Air Source Heat Pumps In New York

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.

Background

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.

Procedure

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.

Case Study

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.

Conclusion

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.

 

Five Reasons Why the Catastrophic Anthropogenic Global Warming Story is Wrong

Scott Adams (of Dilbert comic fame) recently did a video about climate persuasion titled “Scott Adams solves the climate debate and saves the world (really)”, available here on periscope or here on twitter that has sparked folks to distill their arguments for and against climate change. After hearing about this and faced with the constant barrage of media stories about the latest inevitable climate doom story I thought it would be appropriate for me summarize five reasons why I think that the political agenda to transform the energy system of the world is not supported well enough by sufficient scientific evidence to proceed.

The majority of the claims that anthropogenic carbon dioxide emissions are the cause of the observed warming and that we only have a short time to do something or else are based on projections from global climate models (GCM). As noted below I do have relevant experience, education and background to inform my opinion. However, I believe that anyone who does research based on their personal experience, background and education can reach their own informed opinion of these claims if they actively try to get both sides of the story.

First a bit of background of these models. If you want details, Dr. Judith Curry did a detailed overview that includes a summary description. For my purposes all you need to know is that these models are a variation upon the meteorological models that provide predictions that everyone uses when you make decisions based on weather forecasts. There are differences but they use the same physical relationships such as momentum, conservation of heat and conservation of mass.

I have M. S. and a B. S. degrees in meteorology and in my fourth semester of Weather Analysis & Forecasting the laboratory assignment was to break off into teams and write a simple weather forecast model. I have been inside this kind of model but most readers also have some mathematical background that is relevant. In particular you may remember in algebra that if you have three equations you can solve for three unknowns but if you only have two equations you cannot solve them. The problem in meteorological models is that you have more unknowns than you have equations so model developers have to improvise. In particular, instead of using direct relationships for every single factor that affects weather or climate forecasts, meteorologists use parameters to simulate the effects of some atmospheric processes.

The first reason that I am skeptical of any GCM results is the use of parameters which can be thought of as “fudge factors’.   Model developers necessarily have to account for some things that cannot be modeled directly. John von Neumann allegedly summed up the problem stating that “With four parameters I can fit an elephant, and with five I can make him wiggle his trunk”[1]. In other words, he could develop a mathematical model that described an elephant simply by fudging the parameters. Everyone who makes a decision based on a weather forecast has learned that you can trust a forecast for tomorrow better than one several days away. In the 43 years since I graduated the forecasts have become more reliable for dates further in the future because weather forecasters have constant feedback and have been able to adjust the parameters in the meteorological models to improve forecasts based on observations. There is only one global climate system and forecasts made today for 100 years away cannot be checked until 100 years have passed. One insurmountable problem is that the parameters and their use in GCMs cannot be verified as correct in our lifetimes.

Another issue with the parameters is that the focus is on just one of these parameters. Richard Lindzen commented on this:

“Here is the currently popular narrative concerning this system. The climate, a complex multifactor system, can be summarized in just one variable, the globally averaged temperature change, and is primarily controlled by the 1-2% perturbation in the energy budget due to a single variable – carbon dioxide – among many variables of comparable importance.  This is an extraordinary pair of claims based on reasoning that borders on magical thinking.”

My final difficulty with parameters in GCMs is that they are used to model clouds. Dr. Curry explains that in order to solve the physical equations in a global climate the world has to be divided up into a three-dimensional grid. The equations are calculated for each grid cell and repeated to generate a forecast. My particular problem is that the grid cell size needed in order to do these calculations are on the order of 100 km horizontally, the vertical height is often 1 km and they do the calculations every 30 minutes or so. As a result, the models cannot simulate clouds. That single parametrization is a big enough driver of climate that this model component alone could dominate the GCM projections. This uncertainty is well understood in climate science by those who have worked with these models. However, the problems with parameterization is not well understood and its ramifications on the policy decisions is poorly understood by most of those who advocate eliminating fossil fuel use.

