Transforming Climate Ambition into Audit Impact 

Article 2: 

Why are Supreme Audit Institutions making increasing use of Earth Observation and maps?

Auditors typically make extensive use of previous government reports in the area in planning their audits: this is a key part of the audit evidence base.

But if an audit office relies entirely on government self-reporting, it risks producing a report that replicates the established auditee narrative rather than testing it. The UNDP Good Practice Guide on Climate Change Performance and Compliance Audits (published tomorrow) shares good practice in this field and identifies a series of audit offices that have intelligently moved beyond reliance on self-reporting.

A key development is the use of (often free) satellite imagery and open-source geographic information systems (GIS) to allow auditors to verify physical realities directly. This represents a shift from documentary audit evidence to independently verifiable physical evidence. Further use cases are emerging rapidly.

Case Study 1: Verifying Flood Risks and Air Quality in the Czech Republic

A practical demonstration of this approach comes from the Supreme Audit Office (SAO) of the Czech Republic, which has integrated geographic and satellite data across multiple environmental audits. In Audit 19/04 on state support for flood control measures, the SAO went beyond checking if funding guidelines were followed. The audit team extracted active floodplain zones and hazard maps from the regional flood information system and cross-checked against the national real estate registry. By layering these datasets, they independently identified hundreds of buildings constructed inside active flood zones between 2015 and 2018.

In an audit on air quality (Audit 23/07), the SAO evaluated whether public funds spent on dust-reduction projects were effective. The audit team bypassed government reporting by analysing remote sensing data and aerial imagery from Google Earth, national surveying archives, and the European Union’s Copernicus Programme.

The imagery showed that €4.06 million in public funds had been spent ineffectively – leading eventually to a criminal  investigation and a subsequent conviction for subsidy fraud handled by the European Public Prosecutor’s Office (EPPO).

Case Study 2: Measuring Emission Zones in Poland

SAI Poland faced a similar challenge when evaluating whether urban transport strategies were reducing individual car usage and related emissions in major cities. To audit the newly established Low Emission Zone (LEZ) in Warsaw (an area covering 36.7 square kilometres) the audit team utilized open-source software equipped with a deep-learning plugin. They used this to analyse ultra-high-resolution aerial maps across three specific points in time to count the physical volume of traffic.

While the numbers indicated a downward trend, the Polish auditors maintained professional scepticism. They noted that a final conclusion requires normalizing these counts against external variables, such as the specific day of the week the aerial photographs were taken, broader traffic studies, and the regional increase in remote work.

Technology helps – but it does not remove the need for professional expertise and judgement.

Beyond Europe

I have focused here on two European SAIs, on the basis of presentations they made to EUROSAI WGEA in Madrid. This supplements the examples I gave in the UNDP Guide, which came from other parts of the world – impressive applications of Earth Observation from the supreme audit institutions of India, Sri Lanka, Tanzania, Colombia and the Maldives.

Sri Lanka and India used satellite images to assess action on coastal protection – such as the condition of coastal mangroves. Tanzania and Colombia looked at the protection of forest areas. The SAI of the Maldives identified an island that had been washed away because of inappropriate conservation measures. I am looking forward to hearing of more use cases in Bangkok over the next few days.

I discussed the use of Earth Observation in public financial management in an article for the ODI earlier this year – I remain impressed by the leading role of SAIs in applying this technology.

Strategic Implications for SAIs

Changing technology is changing expectations for SAIs. There is an understandable tendency to impose technological change on a top-down basis. Empowering innovation at the level of audit teams can be a productive means of leveraging the skills of staff and of identifying the applications of technology that can be productively deployed.

Using these tools requires a shift in how an audit office manages its resources, but the case studies illustrate that this is achievable. Three specific implications:

  Data sufficiency: Free satellite data (Landsat, Sentinel) is often sufficient for audit-level verification without proprietary systems.

  Automation: Analytical scaling requires basic geospatial automation to prevent evidence bottlenecks.

  Skills retention: Earth observation capabilities require sustained institutional investment, not one-off training.

Conclusion

When an SAI uses independent data to build its findings, it creates strong, more compelling evidence. The audit stops being a debate about political intentions and becomes a statement of physical fact. Evidence of this kind also provides more interesting findings – and this is an underappreciated strength of good audits. Reports are more often remembered for their findings than their recommendations.

Mapping physical outcomes is a huge step forward. Putting this information into a convincing context requires considering the economic incentives embedded in government climate programmes, and the interplay of climate change with public finance.

I will look at SAI work in this area in the next article in this series.