18 May 2019

Case study: Improving climate resilience of a hydroelectric project based on IHA methodology

Hydroelectricity is an important resource to reduce greenhouse gases emissions and mitigate climate change; but hydroelectric projects are also exposed to a wide variety of climate change impacts. To ensure safety, economic viability and minimum environmental impact on the long-term, hydroelectric infrastructure must be able to operate under current and future climate conditions.

In this case study we used the recently published International Hydropower Association’s Climate resilience guide on an hypothetical dam project in order to demonstrate how Callendar can help project owners, financial institutions, developers and governments as they try to assess and reduce climate risks.

International hydropower association's guideline fo climate adapation

IHA’s process overview

Phase 1: Project risk screening

Goal and methodology

This study begins during the feasibility study as projects components are already defined but remain amendable. The goal of this firs phase is to provide a qualitative evaluation of climate risks based on the project characteristics and context, already available information and stakeholder consultation.

It includes the definition of performance indicators for the project. For each indicator a performance goal is set and, when relevant, a tolerable loss margin (ie: the range below the goal that can be accepted by stakeholder). The impact of climate change on each indicator is evaluated based on a bibliographic study and on publicly available data.

Ultimately this phase will determine whether or not climate change could make performance goals unachievable or endanger the project’s safety, economic viability or service continuity. In that case, a formal climate risk assessment is necessary.

Callendar’s added value

Even though this first phase can be performed in-house by many companies, we have access to a large array of past and future climate data and the expertise and know-how to make the screening faster and more cost-effective. Our approach is based on our on-the-shelf climate data visualization tools and allows a quick, qualitative evaluation of climate impacts and uncertainties for the project.

Phase 2: Initial analysis

Goal and methodology

The main goal of this phase is to establish the reference climate for the project. Usually, this baseline should represent current climate conditions.

This raise two issues:

  1. The availability and reliability of past weather observations over a period long enough to establish climate pattern (at least 30 years),
  2. The evolution of climate during the observation period.

As observed weather records are usually scarce and very unevenly distributed, reference climate will be based on reanalysed weather data. These gridded datasets are interpolated by weather models taking into account on past weather observations ; they are available globally with medium to high spatial resolution and subdaily timestep which make them convenient for such study. Observed records, obtained from local weather stations and/or global database (e. g.: Global Historical Climatology Network), will be used to check the quality of reanalysed data and, if necessary, to biais-correct them.

This dataset will be used to search for trends over the last decades. If a trend can be identified, it will be compared with the anomaly computed from regional climate projections, i.e.: the difference between near-future forecasted and past backcasted with the same model.

If the trend derived from past weather data and the anomaly projected by climate models are congruent, it is reasonable to use as a reference a 30-years climate trace which is identical to the reanalysed, biais-corrected dataset adjusted to eliminate the anomaly.

Callendar’s added value

Our experience with reanalysed and observed weather data as well as downscaled climate projections allow us to go beyond data science: we will help you make sense of the results and transform raw data into operational insights.

Phase 3: Stress test

Goal and methodology

This phase is arguably the core of the study. Its goal is to assess quantitatively how the project will performed under a wide range of possible future climates.

According to IHA’s methodology, the stress test is be based on the hydrological model of the project coupled with an technical and economic models. This system is run with sequence of weather reflecting possible future climate as input. The output of each run is then compared with performance goals defined during the first phase. The share of successful runs give an indication of the project’s climate resilience.

In addition to project modelling, the accuracy of this methodology relies on:

  1. Thorough exploration of climate projections to make sure that variations critical for the project is not ignored. This means in particular considering multiple emissions scenarios and for each scenario multiple climate models, including potential outliers. For each couple scenario, model (or ensemble of models), a statistical analysis must be performed to assess the frequency and magnitude of phenomenons relevant for the project such as precipitations, drought, extreme events, etc.
  2. Selection of 20 to 30 climate scenarios: based on the previous analysis a range of climate scenarios will be chosen to reflect the space of possibilities, each scenarios is composed of a set of variables (such as temperature, precipitation, wind, etc.) and their probability density function. This choice is project-specific: for small and/or limited lifespan projects, some scenario can be discarded if they appear unlikely while for critical projects, very implausible scenarios combining the most unfavourable characteristics of various projections can be worth considering.
  3. Generation of sequences of weather for each climate scenario, for example using a stochastic weather generator. These sequence must be adapted to the variables, time scale, spatial resolution and electronic format required by the hydrological model.

Callendar’s added value

As a provider of climate data for industrial decisions, Callendar is particularly well-qualified to support you during this phase.

We can also offer complementary methodologies, such as our Climate Lab: recognizing that all risks can not be captured by models (for example, an apparently benign flood may render roads critical for the hydropower project unavailable and become a major threat), we developed this workshop to simulate the operation of an installation under climate stress and find out potential vulnerability.

Phase 4: Climate risk management

Goal and methodology

The objective of this phase is to reduce the vulnerabilities identified at the previous step while remaining cost-effective. This can be done by adapting the project design or by making sure that these adaptations are possible in the future should the need arise.

Based on the insights gained from the stress test, various combinations of adaptation measures will be proposed. These measures can be selected through expert judgment to address the project vulnerabilities prioritizing risk to the infrastructure safety ; they should allow future adaptation.

The project modified with each possible combination will be subjected to new stress tests using same the models, methodology and weather scenarios as in phase 3. The result will be a matrix representing the gain or loss for each combination of adaptation measures and each weather scenario compared to the unamended project under baseline climate.

The best combination will be selected using three criteria:

  1. Absence of safety risk: combination that fail to eliminate critical safety risk will be eliminated
  2. Tolerable loss: if a combination result in losses outside the acceptable range agreed with stakeholders during phase 1 for one or more climate scenario, it has to be eliminated
  3. Minimum maximum loss: remaining combinations ranked by the maximum loss among all climate scenarios

The combination with the smallest maximum loss is the more resilient while ensuring that the project do not bear safety risk and remains within the range of uncertainty agreeable by stakeholders. If all combination fail this test, the project need to be further adjusted or abandoned.

Case example

In our case, we chose 4 combinations of adaptation measure and tested them for financial results and safety in 4 climate scenarios: baseline, RCP4.5, RCP6.0 and RCP8.5. The result matrix is showed below:

This table can be read as follow:

  • first row, first line: the original project under baseline climate serves as a reference, as a result financial loss are 0
  • second row, first line: under a weather sequence based on IPCC’s RCP4.5 scenario, the original design suffer a financial loss of approximately 2 millions euros per year but its safety is not compromised
  • second row, second line: if the project is modified using the combination #1, it performs better under the same weather sequence with a financial loss reduced to 0.4 millions euros per year

The original design and option #4 have to be discarded because they bear safety risks for at least one climate scenario. Assuming that all other options are within acceptable loss range, the more resilient combination is #2 with a maximum loss of 4M€ per year compared to the unamended project under baseline climate.

Phase 5: Monitoring, evaluation and reporting

Goal and methodology

The adaptation of the project does not end with this study: resilience will be monitored and improved during its whole lifespan and new adaptation measures may become necessary as climate change unfold. This last phase will make sure that:

  • Data, hypothesis and models used during the study are properly documented and the rationale of every choice is explained for later review,
  • A frame is defined to track project’s resilience and performance reviews, coordinated with O&M plan, are scheduled
  • A methodology is set to monitor the evolution of climate in the project’s area
  • Events that should trigger an update of the climate risk assessment are listed, these events may be for example: the detection of a climate trend in observed weather data, new, diverging climate projections, regulatory change, etc.