H1 2026 Evaluation
Croatia Reservoir Transparency
D41Very Poor — Ranked #71 out of 167 countries
weight 30%
weight 20%
weight 15%
weight 13%
weight 10%
weight 8%
weight 5%
Primary source evaluated
DHMZ Hydrology Sector (hidro.dhz.hr) + Hrvatske vode Vodostaji portal
https://hidro.dhz.hr/Dimension breakdown
Data Availability
20% of total score
Croatia does not publish reservoir storage volumes or fill percentages as a standardised public product. DHMZ (hidro.dhz.hr) and Hrvatske vode (vodostaji.voda.hr / mvodostaji.voda.hr) publish daily water-level readings at ~384 automatic gauge stations across rivers, but the outputs are river-stage readings in cm, not reservoir storage (Mm³) or fill %. HEP, which operates all 28 major hydropower reservoirs including Peruća (566 Mm³), Lokvarsko, and Butoniga, does not publish operational storage data publicly — its internal INFOZOK environmental database is not accessible. The ENTSO-E Transparency Platform (16.1.D) theoretically covers the aggregate weekly reservoir filling rate for HOPS/Croatia, but Croatia's submissions to this dataset are sparse and unverified. No dedicated reservoir-fill dashboard exists at a national level. Drinking-water reservoirs (Butoniga, Valtura, Boljunčica) are reported by utilities only for water-quality compliance, not volumetric fill.
Technical Accessibility
15% of total score
DHMZ provides a documented, free XML feed (hidro.hr/hidro_bilten.xml) under the Open Licence of the Republic of Croatia, covering daily hydrological bulletins with no authentication required. A separate XML endpoint index is maintained at meteo.hr/proizvodi.php?section=podaci¶m=xml_korisnici. This represents a genuine programmatic access layer, though it covers river stages and flood indicators rather than reservoir volumes. The DHMZ HIS2000 web application (hidro.dhz.hr) offers an English-language daily report view but no machine-readable download of historical reservoir data. The ENTSO-E Transparency API (RESTful, documented) could in principle provide aggregate weekly reservoir filling for Croatia, but actual data coverage for HR is unreliable. The Hrvatske vode Geoportal (preglednik.voda.hr, tiles4.voda.hr/gis-web/) provides WMS/WFS spatial layers under INSPIRE standards, but these cover flood risk zones and water districts, not real-time storage volumes. No REST API specifically for reservoir fill % exists on data.gov.hr.
Coverage
30% of total score
v1.3.0 capacity-weighted with conservative estimation applied 2026-05-29. Croatia total national reservoir storage capacity ~3,500 hm³ dominated by Peruća (~566 hm³), Croatia's share of Buško Jezero (~782 hm³ total, shared with BiH), Dubrava (~88 hm³), Lokvarsko (~30 hm³), Bajer, Sabljaci, Kruščica (~24 hm³), Lepenica, Vrana, Butoniga (~19 hm³ drinking water), Valtura, Boljunčica, plus the Varaždin/Čakovec impoundments on the Drava cascade — ~12-15 reservoirs above 10 hm³. COVERED capacity ≈ 1,120 hm³ on a conservative basis via HEP partial annual reporting for strategic hydropower reservoirs + DHMZ hidro.dhz.hr daily XML feed for selected gauges + Hrvatske vode vodostaji.voda.hr daily readings + ENTSO-E 16.1.D aggregate when populated. coverage = round(100 × 1,120 / 3,500) = 32. The conservative downward revision from 40 reflects that HEP's INFOZOK environmental database remains internal, that the published metric is generally river-stage rather than reservoir storage volume, and that no fill-percentage product exists for the long tail of medium reservoirs and drinking-water storage.
Historical Depth
13% of total score
DHMZ maintains national archives of hydrological data since its founding in 1947, and river-gauge historical records extend back several decades. The periodic Meteorological and Hydrological Bulletin (published monthly, with archived PDFs back to at least 2022 publicly online, and older issues in institutional repositories) contains some reservoir-level information embedded in context. The EU Water Framework Directive monitoring network, operational in Croatia since 2013–2014 accession, provides another 10+ years of structured records. However, for reservoir-specific storage volumes or fill percentages, the historical online record is very thin: there is no publicly queryable time-series of Mm³ stored in specific reservoirs. Historical data exist within DHMZ and HEP internal systems but accessing them requires formal data requests. Academic papers (e.g., Peruća water quality, 2019–2021 monthly data) demonstrate that multi-year records exist and have been shared for research, but are not routinely published as open datasets.
