Development The planning analysis for development of a new hydropower project must take into account the effects of climate change on the available water resource. Peter Rae talks to Hydro Review about challenges, risks and how power markets might be affected.
Peter Rae is principal of Peter Rae Hydro Consulting Ltd. in Canada. He has more than 38 years of experience with planning, design and development of hydropower and water resources projects. Rae has worked on all phases of project development, from conceptual and feasibility studies to development activities and construction management. In this interview, Rae provides his opinions about how climate change will influence the planning, development and operations of water resources projects, with a particular focus on hydropower.
Q: How exactly can climate change affect power generating facilities?
Rae: Climate change is a risk that will affect energy generation systems worldwide, although the nature and quantum of the change may be open to some interpretation depending on the technology used.
For an electricity generating facility, any change to the climate will affect operating yield and costs. In the case of thermal plants, the change affects the cooling cycles, whereas solar and wind facilities are affected by cloud cover and wind patterns.
Hydropower is directly affected by climate change in the fact that it alters the amount of water available for generation and potentially the seasonal distribution of runoff.
In all cases, there will be impacts on the cost of energy production and plant operations. Firm energy yield will be affected in all cases. Climate change will also affect design risks for new hydro plants in terms of changes that will result in greater uncertainty concerning flooding, which will in turn affect project costs.
Q: Can you go into greater detail on how hydropower specifically can be affected?
Rae: Hydrological analysis for hydro projects normally assumes homogeneous and stationary data records. The assumption is that the overall climate of the watershed is unchanging and that a sufficiently long hydrologic record can display the inter-annual and seasonal variability of the net water yield at a particular site. The use of a long record has been assumed to provide a reliable measure of the bounds of energy production, as well as its long-term mean.
Feasibility of the development of a new hydropower project typically is predicated on statistical analysis of the hydrological record, or more rarely using rainfall-runoff models with a stationary and homogenous input for the meteorological database. The hydropower industry has developed an array of more or less standardized methods to test and quantify the hydrological database.
In a stationary system, generation planning can be performed by assuming that in the future the system will have similar characteristics to those represented by the historical record. This assumption also applies to other power stations that might be considered in the generation expansion for the power system being studied. For example, the heat rate for thermal power stations would often be taken as constant with time, albeit with minor adjustments for changes in mechanical efficiency or reliability.
But planning in a universe affected by a changing climate must assume a non-homogeneous power system and non-stationary conditions. Watershed yield may be affected by significant changes in the seasonal distribution of precipitation and the effects of changes to snow pack storage. Even if the same volume of runoff is available at some sites, the loss of snow pack could reduce the potential energy yield. The result can be a long-term change in the character of the energy yield and the associated firm energy available from the system.
Precipitation itself will likely be affected by long-term trends that will change the gross water input to the watershed. Other important effects for hydro plants will be changes in evapotranspiration and evaporation water losses.
The new normal is moving toward a situation where project and power system evaluation need to consider the non-homogeneity of the hydrologic system and the change in key elements of the climate system. The effect for individual power stations is a change in net energy yield and its seasonal distribution with time. Note that the changes will occur over a period of decades but the hydrologic system may reach certain tipping points where its character shifts dramatically during a short period.
Q: I imagine that with these considerations, analytical approaches for hydropower development studies need to change as well. Can you elaborate on this?
Rae: Analytical methods for assessment of projects in a world with climate change should more properly support the evaluation of potential future revenues. The climate change trend is assessed using global circulation models (GCMs), which simulate global changes over a long period. GCMs have been developed and are used by a number of organizations internationally that are engaged in long-term climate research. These models do not provide sufficient detail for individual project evaluation, but the scenario outputs can be used to predict changes in parameters that are of use for hydrological analysis. One concern is that different models provide different results, although the consensus results are similar in terms of the overall trends. A suite of models may be required to assess the range of possible future scenarios, especially if the project is in a region where the models do not provide a consistent result. The application of the GCMs will likely remain within the realm of climate science rather than being adopted as project-scale tools.
At the project level, we are in a phase where conventional analytical methods and tools must be adapted for the non-stationary environment. As an alternative to conventional statistical modeling, a deterministic modeling approach can be used to assess the effects of climate change at a project level. A basin-scale model can be developed to represent the rainfall-runoff process for a recent period, which can be taken as being reasonably stationary. The model should be calibrated and then verified for a separate period of the record. Future scenarios can then be developed from the model by introducing trended parameters for some of the key inputs, such as precipitation and temperature.