My second reason for not trusting these models is related to my experience in air pollution meteorology and work I did as a consultant to EPA evaluating the performance of air quality dispersion models. Complexity-wise those models are orders of magnitude simpler than climate models and they are simple enough to be directly verified. All air quality dispersion models are based on results from field studies that injected tracers into stack effluents, measuring the downwind concentrations in a test array under a wide array of meteorological conditions and then developing coefficients for pollution dispersion. I worked on a project where we compared the known emissions and observed ambient concentrations to model projections at power plants. The results showed that the models in use at the time adequately predicted maximum concentrations so EPA was comfortable that they were working correctly. The frightening result in my mind is that it was not uncommon for the model to predict a maximum concentration close to the maximum observed value but the meteorological conditions for the predicted maximum would be different than the meteorological conditions for the observed maximum.

Consider the differences between the GCMs and air quality models. Air quality models use parameters that are based directly on observations, incorporate known emissions, and have been extensively verified in field studies but those models could get the right answers for the wrong reasons. GCMs use parameters that are based on model developer opinions, have to estimate future emissions, and cannot be tested in the climate system. We are supposed to believe the models are getting the right answers for right reasons. I don’t think that is a reasonable assumption for modeling that is the basis for converting the entire economy away from fossil fuels.

My third reason for not accepting the commonly held belief that catastrophe is inevitable is that the projections that make that claim are only one answer in a wide range of potential outcomes. Consider for example that the Intergovernmental Panel on Climate Change (IPCC) does not give a single value for the sensitivity of atmospheric temperature to carbon dioxide. Instead they give a range of potential warming for a doubling of atmospheric concentrations of carbon dioxide of between 1.5 and 4.5 degrees C. Nic Lewis does a nice job discussion climate sensitivity here.  The GCM projections cover a wide range of potential outcomes from benign to catastrophic. Without going too deep I want to point out that the damage claims for increased carbon dioxide depend on the shape of the distribution of this sensitivity. Damage affects society when costs are greater and those estimates are strongly affected by the probability of extreme damages. The first problematic aspect of this issue is that the projections for high impacts rely on a relatively high probability of extreme impacts. Although recent research has shown that the likelihood of extreme outcomes is lower than previously thought, those results have not been incorporated into the damage estimates. If they were considered, then the costs currently claimed would be reduced.

More problematic to me than the range of possible model outcomes is the use of the worst-case representative concentration pathway as the business as usual scenario. The IPCC developed a set of four pathways to concentrations representing the range of radiative forcing (i.e. greenhouse effect) in the literature for 2100. The range of possible future atmospheric forcing levels runs from relatively low levels of carbon dioxide to the highest forcing they thought possible. The problem is that these concentrations had to be related back to emission scenarios and the worst case representative concentration pathway with a forcing of 8.5 watts per meter squared is so high that the emission scenario necessary to get that level is not credible. Many of the really scary projections that get the headlines and dominate the narrative why we need to reduce carbon dioxide emissions use RCP 8.5 as business as usual. We already know that the likelihood of that future emission scenario is extremely unlikely. When coupled with models that give a range of outcomes from benign to problematic, I can only conclude that while it is not impossible that there could a catastrophic impact the probability is so low that it should not drive policy decisions.

My fourth reason for not trusting the models that claim carbon dioxide is the primary driver of the recently observed warming is that even for the limited results we have from the models that can be compared to climatic system they don’t do well. Dr. Curry has explained that inconsistency well:

Between 1910 and 1940, the planet warmed during a climatic episode that resembles our own, down to the degree. The warming can’t be blamed on industry, she argues, because back then, most of the carbon-dioxide emissions from burning fossil fuels were small. In fact, Curry says, “almost half of the warming observed in the twentieth century came about in the first half of the century, before carbon-dioxide emissions became large.” Natural factors thus had to be the cause. None of the climate models used by scientists now working for the United Nations can explain this older trend. Nor can these models explain why the climate suddenly cooled between 1950 and 1970, giving rise to widespread warnings about the onset of a new ice age.

The final reason that I believe that the political agenda to transform the energy system of the world is not supported by sufficient evidence to be credible is the suggestion that climate change is easy to solve because renewables are a viable solution. As convinced as I am that the climate science does not support this agenda I believe the suggestion that wind and solar can solve our energy problems is even more of an exaggeration. In fact that issue is the primary driver why I blog and have written so much about New York’s climate change plans. However, don’t listen to me, listen to Bill Gates who states “The idea that we have the current tools and it’s just because these utility people are evil people and if we could just beat on them and put (solar panels) on our rooftop—that is more of a block than climate denial,” Gates said. “The ‘climate is easy to solve’ group is our biggest problem.” Another problem is that the renewable “solution” very likely has very significant environmental impacts that are generally ignored. Michael Shellenberger had a Ted talk “Why renewables can’t save the planet” that addresses this issue.

