Latest January Sensor Readings with Temperature Changes

Write a query to optimize a slow-performing database join on a table containing millions of sensor telemetry logs.

Use the provided PostgreSQL tables and restrict analysis to January 2026. Return only active sensors that have telemetry during that period.

Output

1. One row per qualifying sensor with sensor_id, site_code, sensor_name, latest_recorded_at, latest_temperature_c, previous_temperature_c, and temperature_delta_c
2. Use the most recent reading, breaking timestamp ties with the greatest log_id
3. Order by sensor_id ascending

Output details

• Treat January 2026 as recorded_at >= TIMESTAMP '2026-01-01 00:00:00' and recorded_at < TIMESTAMP '2026-02-01 00:00:00'.
• Define previous_temperature_c as the preceding reading for that sensor within January, ordered by recorded_at ascending and log_id ascending.
• Return previous_temperature_c as NULL when no preceding January reading exists.
• Return temperature_delta_c as latest_temperature_c - previous_temperature_c, yielding NULL whenever either value is NULL.