New York’s proposed Community and Climate Protection Act has a goal for “the state of New York to reduce greenhouse gas emissions from all anthropogenic sources 100% over 1990 levels by the year 2050, with an incremental target of at least a 50 percent reduction in climate pollution by the year 2030”. In order to reach that ambitious CO2 reduction goal all sources of CO2 emissions have to be reduced. One energy sector with relatively large emissions is residential home heating and the clean energy alternative for home heating is electric heat pumps. In this post I explain why I think that air source heat pump deployment in New York coupled with the simultaneous goal to eliminate greenhouse gas emissions is fatally flawed based on a case study for conversions near Caledonia, NY.
How Stuff Works explains that “heat pumps use a small amount of energy to move heat from one location to another”. Air conditioners cool our homes by removing heat from the air inside and moving outside. An air-source heat pump acts like an air conditioner in the summer and in the winter works in reverse moving heat from the outside air into the home to warm it. Obviously this kind of heat pump eliminates the need to have two separate systems and advocates tout its energy savings too. According to the Department of Energy (DOE):
An air-source heat pump can provide efficient heating and cooling for your home. When properly installed, an air-source heat pump can deliver one-and-a-half to three times more heat energy to a home than the electrical energy it consumes. This is possible because a heat pump moves heat rather than converting it from a fuel like combustion heating systems do.
Air-source heat pumps have been used for many years in nearly all parts of the United States, but until recently they have not been used in areas that experienced extended periods of subfreezing temperatures. However, in recent years, air-source heat pump technology has advanced so that it now offers a legitimate space heating alternative in colder regions.
For example, when entire units are replaced in the Northeast and Mid-Atlantic regions, the Northeast Energy Efficiency Partnerships (NEEP) found that the annual savings when using an air-source heat pump are around 3,000 kWh (or $459) when compared to electric resistance heaters, and 6,200 kWh (or $948) when compared to oil systems. When displacing oil (i.e., the oil system remains, but operates less frequently), the average annual savings are near 3,000 kWh (or about $300).
Reading this statement gives the impression that this technology is a “no regrets” solution for replacing oil heating CO2 emissions because it saves money for home heating. However, there is a critical caveat for New York State. Air-source heat pumps only work when they move heat and when it is really cold (below zero degrees Fahrenheit) there is no heat in the air to move.
The American Council for an Energy-Efficient Economy published a paper that illustrates this issue with air source heat pumps: Field Assessment of Cold Climate Air Source Heat Pumps (ccASHP) (https://aceee.org/files/proceedings/2016/data/papers/1_700.pdf). The report describes a Center for Energy and Environment field study in Minnesota where cold climate air source heat pumps were directly compared to propane and heating oil furnaces. The report notes that “During periods of very cold temperatures when ccASHPs do not have adequate capacity to meet heating load, a furnace or electric resistant heat can be used as backup.” Figure 2 (ASHP Supplemental Energy Use) from that document graphically shows the problem. In this field study homes were instrumented to measure the heat pump and furnace backup usage. Backup furnace usage was relatively low and the heat pump provided most of the heat until about 20 deg. F. For anything lower, heat pump use went down and the furnace backup went up. Below zero the air source heat pumps did not provide heat and furnace backup provided all the heat.
I believe that there are two problems with the plan to deploy air source heat pumps. I suspect but will not try to evaluate that because a fossil fired furnace or electric resistant heat must be used as backup in a typical New York State winter the cost savings from a more efficient heat pump are offset by the need to maintain a second heating system. The other problem is what might happen to peak electrical loads if electric resistant heat is the preferred backup system. The analyses that I have reviewed point out that converting a natural gas system to an electric heat pump system increases operating costs because natural gas is so low. Propane or fuel oil conversions save money so would be the first to convert because of the higher costs of propane and fuel oil. However, I am not sure that homeowners who convert would want to maintain an oil or propane furnace simply because of the storage system requirement. Consequently, I believe radiant electric heat will be the preferred option for air source heat pump conversions. If residential home heating is electrified significantly electric load will increase and I wonder what could happen to load when the efficient heat pump is replaced with radiant electric heat when the temperatures get really cold.
I hypothesize that the combination of widespread air source heat pump deployment and increased reliance on wind and solar renewable energy will create unacceptable reliability issues during peak winter load periods. I evaluated energy usage for one week before and one week after the 2017-2018 peak winter day (January 5, 2018). I had previously analyzed data near Caledonia, NY and will use that for this analysis.
I used two sources of data. Electric load data for New York State are available from the New York State Independent System Operator and meteorological data are available from the NYS Mesonet meteorological system. The NYS mesonet is a network of 126 weather observing sites across New York State. The official website of the Mesonet includes a tab for live data that brings up station information for the 125 operating individual sites that shows that available data include wind direction and speed, temperature at two levels, relative humidity, precipitation, pressure, solar radiation, snow depth, and camera images. I obtained hourly and 5-minute archived meteorological data for two sites near Caledonia, NY where a 180 MW solar farm has been proposed.
The Winter 2017-2018 load peak occurred during an intense cold snap. From December 29 to January 8 the temperature did not get above freezing and there were four days with below freezing temperatures as shown in the table of Daily temperature and load statistics. Note that the highest load did not occur on the coldest day. This was because the coldest day was a Saturday when business loads are lower. Also note that the New Year’s holiday occurred during this period which also reduced the load. The graph of load, temperature and wind speed for winter peak 2017-2018 shows how hourly load varies with temperature over the 15 day peak period.