To sum up. The rationale used to justify the need to convert the energy system of the world is that carbon dioxide will cause inevitable catastrophe and we can be saved if only we implement renewable wind and solar which will be easy to do. I don’t believe the science supports inevitable catastrophe because those projections are based on global climate models. Those models use too many fudge factors that can give too many results that can never be tested, much simpler models that are based entirely on observations can give the right answer for the wrong reason and the model results to date do not adequately predict the one climate experiment we can test. Most of the catastrophic outcomes that dominate the political and media narrative depend on an emissions scenario that is not credible. I do not believe that diffuse and intermittent solar and wind can be used to replace reliable and affordable electric power much less generate enough energy to convert transportation, heating, and industrial use of fossil fuel to electricity.

[1] Attributed to von Neumann by Enrico Fermi, as quoted by Freeman Dyson in “A meeting with Enrico Fermi” in Nature 427 (22 January 2004) p. 297

 

Solar Energy Issues in Upstate New York

Led by Governor Andrew Cuomo, New York’s solar ambitions are a key component in his agenda for to ensure “vital progress on the climate” is continued. This is a post on one aspect of the NY-Sun program. I am a retired meteorologist who worked in the electric generation sector for over 35 years and I know that New York is not a particularly sunny place in the winter so I wanted to check out potential issues with solar variability during the peak summer and winter loads. The opinions expressed in this post reflect my personal opinion.

Introduction

NY-Sun is supposed to make solar affordable for all New Yorkers. According to the NY Sun section in the NYS website Leading on Climate Change and Protecting our Environment:

  • NY-Sun is developing a sustainable, self-sufficient solar industry in the State by incentivizing New Yorkers, businesses, and communities to invest in solar energy.
  • The Governor’s $1 billion NY-Sun program has grown solar power in New York State by nearly 800% since 2011, and has reduced greenhouse gas emissions by nearly 25%.
  • The program aims to add more than 3 gigawatts of installed solar capacity in the State by 2023, enough solar energy to power 400,000 homes.

(Proof reading this before publication I was struck by these claims so I posted on them at my companion site.)

In order to evaluate the effect of solar variability on the transmission grid I needed an example facility. New York State’s permitting process for power plants of 25 megawatts or higher has extensive requirements for public involvement. Invenergy Solar Project Development LLC has started the permitting process for construction of Horseshoe Solar Farm which I will use as my example. According to the Horseshoe Solar Farm – Public Involvemen Program it will be a 180 MW Solar Electric Generating Facility Located in the Town of Caledonia, Livingston County, New York. Eventually I will address this particular project in detail later but this post only looks at potential performance during peak periods.

It is interesting to note that the New York Independent System Operator had this to say about solar photo voltaic (PV) facilities in their 2018 Load & Capacity Data report:

The actual impact of solar PV varies considerably by hour of day. The hour of the actual NYCA peak varies yearly. The forecast of solar PV-related reductions in summer peak reported in Table I-9 assumes that the NYCA peak occurs from 4 p.m. to 5 p.m. EDT in late July. The forecast of solar PV-related reductions in winter peak is zero because the sun sets before the assumed peak hour of 6 p.m. EST.

Because reliability planning necessarily focuses on peak periods I decided to look at the loads on the 2017 peak summer day (July 19, 2017) and the 2017-2018 peak winter day (January 5, 2018). The load data are available from the NYISO on hourly or 5-minute intervals. I decided to estimate the capacity from this facility during these peak periods (one week before and one week after the peak day) using the hourly data. I also wondered about the short term variations so I used the five minute data for the 72 hours around each peak day.

Analysis

In order to estimate the solar generation output from this facility I used solar radiation data from two nearby NYS Mesonet meteorological systems (Rush and York). 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. You can change the station by clicking on any dot on the state map. Data available include wind direction and speed, temperature at two levels, relative humidity, precipitation, pressure, solar radiation, snow depth, and camera images. I got archived solar insolation data on an hourly and 5-minute interval for two nearby sites that I used to estimate solar generation.