Update Frequency
10% of total score
For the data that is published — river water levels via DHMZ and Hrvatske vode — updates are genuinely daily, with the XML bulletin (hidro.hr/hidro_bilten.xml) refreshed each day and the Vodostaji app/portal reflecting near-real-time gauge readings. The ENTSO-E 16.1.D dataset, if populated for Croatia, is weekly. However, because actual reservoir storage volumes are not published, this daily/weekly rhythm applies to proxy indicators (gauge heights) rather than to fill percentages. HEP annual reports provide one summary per year. The DHMZ monthly Meteorological and Hydrological Bulletin provides a monthly narrative with some reservoir context. Score reflects that while the underlying sensor network is daily, the storage-specific data gap means the useful public update rate for reservoir fill is effectively annual (HEP annual report) or absent.
Methodological Transparency
8% of total score
Croatia has made meaningful steps toward methodological transparency through EU WFD compliance. A detailed guidelines document for hydromorphological monitoring of heavily modified water bodies has been published by Hrvatske vode. A peer-reviewed assessment of hydromorphological conditions of Croatian reservoirs (MDPI Water, 2023) documents capacity figures, classification criteria, and monitoring methods for major reservoirs. DHMZ data is licensed under the Open Licence of the Republic of Croatia, with attribution requirements stated. However, the capacity figures and stage-volume curves for individual hydropower reservoirs are not published by HEP; the INFOZOK environmental database is internal. The regulatory harmonisation study (ResearchGate, Cosic-Flajsig) documents the gap between EU WFD requirements and actual practice. Measurement methods for karst catchments are discussed in scientific literature but not as standardised public metadata on any national data portal.
Language and Usability
5% of total score
The primary data interfaces are entirely in Croatian. The DHMZ main website has an English landing page (meteo.hr/index_en.php), and the daily hydrological report is accessible in English via a URL parameter (hidro.dhz.hr/hidroweb/skripte/hidrobazahtml.py?funkc=bilten&jezik=en), but the content translates river-stage alerts, not reservoir storage figures. The Vodostaji portal and mobile app are Croatian-only. The voda.hr website has a basic English section ('About us', organisational description) but the Geoportal and data download interfaces are in Croatian. The ENTSO-E transparency platform provides English-language interface and documentation for whatever Croatia reports there. Data.gov.hr has an English CKAN interface for dataset discovery but the uploaded datasets and their metadata are in Croatian. Overall, an English-speaking analyst can navigate to Croatian water data only at a superficial level; extracting actual reservoir time-series requires Croatian proficiency or institutional contact.
Evaluator notes
Croatia presents a paradox of solid monitoring infrastructure combined with very limited public data output for reservoir storage specifically. The country operates a capable network of ~384 automatic hydrological gauge stations (Croatian Waters / Hrvatske vode) plus ~47 DHMZ automatic hydro stations, with data published daily via XML API and a public portal (vodostaji.voda.hr). The DHMZ XML open-data endpoint (hidro.hr/hidro_bilten.xml) is a genuine programmatic access layer licensed under the Republic of Croatia Open Licence. However, all of this infrastructure reports river water levels and flows — it does not publish the reservoir storage volumes or fill percentages that are the core metric for RTI. HEP (Hrvatska elektroprivreda), which controls all 28 major hydropower reservoirs with a combined storage of approximately 1,050 Mm³, does not release operational storage data publicly; its internal INFOZOK system handles environmental compliance data. ENTSO-E 16.1.D (weekly aggregate reservoir filling rate, reported by TSO HOPS) is the one channel where a European-standard reservoir fill metric could appear for Croatia, but coverage and reporting consistency for Croatia on this dataset are weak. A genuine structural constraint is the Dinaric karst hydrology underlying most of Croatia's large reservoirs. Karst aquifer systems create highly heterogeneous catchment boundaries, with significant subsurface exchange between apparent drainage basins — this makes simple volume accounting less reliable and may partly explain why operators have not committed to public fill-percentage reporting. Peruća (566 Mm³, by far the largest reservoir), Lokvarsko, and the Lika-Gacka system all sit in karst terrain where hydrological monitoring requires dense measurement networks and is difficult to reduce to a single 'fill %' figure. This is a technically valid challenge, but comparable karst-dominated countries (Slovenia, Montenegro) still publish at least annual aggregate storage statistics. From a transparency improvement perspective, the path of least resistance would be: (1) HEP publishing weekly aggregate hydro storage figures to ENTSO-E 16.1.D consistently, matching the obligation already in place; (2) DHMZ extending its XML feed to include reservoir-specific water levels with published stage-volume curves so that fill % can be computed by third parties; (3) data.gov.hr hosting a structured reservoir metadata dataset (capacity, operator, basin) as a companion to the existing water district spatial data. The EU Water Framework Directive reporting cycle and the climate-resilience impetus from the DHMZ 2024–2028 strategic plan (KlimProPos/KISS open-data initiative) provide institutional levers for these improvements.
Evaluated by Jaime Delgado · 2026-05-29 · Methodology v1.3.0
Compare with
Other countries with grade D: