v0.4.1 💽 Persistent memory and safer recovery​

Version v0.4.1 focuses on persistent memory — one of the most important reliability improvements in SILAM Pollen.

SILAM Pollen now saves the last successfully loaded data so sensors and forecasts can recover faster after a Home Assistant restart or a temporary SILAM/network issue.

With this update, you get:

• faster recovery for sensors and forecasts after Home Assistant restarts;
• fewer empty or unavailable states during startup;
• more stable behavior during temporary SILAM or network issues;
• fewer unnecessary requests to the SILAM service;
• more resilient SMART mode when datasets are temporarily incomplete or unavailable;
• clearer diagnostics and better Home Assistant Repairs warnings.

💽 Persistent memory​

SILAM Pollen now stores the last successfully loaded data and can restore it when the saved data still matches the current integration settings.

This helps keep pollen sensors and forecasts available more consistently, especially after a Home Assistant restart or a temporary network/SILAM issue.

🧠 More stable SMART mode and fewer SILAM requests​

SMART dataset selection is now more resilient when SILAM datasets are temporarily incomplete or unavailable.

Together with persistent memory, this also helps reduce unnecessary requests to the SILAM service. When the integration can safely restore or reuse already available data, it avoids downloading the same forecast again unless a real live update is needed.

🗂️ More reliable SILAM service and data checks​

SILAM Pollen now handles SILAM service availability, published datasets, and runs catalog checks in a more centralized and consistent way.

This helps the integration better understand the current SILAM state, while allowing SMART mode to choose data sources more safely and avoid outdated or unavailable datasets.

🩺 Diagnostics, Repairs, and update services​

This release also improves day-to-day maintenance:

• a new SILAM service status diagnostic sensor helps show whether SILAM, the selected dataset, or the data itself is currently available;
• Home Assistant Repairs can warn about unavailable manually selected datasets or deprecated legacy entities;
• manual update services were improved for the Home Assistant UI;
• the Full Refresh service helps check SILAM directly, without using saved data;
• existing automations and scripts remain compatible.

🌐 AI-assisted translation updates​

Translations were updated with the help of AI to cover the new diagnostics, services, Repairs, and setup messages.

If you are a native speaker and notice inaccurate or unnatural wording, corrections are very welcome.

v0.4.0 🧠 Smarter forecasts that adapt automatically​

Version v0.4.0 takes SMART dataset selection to the next level.

SILAM Pollen now monitors which forecast data is actually available and can automatically adapt when the preferred dataset is temporarily incomplete.

This means:

• fewer gaps during SILAM dataset updates,
• less manual dataset management,
• fewer unnecessary downloads,
• more stable and predictable pollen forecasts.

🗂️ Shared runs catalog manager​

SILAM Pollen now uses a shared catalog manager for all configured locations.

Catalog information is cached and reused, reducing duplicate requests and providing a consistent view of the latest available SILAM data.

🧠 Coverage-aware SMART dataset management​

SMART mode now manages datasets continuously while the integration is running.

If the preferred dataset cannot provide the required forecast range, SMART can:

• switch to another suitable dataset,
• keep the forecast updating automatically,
• return to the preferred dataset when new data becomes available.

🏷️ Dataset source tracking​

Forecast data now keeps track of which dataset it came from.

This allows SILAM Pollen to combine data from multiple sources safely while preserving higher-priority forecast values.

🧩 Forecast completion from multiple datasets​

When the preferred dataset does not provide the complete forecast window, SILAM Pollen can fill the missing period using other available datasets.

Only missing forecast data is added, without overwriting values already received from the primary source.

🌍 SILAM Northern Europe (v6.1)​

The regional SILAM Northern Europe dataset has been updated to v6.1 and now includes all improvements introduced with the SILAM pollen v6.1 generation.

🗑️ Legacy datasets retired​

The following legacy datasets are no longer available for SMART or manual selection:

• SILAM Europe (v6.0, legacy)
• SILAM Northern Europe (v5.9.1)

The migration approach proposed in PR #23 was expanded to cover all supported legacy dataset replacements.

🩺 Safer and clearer setup​

When a new integration is added, SILAM Pollen now checks whether the SILAM service and pollen catalogs are available.

The setup interface shows a clear status and provides localized guidance when the service, network connection, or selected location cannot be validated.

🗣️ Expanded language support​

• Added a Polish translation, contributed through issue #24 by @ArekKubacki.
• Updated the Slovak translation through PR #27 by @misa1515.

v0.3.2 🧠 Smarter pollen updates for the new season​

Version v0.3.2 strengthens the SMART update logic, making pollen forecasts more reliable and easier to use as the new season begins.

With SMART dataset selection, SILAM Pollen no longer relies on a fixed configuration.
Instead, it automatically chooses the most suitable pollen dataset for your location, based on actual coverage — helping ensure that the forecast you see is relevant to where you live.

In practice, this means:

• less manual setup and guesswork,
• fewer cases of misleading or irrelevant pollen data,
• smoother behaviour when datasets change between seasons.

Manual dataset selection is still available, but SMART mode is recommended for everyday use.

🌍 SILAM pollen v6.1 datasets​

This release introduces support for the latest SILAM pollen v6.1 datasets, which form the basis for forecasts in the upcoming pollen seasons.

Compared to previous versions, v6.1 datasets provide:

• updated SILAM v6.1 model outputs for the current season,

• a refreshed European baseline (SILAM Europe pollen v6.1),

• a new high-detail regional dataset where available.

• SILAM Finland pollen v6.1 offers the highest spatial detail within its coverage area
  🟦 Blue zone — coverage of SILAM Finland (v6.1) with up to 0.8 km resolution, especially useful for users in:

  • Finland
  • Saint Petersburg and Northwestern Russia
  • Northern & eastern Sweden (including Stockholm)
  • Northeastern Norway
  • Northern Estonia (including Tallinn and Narva)

When available, this dataset is automatically selected by SMART mode, with manual selection still possible in the integration options.

⚠️ Deprecation​

v0.3.1 🧠 Smarter data updates.​

Version v0.3.1 introduces a major internal improvement in how SILAM Pollen detects and updates forecast data.

The integration is now run-aware and checks the SILAM runs catalog before fetching data. This allows it to reliably determine whether a new model run is actually available, instead of blindly re-downloading the same dataset.

If the current run hasn’t changed:

• cached data is safely reused,
• full XML downloads are skipped,
• and only genuinely missing forecast hours are fetched when the forecast window moves forward.

This results in:

• significantly fewer unnecessary network requests,
• faster and more predictable updates,
• reduced load on both Home Assistant and the SILAM data infrastructure.

These changes are especially important ahead of the pollen season, ensuring timely updates while keeping the integration efficient and robust.

v0.3.0 🚀 One glance — the full picture of weather and pollen.​

v0.3.0 delivers a cohesive experience: one card brings together weather and pollen with a familiar look, fast response, and clear presentation. The integration now ships with the dashboard card included: a local JS module is bundled — no CDN or external dependencies (you only need to add it once under “Resources”). The Basic panel preserves the simplicity of the stock Weather card, while the Extended panel unfolds all forecast and allergen layers — tidy, informative, and straight to the point.

• 💠 Experimental Lovelace card (local build)

  • Self-contained JS bundle, no CDN, works offline.
  • Built-in visual editor (LitElement) — configure right in the UI.
  • Live WebSocket updates — no polling delays.
  • Automatic i18n: language, formats, and units from Home Assistant.
  • Native MDI icons and theme support.
  • Compatible with any weather.*, shines with weather.silam_pollen_*_forecast.

• 🖼 Basic panel

  • Mirrors the stock Home Assistant Weather card for maximum visual consistency.
  • For current weather, shows current conditions and one primary attribute of your choice — just like the stock card — and additionally lets you select an unlimited number of draggable attributes (toggle row or column layout).
  • When forecast is enabled for SILAM, the forecast row shows the pollen index icon and state instead of weather condition and temperature; for other weather integrations, behavior is identical to the stock card.
  • Supports a scrollable list of forecast slots if they don’t fit horizontally.
  • Number of slots is configurable in the editor: up to 12 via slider or unlimited via manual input.
  • Selecting attributes in the additional block unlocks wide possibilities to visualize weather data the way you like.

• 📊 Extended forecast panel

  • Modes: Standard, Focus, Minimal.

  • 🌿 Allergen-level forecasts

    • A dedicated forecast line for each selected allergen.
    • Peak concentrations highlighted in 12-hour and daily views.
    • Show details on tap on pollen blocks: opens the sensor card for the tapped allergen; behavior is configurable in the editor.
    • Trend logic + icon — clear up/down indication.

  • ⛅ Full forecast layers

    • Temperature and apparent temperature, precipitation (probability and amount), UV index, wind (speed, gust, direction), cloud coverage, humidity, dew point, pressure.
    • Meteo risk strip (experimental): fog (radiation/advection), dew, frost, icing/freezing rain, heavy rain/snow, heat, wind chill, strong/storm wind, “barometric saw”.
    • Slim strips for humidity and dew point, clear labels, refined typography.
    • Gestures: tap / hold / double tap (navigate, link, call service, perform action).

---

🌿 Updated pollen sensor model​

• 🌸 Pollen Forecast Sensor — now primary and out of beta
• 🗂️ Pollen Index — deprecated
  • Marked as deprecated.
  • Available only via the Legacy toggle (off by default for new installs).
  • Migration preserves compatibility: existing entries get legacy=true.

v0.2.7 🚀 Major update of the "Pollen Forecast BETA" sensor!​

After its beta period, the forecast sensor is heading for a stable release—bringing more accuracy, data and possibilities for your dashboards.

• 🔄 Re-worked algorithms in "Pollen Forecast BETA"

  • The state now reflects the current pollen index from the now block, not the first hourly step.
  • Hourly forecasts aggregate the index and allergen levels by maximum within each three-hour window (was median).
  • Daily and 12-hour values are calculated using an observational percentile (no interpolation):
    ≥ 18 points → 80th percentile · 12–17 points → 70th · < 12 points → maximum.

• 🌸 Forecast sensor is always created
  Sensor weather.silam_pollen_{Zone Name}_forecast is present regardless of options—now data are always available; hourly, twice-daily (12 h) and daily forecasts appear only when the forecast option is enabled.

• 🌅 Daily forecast & allergen peaks
  The sensor now includes a daily forecast (up to five days) based on the observational percentile.
  New attribute allergen_peaks reports peak allergen concentrations for both daily and twice-daily windows—shown when specific allergens are enabled.

• ➕ New attributes for "Pollen Forecast BETA"
  next_condition, pollen_<allergen>, altitude, date, responsible_elevated—extra context for automations and dashboards.

• ⏱️ Diagnostic sensor “Forecast Horizon”
  The new sensor.silam_pollen_{Zone Name}_forecast_horizon shows how many hours the current forecast (state) actually covers and what forecast length (forecast_duration) you asked for.

• 🖼️ Supported in pollenprognos-card v2.4.1+
  The card now renders the daily forecast via weather.get_forecasts.

  Version compatibility
  • pollenprognos-card ≥ v2.4.1 requires silam_pollen ≥ v0.2.7
  • Older card versions (≤ v2.4.0) remain compatible with silam_pollen ≥ v0.2.5.

v0.2.6​

• 🆕 native_name Attribute
  All allergen sensors now include a native_name attribute that reflects the original SILAM API key (e.g. alder_m22, grass_m32, etc.), ensuring better compatibility with external tools and visualizations.

• 🔧 Unified entity_id Naming
  Entity IDs (including defaults and resets) now correspond exactly to the keys in the localization files. For example, sensor.silam_pollen_{Zone Name}_grass remains identical across all languages and for every sensor in the integration.

v0.2.5 🌟​

This is a truly significant update for SILAM Pollen!

• 🎉 Default HACS integration
  Congratulations to us all — the integration is now included in the official HACS repository by default, making installation easier than ever!

• 🖼️ Beautiful dashboards
  Great news for dashboard lovers: @krissen has added SILAM Pollen support to the pollenprognos-card (since v2.3.0) — now current conditions and pollen forecasts are available right on your dashboard!
  Learn more here 👈

• 📈 Long-term statistics
  Pollen sensors and the new diagnostic sensor now collect and display historical data.

v0.2.4​

• 📖 Default README in English
  The README file is now presented in English by default.

v0.2.3​

• 📦 Prepared for HACS publication
  Changes made to host the integration in the default HACS catalog.
• 🌐 Added Slovak translation
  Thanks to @misa1515 for implementing the Slovak localization!
• 🌐 Added Dutch translation
  Thanks to @rubdos for implementing the Dutch localization!