In temperate climates, the seasonal accumulation and melting of snow pack is an important part of the total project storage cycle. Simulation of this process is essential to estimate firm energy. Climate change scenarios that involve even a small increase in basin temperatures will affect the snowpack, with potentially significant effects on the firm yield of a project. Deterministic modeling of snow accumulation and melt is required.
Key parameters can be defined by sampling from a selection of the available GCMs. It is important to recognize that any scenario taken from the model is only one possible future and the actual outcome may differ because of the inherent uncertainty in the prediction of climate change.
This downscaling process from the GCM to the project watershed requires more effort than a conventional analysis. Prediction of realistic future scenarios for energy yield will require careful assessment of the ranges of potential hydrological responses to the changes in precipitation, evaporation and other meteorological inputs.
The outcome of the hydro-meteorological analysis will be a time varying series that may be non-homogeneous due to the effects of changes in such elements as snow pack and evaporation. Use of this product for assessment of a project must then consider the hydrological variability. Consideration should be given to simulation of a family of future hydrologic scenarios that can be analyzed for energy yield in the same manner as would be applied for stochastic hydrological methods.
Several scenarios for input meteorological parameters can be devised with a trend component. These can then be simulated to provide a family of output scenarios that combine inter-annual variability with the longer-term climate change trend.
Analysis of this trended family of scenarios will then provide a variety of inputs to the simulation of reservoir and power station operation.
Q: How can the uncertainty surrounding climate change affect potential investors in hydro project development?
Rae: The investor wants to ensure the hydro project can yield a rate of return that confirms the investment is superior to alternatives. The rate of return will depend on the capital cost of the project, financing terms, tax conditions and future revenue.
With climate change and an associated increase in runoff with time, the investor can see an increase in the rate of return. However, whether this increase can be monetized depends on whether the power market will recognize the value. If all projects within a region are subject to the same effects, then the relative ranking of one project vis-à-vis other similar projects will not change. The challenge is to identify whether there are changes to other elements of the power demand or supply mix that will increase the potential revenue from a project. Note that a project might actually obtain higher revenue even if the gross energy yield is reduced, if the value of the available yield is increased because of the timeliness of the generation.
The effect of the long-term increase in energy yield is muted by the discounting of the future revenues. The normal approach for economic and financial analysis of projects relies on discounted cash flows where events far in the future have progressively less effect on the rate of return. In a changing climate world, many projects will still be found viable even though a long-term trend may reduce their yield. The near term revenues may still be sufficient to justify investment.
A key consideration for the investor is whether the energy value in the market remains constant, regardless of whether energy yield increases or decreases. It is a fact that the market value of the resource will change, and the investor will want to find some methods to predict the characteristics of the future power market so that the revenue can be assessed.
It is possible that the uncertainty about the effect of climate change on power market pricing may be greater than the uncertainty in the evolution of the energy yield. In fact, if climate change is likely to reduce generation from existing sources because of the effects on reduced rainfall, evaporation losses, thermal power efficiency (due to higher temperatures) or reduced snow pack, then it is possible that energy prices will increase due to the effects of shortage, especially if the capabilities of the plants available in the power system are not well-suited to the evolving climate because of a lack of storage and loss of firm energy.
The project investor holds the primary revenue risk due to their requirement to generate a long-term rate of return. The risk of climate change is mitigated to some degree by the discounted value of future dividends, but this does not alter the fact that climate change can significantly affect the returns. Prediction of the effects of changes in yield is required as part of a competent and comprehensive investment risk analysis.
Risk sharing is required to ensure that investors are available to expand power systems. In this regard, the principle of allocating risks to the party best able to manage the risk should apply after a project is completed. The individual project investor has limited opportunities to mitigate their risks in the event of a declining energy yield. The opportunity to affect the layout or cost of the project occurred during the development period, when the optimization of the plant was undertaken.
The long-term role of the plant in the system depends on the characteristics of other projects available for dispatch. The investor is interested in maximizing revenue by dispatch of power and energy to the market, which ideally would value both of these elements of the plant.
Q: How would commercial lenders for hydropower project be affected by climate change?