I calculated the electrical generation output from the 180 MW Horseshoe Solar Farm based on my internet research. I think it is a pretty good estimate but if someone reading this could confirm that or tell me what I am doing wrong I would appreciate it. The mesonet stations provide solar insolation measured in watts per meter squared. I assumed that the 180 MW of solar cells produced 180 MW when the solar insolation equals 800 watts per square meter (the PVUSA test condition) and I did not account for any other factors such as the cell temperature or any losses. So my naïve formula was simply the observed input solar insolation times 180 divided by 800.

The first question to address is how much power can we expect from a 180 MW facility in upstate New York during the peak periods? Frankly, I would expect this location to be pretty good relative to other central to western New York sites. Locations closer to Syracuse would be worse and locations due east of Lake Erie or Lake Ontario would be much worse because of lake-effect clouds. The Estimated Solar Generation from Horseshoe Solar Farm table lists the results. In the summer the results are pretty good. According to my methodology peak output could be over 200 MW when insolation is highest and the capacity factor over the entire 15-day peak period is over 23%. However, in the winter the solar generation output would be abysmal. The peak generation hour was only 103 MW and the capacity factor of the 15-day peak period would be no more than 7%.

The second question is what about the shorter term variability in solar generation output. The Genesee Load and Horseshoe Solar figure lists 5-minute load (MW) data for the Genesee Control Area Total Load (pink circles), estimated Horseshoe solar potential output using the Rush NYS Mesonet station solar radiation data (blue box) and the estimated Horseshoe solar potential output using the York NYS Mesonet station solar radiation data (red cross).

There are several issues. Note that the scales are different so for starters this solar farm makes little difference to the energy needs of the Genesee NYISO control area load.  As noted by NYISO the solar potential energy diurnal cycle does not match the load peak well because actual load peaks after solar generation peaks. The estimates of solar generation from both sites shows quite a bit of variation during the day. For example, the peak insolation at 12:55 EDT is 1,046 watts per square meter but ten minutes later it is down to 491. I estimate that will translate into a swing of over 100 MW that some facility somewhere has to cover. Also note that there are numerous periods when the estimated solar generation using Rush data differ from estimates using York data. This suggests that it was a partly cloudy day with significant variations in solar insolation. Because these are 5-minute averages the instantaneous variations are likely larger.

Ramifications

As I suspected these data show that the proposed NY-Sun solar buildout in Upstate New York will not end well. With respect to the abysmal capacity factor of the winter peak keep in mind that Cuomo’s Reforming the Energy Vision includes plans to convert residential home heating to electrical heating and there are activists that not only want to do that on an accelerated schedule but also are insisting that all the electrical power be produced by wind and solar energy only. Theoretically, battery storage could provide all the renewable power. However, only the energy innumerate or energy naïve could possibly think that solar energy coupled with battery storage could provide enough energy to heat all upstate residences during winter peak periods without massive overbuilding of both solar farms and battery storage simply because solar potential in the winter is so poor.

The summer short-term period data illustrate the fallacy that solar is cost-equivalent to coal or gas or whatever the claim is today. Fossil-fired facilities provide near constant energy but these data show that solar has huge variations. When considered alone, if this facility or any other solar facility gets built someone else on the grid has to provide support so that power sent to the grid is near constant. Battery energy storage can provide that service and frankly given the degree of intermittency on a day like the peak summer day would be the only solution that might work. However, any battery solution at least doubles the cost of solar. Not surprisingly it is even worse. If you dedicate a battery array to providing smooth power you cannot use the battery array for storage. The State really needs to explain how they propose to incorporate 3 gigawatts of solar.

On the other hand, if battery storage is a requirement for a solar facility these impacts are addressed. That way their wildly fluctuating output does not impact the grid and the units can be dispatched to match the observed load. Of course if that is required the price of solar at least doubles so it is unlikely that prudent and politically inconvenient approach will be adopted.

NYS Energy Profile Patterns and Trends

On March 1, 2019 the New York State Energy Research & Development Authority (NYSERDA) Energy Analysis program published Patterns and Trends – New York State Energy Profiles: 2002-2016 which they described as a “comprehensive storehouse of energy statistics and data on energy consumption, supply sources, and price and expenditure information for New York State.”   I agree and strongly recommend that anyone who has any interest in New York State energy download the document and check it out.