In order to estimate how much renewable energy would be available during these conditions I converted to solar insolation and wind speed into power generated in MW using example utility-scale facilities. For solar power I used the 180 MW Horseshoe Solar Farm estimated output because it is near the NYS mesonet stations. In my analysis of Solar Issues in Upstate New York using that facility I assumed that 180 MW of power would be generated when the solar insolation equaled 600 watts per square meter and power output the rest of the time would be proportional so observed solar insolation. I believe that is a conservative assumption but would welcome comment.
There aren’t any wind farms nearby. So I estimated power output for a 100 MW wind farm. I found a reference that stated “Wind turbines start operating at wind speeds of 4 to 5 metres per second and reach maximum power output at around 15 metres/second”. I assumed that below 9 mi/hr wind output was zero and that power output was proportional to the wind speed difference between 9 mi/hr and 33 mi/hr consistent with that reference. The NYS mesonet measures wind at 10m and I assumed that the wind farm hub height was 90m. I modified observe wind speed using the wind profile power law with a coefficient of 1/7 to account for the relationship between wind speed and height.
I used Field Assessment of Cold Climate Air Source Heat Pumps Figure 2 (ASHP Supplemental Energy Use) to estimate the amount of power needed when an individual home convert to an air source heat pump and uses radiant electric heat when the heat pump becomes ineffective (assumed to be 15 deg F). I crudely digitized the lines in Figure 2 and calculated the best fit lines for ASHP Consumption and Furnace Backup Consumption. I converted the energy use to electrical energy by converting Btu to watts by dividing the Btu energy use by 3.41. The Energy Use for Residential Home Heating Electrification Table Table illustrates my concern that residential home heating conversion to air source heat pumps has the unintended consequence that when it gets below 15 deg F and consumers really need to heat their homes that the rate of energy use increases over six times per five degree drop in temperature.
The purpose of this analysis is to determine if there are problems if the 100% renewable solar and wind target is coupled with widespread implementation of residential home heating with air source heat pumps. The Housing Units by Space Heating Fuel Table lists the number of occupied housing units for two counties near Caledonia. The Field Assessment of Cold Climate Air Source Heat Pumps report states that liquefied propane (LP gas) and fuel oil or kerosene space heating are the most likely sectors to convert to heat pumps because of fuel cost savings. There are 18,244 housing units that burn those two fuels. I calculated the electricity required for 10%, 15% and 25% conversions for 18,244 housing units.
The figure entitled Residential Home Heating ASHP Conversion and Renewable Power Case Study shows the relationship between home heat electrical load and meteorological conditions affecting renewable wind and solar power. Colder days in Upstate New York often occur on clear, windless nights. When the sun rises the temperature increases quickly. Although cloudless skies maximize solar power the sun is low in the sky and the days are short so the power output is low. Of course the cold weather increases the need for home heating energy.
The Cumulative Renewable Charging and Discharging Margins graph attempts to estimate energy storage requirements. Clearly the only way that solar and wind can be expected to cover winter peak loads is by incorporating energy storage. During this windless case study energy storage needs to discharge to cover the residential home heating power requirement as shown in blue. During the day solar power recharges the energy storage as shown in red. In this case study the maximum storage needed was 372 MW-hr on hour 82. It turns out that renewable excess power charged to the system before this case study was sufficient to cover that requirement.
This case study illustrates my concern that wide-spread implementation of air source heat pumps coupled with increased use of renewables will be difficult. In this analysis the meteorological conditions on New Year’s Eve 2018 show that the proposed Horseshoe solar facility with a nameplate capacity of 180 MW and a wind farm with a nameplate capacity of 100 MW would have been just able to cover the conversion of 2,737 homes to air source heat pumps. However, energy storage capable of at least 372 MW-hr has to be available somewhere. There already are 47,000 homes using electricity and another 15,000 homes that are supposed to be cost-effective candidates for conversion just in two local counties. Most importantly, this is just one component of residential electricity load which is one component of total load.
The Horseshoe Solar Farm – Public Involvement Program claims that the facility will provide enough electricity to meet the average annual consumption of 33,000 or 50,000 households, based on average annual household electric consumption of 10.8 MWh for the U.S. and 7.2 MWh for New York State, respectively. I bet that these household electric consumption averages do not reflect an electrically heated home in cold regions. If I guess that the average consumption for this 15 day period is a decent number for the heating season and assume a 90 day heating season that more than doubles the electric consumption for a New York State household. In other words there is no way Horseshoe Solar Farm is going to provide enough electricity for 50,000 homes using air source heat pumps.
Even though this is a crude “back of the envelope” analysis, the sobering results suggest that the Legislature should do a complete winter peak analysis correctly before codifying reductions that eliminate fossil fueled power plants and require the conversion of residential home heating to meet some arbitrary CO2 reduction goal. According to Patterns and Trends – New York State Energy Profiles: 2002-2016 there are over a million homes currently using fuel oil or kerosene, 500,000 homes using electricity and another 200,000 using propane in New York State.
Based on my analysis I think that even moderate air source heat pump deployment for the residential home heating sector in New York coupled with the simultaneous goal to eliminate greenhouse gas emissions using extensive deployment of wind and solar power is fatally flawed. I cannot imagine how much wind power, solar power and energy storage would have to be deployed to cover the winter peak, much less the winter peak adding significant electrification of residential home heating, for the entire state because those renewable resources are very weak during winter peak load periods. It is incumbent upon the advocates for the Climate and Community Protection Act to determine what renewable resources will be required and how much they will cost before their legislation is considered by the Legislature.