Rae: Organizations that lend money for energy development have an important voice because of the fact that project financing typically involves limited-recourse funding schemes. In theory, the lenders share some project risks by securing the loans against the future revenues of the project. In practice, lenders are diligent in ensuring a project can generate sufficient revenue so that debt service can be covered with a secure margin.
Firm energy is often used for analysis, and the interests of the lenders are very much limited to the initial years immediately after the commercial operations date, when the debt service coverage ratio will be at the lowest levels and the project has not accumulated reserves from energy yields exceeding the firm yield. The effect of tariff inflation, project reserves, and debt retirement normally acts to increase this ratio from the minimum level in the first year or two after commercial operations.
Given the short time scale of their risk exposure (typically 10 to 15 years), commercial lenders are not in a position to share much of the climate change risks and generally do not have a strong incentive to value climate change as a significant risk. The amount of climate change within this period is small relative to the normal hydrological variability.
The situation is quite different for the multilateral development banks (MLBs) that have a long-term interest in the economic development of the host country. MLBs usually seek to ensure that the energy infrastructure developed will have a viable long-term role in the power system. The composition of the overall power system and the characteristics of individual plants are of interest in this analysis. This view of the societal value of the project emphasizes the need to consider the effects of climate change. The challenge is to ensure that the project will be a valuable component of the power system even if climate change results in a different energy yield and power market.
Q: What about companies that insure power projects? Does climate change affect them significantly?
Rae: Insurers to power projects should have an interest in the effects of climate change due to coverage for physical damages, which could result from flooding or other weather conditions during operation. The challenge is to value the insurable risks in a climate change scenario. The approach is dealt with through an underwriting system where policies are renewed at regular intervals through the life of the project. The insurers do not, therefore, have a long-term stationary interest in the project but are able to adapt their coverage and pricing to the changing circumstances.
Q: It seems that effects would also be felt by the entity that regulates the power system. Can you discuss that?
Rae: In some jurisdictions, a regulator controls access to the generation system and functioning of the market. The role of the regulator is to make decisions about the present and future power system needs. The regulator can control the generation mix by selecting expansion options based on their capability to supply the anticipated demands.
With respect to hydropower, the effects of climate change on the power system are mitigated to some extent by the normal tendency to add generation to accommodate growth or for fleet replacement. These investment decisions will have the advantage of knowledge of the systems at the time of commitment and the capability of the existing generation plants in the system. The expected outcome will, therefore, be that the new plant will adapt the power system in light of the changes in climate evident in the future, with the result that financial decisions with a relatively short time horizon are still valid.
Hydropower is also a flexible resource that offers a variety of products of value. While energy may be diminished in some climate change scenarios, these other secondary benefits will remain, although investors are somewhat at risk if the benefits are not monetized in the market. Greater thought will be required about market mechanisms that mitigate investor risk in an era of climate change.
The regulator has a key role in managing the risks of climate change by ensuring that the system develops in a manner that mitigates the detrimental effects. Most of the climate change risk is best managed at the aggregated level of the power system. The system is able to balance future additions to complement existing resources to adapt to the changing nature of the climate.
Q: You mentioned market mechanisms to mitigate risk surrounding climate change. What needs to happen here?
Rae: In many jurisdictions, a power market acts as the primary control on generation expansion and the revenue achieved from projects. In these cases, investors often have the option of some combination of fixed power sales contracts and sales to a spot market. Other market designs provide for all generators to bid energy and capacity, with all participants receiving a market clearing price, which would be the highest cost of energy dispatched to serve the demand.
With climate change, power markets are likely to be a good method to allocate the risk for project investors. The market will adjust the value of energy to account for scarcity or surplus, which will provide signals to potential investors for new plant additions. The difficulty is the time scale required for development and the need for advance projects of markets to justify investment.
The theoretical structure and operation of markets will enable the adaptation of the power system required for climate change.
Water resources projects are long-term investments that will have an ultimate time scale affected by climate change. Planning must consider the long-term role of projects given the potential environmental and social impacts.
Markets tend to encourage a short-term focus for investment because of the incentives to focus projects on the predictable energy revenues available. A longer-term financial assessment may be more important in a changing climate because of the need to adapt the power system to the future reality of the climate rather than to the smaller increments likely to occur.