UF loading, SDI |
| Cu / Zn (trace) | From corrosion-inhibitor breakdown | SKKY metals limits |
The Defensible 2026 Process Train for a Bursa Data Center
The defensible 2026 train for a Bursa blowdown stream runs equalization → DAF or lamella clarification → multi-media filtration → UF → RO, with reject branched either to brine concentration/crystallization (ZLD) or to controlled sewer per BUSKI permit (Ecologix, 2025).
Step 1 — Equalization basin. 24–48 hours of HRT to buffer TDS, pH, and temperature swings so downstream equipment sees a stable feed. Mix with slow-speed paddles; avoid aeration to prevent CO₂ stripping that would shift calcium carbonate equilibrium. Step 2 — DAF or lamella clarification. Removes suspended solids, oils, and flocked metal hydroxides before they load the filters. The DAF clarification step for cooling tower blowdown (4–300 m³/h DAF range) handles variable solids well; lamella plates at 20–40 m³/m²·h are an alternative when footprint is tight. For DAF sizing in detail — air-to-solid ratio, recycle rates, retention times — see the DAF design parameters 2026 guide. Step 3 — Multi-media filtration. Anthracite/sand/garnet polishing to SDI <5 ahead of UF; the multi-media filtration polishing stage protects membrane life. Step 4 — UF as RO pretreatment. 0.03 µm PVDF membranes rated for up to 300 NTU feed and delivering consistent SDI <3; the UF pretreatment ahead of the RO typically operates in the 2,000–40,000 L/h range per skid. Step 5 — RO for blowdown-to-makeup reuse. The RO unit for blowdown-to-cooling-tower reuse is sized at 75–80% recovery as the conservative default; pushing past 90% requires controlled-salt-precipitation upstream (IDE Tech, 2025). Step 6 — Reject management. RO concentrate either feeds a brine concentrator + crystallizer (ZLD) or is discharged under a BUSKI permit when salinity, temperature, and metals are within acceptance limits. Across all wet stages, the PLC-controlled chemical dosing skid handles pH correction, antiscalant, and biocide feed to keep recovery and flux on target.
The same process train scales linearly with IT load, but the equipment room footprint, equalization volume, and reject-handling complexity change step-wise. At CoC 5–6, a 20 MW Bursa site needs ~720,000 L/day of makeup and ~180,000–220,000 L/day of blowdown — this fits a single skid-mounted train (one DAF, one multi-media, one UF, one RO) in a single equipment room with the chemical dosing skid alongside. The 20 MW case is the right place to start the BOM. A 100 MW AI-ready campus needs ~3,600,000 L/day of makeup and ~900,000–1,100,000 L/day of blowdown, which pushes the design into parallel trains, dedicated equalization sized for 48 hours, and serious reject handling.
Reuse, Discharge, or ZLD: The 2026 Bursa Decision
Direct discharge fees in water-stressed regions run $5–15 per thousand gallons (Genesis Water Tech, 2025). For a 20 MW Bursa site cycling 180,000–220,000 L/day of blowdown, that is a real operating line, not a rounding error — and BUSKI's tiered industrial tariff in 2025–2026 has moved in the same direction as ISKI's, penalizing the highest-volume users (Ecologix, 2025).
On-site treatment becomes mandatory when any of three conditions is met (Ecologix, 2025): effluent exceeds municipal limits; the site is in a water-scarce area committing to ZLD or ≥70% recovery; the local WWTP cannot accept the daily volume. Bursa NOSAB/BOSAB tenants in 2026 routinely hit conditions 2 and 3 — the Susurluk sub-basin's seasonal supply profile and the shared industrial-park WWTP load acceptance both factor in. Blowdown-to-cooling-tower reuse at 70–80% RO recovery is the lower-CAPEX, lower-parasitic-load default for most 2026 Bursa builds; ZLD is reserved for hyperscaler mandates, sub-basin moratoria, or sites with a corporate water-positive target. ZLD adds 2–3× the CAPEX and a meaningful parasitic load through brine concentrators and crystallizers — only justify it against alternative water supplies or discharge prohibitions, not against reuse-with-permit (Ecologix, 2025).
