How Clean Would Site-C Hydro Power Be?

As most citizens of British Columbia realize by now, the Site-C hydro power project along the Peace River has been approved by both the Federal and the Provincial Government. After reading the aforementioned approvals I noticed the project was always referred to as being the “Clean Energy Project”. This notion of clean they refer to is commonly understood as being due to the operation phase of hydro power not requiring any solid fuel substrate to convert into electricity and therefore does not emit undesired air emissions during energy conversion such as particulate matter, nitrogen oxides, sulphur dioxide, dioxins and heavy metals. The Site-C project’s level of ‘cleanness’ can also be referred to in terms of life cycle greenhouse gas emissions which is an important metric when comparing several power production alternatives. I became very curious to understand, how ‘GHG-clean’ would the Site-C project likely be.

GHG Intensity of Site-C Hydro Power as Calculated

Well, it turns out Site-C hydro power would likely be quite clean. Upon reviewing the ‘Report of the Joint Review Panel Site C Clean Energy Project‘ information was limited on how BC Hydro undertook the life cycle greenhouse gas (GHG) analysis of the project. After a little more digging I found the original technical report for which the life cycle GHG intensity of 10.5 grams CO2eq. per kWh (gCO2eq./kWh) is based on. At face value British Columbians should  feel fairly good about this low GHG intensity factor considering natural gas and coal power have GHG intensity factors that typically fall in the range of 350-550 and 900-1200 gCO2eq./kWh, respectively. The IPCC Special Report on Renewable Energy provides the following concise figure explaining life cycle GHG intensity of different power sources as depicted below:


Figure 1: Life Cycle GHG Intensity Ranges for Several Power sources (units: gCO2eq./kWh) (source)


Despite the Site-C hydro power system likely having a low GHG intensity it is a worthwhile exercise to gain at least a simplified understanding of how this 10.5 gCO2eq./kWh factor was derived. The GHG technical report split GHG project emissions into two main categories: hydro power reservoir construction and operation. With a series of sensitivity analyses included, the likely annual average emissions from these two activities were estimated to be 124,600 tonne (t) CO2eq. per year due during the construction phase (for 8 years) and 43,400 tCO2eq. (net) per year during the operation phase (for 100 years). Since the capacity of Site-C is designed to be 1100 Mega-Watts (MW) and the average annual energy generation is expected to be about 5100 Giga-Watt-hrs (GWh) we can calculate the report’s estimated likely GHG intensity factor of 10.5 gCO2eq./kWh:  


10.5 gCO2eq./kWh = (43,400 tCO2eq./yr + 124,600 tCO2eq./yr*8yr/100yr)/(5100 GWh/yr*106kWh/GWh)*106g/tonne


The GHG emissions during the 8 year construction period were due to material requirements, fuel combustion and electricity demand and these emissions contributed 1.95 gCO2eq./kWh to this 10.5 value. The remainder and majority of the GHG intensity (8.5 gCO2eq./kWh) was due to the net GHG emissions that stemmed from direct land use change (emissions that occur during the operation phase). Let’s explore this part of the GHG intensity factor next. 

Net GHG emissions from Land Use Change

The Site-C reservoir requires a significant area to be permanently inundated. Stretching 83 km along the Peace River the surface of the reservoir is expected to span an area of 9,330 hectares (ha) and this would require the flooding of 6,345 ha of land. The figure below shows the land-type breakdown of the land area that would be slated for inundation. 


Figure 2: Breakdown of the land-types to be flooded as a result of Site-C Hydro Project (source)


As shown above the majority of land inundated would be sparse trees while about 570 ha of land to be flooded is considered as cultivated agricultural land. The idea of net GHG emissions attributed to the Site-C project due to inundation was calculated by taking the difference in GHG emissions/sources between the inundation area as it is today and the inundation area upon project completion where operational GHG emissions are estimated across a 100 year operational time horizon. In terms of current land productivity the technical report assumed that apart from the cultivated land there were also livestock/grazing activities considered where a current estimate for beef (1,000 head) and dairy cattle (300 head), swine (300 head), horses (150 head) and bison (300 head) was made. GHG emissions from current activities (i.e. farming) upon areas planned for flooding were then estimated and subtracted from the operational GHG emissions expected from the reservoir over the 100 year life time. The figure below illustrates the possible GHG emission sources from a typical hydro reservoir.


Figure 3: A schematic of the various GHG sinks and sources in a hydro reservoir system (source)


As illustrated in the figure above there are a series of complex mechanisms that need to be considered when estimating GHG emissions and sources. A recent study by Professor Edgar Hertwich attempts to integrate these emissions into a life cycle assessment perspective. Here Prof. Hertwich reviewed and analyzed the generation of greenhouse gas emissions from reservoirs for the purpose of technology assessment, relating established emission measurements to hydro power generation. What he found was the global average emissions from reservoir hydro power are estimated to be around 85 gCO2/kWh and 3 gCH4/kWh. Despite this relatively high average in net GHG emissions (equates to 148 gCO2eq./kWh) the study’s recommended numbers for boreal reservoirs (0.97 kg CO2/m2/yr and 40 g CH4/m2/yr) leads to a GHG intensity factor of 21 gCO2eq./kWh.

These factors suggest that the likely GHG intensity assumed for the Site-C hydro power system might be underestimated though the level of uncertainty in calculating net GHG emissions from a reservoir makes it difficult to say who is more correct. At first glance, I think the Site-C GHG technical report undertook a thorough GIS based site-specific analysis utilizing as much localized data as possible and their likely GHG intensity factor of 10.5 gCO2eq./kWh is probably more accurate because of the details that went into this technical report. All in all, Prof. Hertwich highlighted the importance of this emission source, the major limitations of trying to measure these emissions and the high level of uncertainty involved in these calculations. He also found that the power density (plant capacity divided by reservoir area) of the hydro power plant has a strong correlation with reservoir GHG intensity. He cautioned that any hydro reservoir plants with a power density of around 4 W/m2 or less should be avoided. It turns out that the Site-C hydro power plant would have a power density of 20 W/m2 which is well above this CDM project eligibility boundary that Prof. Hertwich generally agreed with.

Followup Questions

Although I do think the Site-C hydro power system would be relatively clean in terms of GHG intensity I do think it is only fair to explore the alternatives and to better understand their associated GHG intensities. For instance, the Canadian Geothermal Energy Association proposed a more cost effective geothermal power alternative and Clean Energy BC proposed a more affordable small-to-medium scale renewables alternative as well. While these projects suggest less cost to BC tax payers, could these projects also equate to less GHG intensity? Another question on my mind is what about the 570 ha of cultivated land and the additional land required for grazing and livestock activities currently taking place on the to be inundated Site-C project land? In a typical LCA study involving presently productive land use change the GHG emissions due to indirect land use change should also be considered in the GHG intensity calculations. In subsequent blog posts I will explore these followup questions. Please stay tuned.