| Path | When it fits a Bursa 2026 build | CAPEX posture | Operational risk |
|---|---|---|---|
| Direct sewer (BUSKI permit only) | Blowdown within SKKY Table 16 / BUSKI connection limits; low-tariff tier | Lowest | Tariff escalation, moratorium risk in Susurluk sub-basin dry season |
| On-site reuse (RO 70–80% recovery) | Default for 20–100 MW Bursa builds; co-located with cooling tower | Moderate | Membrane scaling if antiscalant program drifts |
| Partial ZLD (RO + brine concentrator) | Sites with >80% recovery mandate but no full crystallizer need | High | Parasitic load, brine disposal logistics |
| Full ZLD (crystallizer) | Hyperscaler water-positive mandate, sub-basin moratorium, no sewer option | Highest (2–3× reuse train) | Crystallizer uptime, solid waste handling |
Bursa-Overlay Economics: What This Costs in 2026
Genesis Water Tech (2025) published a worked 15 MW water-stressed example at 60% blowdown recovery: $200,000 CAPEX, 6.7-year simple payback on water alone. Once avoided discharge fees, energy recovery from concentrate-side pressure exchangers, and corporate sustainability targets are included, payback typically compresses to 3–5 years. A 50,000 GPD RO skid treating blowdown runs $250,000–$500,000 installed, with operating cost $1.50–$3.00 per thousand gallons treated (Genesis Water Tech, 2025).
For a Bursa site, the overlay is BUSKI industrial tariffs, Turkish-lira-denominated discharge fees, and any EU-tenant CSRD-style disclosure pressure. The CAPEX line for DAF + multi-media + UF + RO plus the chemical dosing skid is an insurance policy with measurable monetary value, not a sustainability add-on — meeting SKKY Table 16 eliminates permit risk and the shutdown exposure that comes with a single non-compliance event. For a Bursa project, request the current BUSKI connection rules, the latest SKKY Table 16 industry limit values, the site's specific industrial tariff tier, and the receiving WWTP's load acceptance — site-specific design values must be verified against current permits and final equipment proposals. For peer reference points where the same six-step train applies, the Rawalpindi data center blowdown treatment guide covers an arid-basin case, and the Ibadan data center blowdown treatment guide shows a tropical-municipal discharge context.
Frequently Asked Questions
How much does a blowdown treatment train cost for a 20 MW Bursa data center in 2026?
For a 20 MW Bursa site producing 180,000–220,000 L/day of blowdown, a single skid-mounted DAF + multi-media + UF + RO train with a PLC-controlled chemical dosing skid is the typical BOM starting point. Genesis Water Tech (2025) puts a 50,000 GPD RO skid at $250,000–$500,000 installed, and a 15 MW water-stressed example at $200,000 CAPEX with a 6.7-year simple payback on water alone, compressing to 3–5 years once avoided discharge fees and energy recovery are included. Request itemized vendor quotes for DAF, multi-media, UF, RO, and the dosing skid against your specific BUSKI tariff tier and the project's discharge-fee line — site-specific values must be verified against current permits and final proposals.
How do I select a treatment-train supplier for a Bursa 2026 build?
Shortlist suppliers who can document prior deliveries of the full six-step train (equalization, DAF or lamella, multi-media, UF, RO, chemical dosing) to a water-stressed Marmara-basin or comparable Mediterranean site, and who will provide performance guarantees tied to the SKKY Table 16 limits and BUSKI's local connection rules — not generic brochure numbers. Require influent testing on a feed sample from the Bursa supply, a written recovery and flux warranty on the RO unit at the design cycles of concentration, and a confirmation that the antiscalant program is compatible with the silica and calcium-sulfate scaling envelope. Compliance risk in 2026 is more about permit alignment than hardware — pick a vendor who will sign off on the BUSKI permit package, not just the equipment.
What influent parameters should a Bursa engineer design the train against?
Design against blowdown at 1,500–3,000 ppm TDS, pH 7.5–9.0, silica 20–80 mg/L as SiO₂, temperature 30–40°C, suspended solids 10–50 mg/L, plus residual biocides, phosphonates, and trace Cu/Zn from corrosion-inhibitor breakdown (Ecologix, 2025; Genesis Water Tech, 2025). The three scale-formers that cap conventional RO recovery are silica, calcium carbonate, and calcium sulfate (IDE Tech, 2025) — these define the antiscalant program and the recovery ceiling.
When does a Bursa site need ZLD instead of RO reuse?
ZLD is justified when BUSKI imposes a sewer moratorium, when the site sits in a Susurluk sub-basin segment with explicit ZLD requirements, or when the operator has a corporate water-positive mandate that makes any liquid discharge unacceptable (Ecologix, 2025). For most 2026 Bursa builds, blowdown-to-cooling-tower reuse at 70–80% RO recovery is the lower-CAPEX, lower-parasitic-load default — ZLD adds 2–3× the CAPEX and a meaningful parasitic load through brine concentrators and crystallizers, so the justification has to be a permit or corporate constraint, not a sustainability slogan.