For numbers geeks like me one of the features that I really love is the fact that the tables are linked to spreadsheets. For example, Table 3-1b: NYS Primary Consumption of Energy by Sector in the report only lists data from 2002 to 2016 but when you click on the table and download the spreadsheet data back to 1980 are available. Moreover, the data are in a spreadsheet so you can process it as you wish.

It is very disappointing that there was a 26 month lag between the end of the report period and the publication of the data. I consider this one of the best products of NYSERDA but apparently under the rule of the Cuomo Administration it is not a priority. Before the Cuomo Administration this report came out 13 months after the end of the reporting period. The 1997-2011 report was 18 months later and the last three reports were dated in October so were 22 months late.

I will be using these data in future posts but I cannot help but show how useful it is in one example. National Grid’s “Northeast 80×50 Pathway” is a blueprint for the region to reduce its greenhouse gas emissions 80 percent below 1990 levels by 2050 (“80×50”). The first of its kind in the Northeast, the Pathway spans seven states (New York and the six New England states), and addresses the three main sectors of emissions: transportation, electricity, and heat. If you search on “National Grid Northeast 80 by 50 Pathway” you will get a list of fawning articles talking about how great this is but I am a customer worried about home heating and they are talking about a plan for that.

The politically correct approach for reducing emissions from home heating is to convert to heat pumps. According to the pathway:

Heat pumps are very different from standard electric resistance heaters. Compared to traditional “baseboard” technologies, heat pumps achieve a 50-80% reduction in electricity use by moving heat rather than creating it. They use conventional refrigeration technology to absorb heat from one source (air, ground, or water), transfer it to another source, and raise it or lower it to a temperature suitable for space heating (or cooling) and hot water. Heat pumps still face major adoption challenges. In particular, ground-source heat pumps need to achieve cost declines to become more accessible to customers, and air-source heat pumps need to be paired with proper building insulation.

The question is just how much energy are we talking about. The Patterns and Trends document includes a table of NYS Net Residential Consumption of Energy by Fuel Type that shows just ambitious changes to home heating will be. Total residential consumption of energy typically totals 800 TBtu or 800 trillion British thermal units. Thomas Fuller writing about the challenges of meeting the Green New Deal notes that a BTU is a unit of energy–strictly speaking the energy required to raise the temperature of a pint of water from 39 to 40 degrees Fahrenheit. About the amount of energy released from a burning wooden match. There are around 115,000 BTU per gallon of gasoline and one TBtu is equivalent to 39,000 cars driving round trip between New York and Los Angeles if the cars get 25 miles per gallon.

I graphed the data to see the NYS Residential Energy Consumption trends. Natural gas use has increased over time, coal has disappeared and petroleum products have decreased.  Today Natural gas provides 58% of the home heating energy, petroleum products 16%, and wood around 2%. Electricity, solar and geothermal provide 24%. National Grid’s pathway suggests that petroleum product heating should be converted to electricity which would mean that 128 TBtu of energy needs to be replaced. While that seems plenty ambitious to me there are those that think that we need to convert all home heating to electricity, solar and geothermal. I think that the disruption and expense of complete conversion far out-weighs any benefits but that will be the subject of another post.

NY Green New Deal – NYS 2010 Climate Action Plan

This is one of a series of posts on Governor Andrew M. Cuomo’s New York State Green New Deal. The announcement noted that it will create the State’s first statutory Climate Action Council, comprised of the heads of relevant state agencies and other workforce, environmental justice, and clean energy experts to develop a plan to make New York carbon neutral.  Not mentioned was the fact that there was a previous Climate Action Council that was not created by statute. This post will highlight the draft plan produced by the first Climate Action Council in late 2010.

According to the New York State Department of Environmental Conservation (DEC):

Executive Order No. 24 set a goal to reduce greenhouse gas (GHG) emissions in New York State by 80 percent below the levels emitted in 1990 by the year 2050. The Executive Order also created the New York State Climate Action Council (CAC) with a directive to prepare a climate action plan. The climate action plan would assess how all economic sectors can reduce greenhouse gas emissions and adapt to climate change. The Plan would also identify the extent to which such actions support New York’s goals for a clean-energy economy.

On November 9, 2010, the CAC released an Interim Report that had been prepared with assistance from the New York State Energy Research and Development Authority (NYSERDA), the Department of Environmental Conservation (DEC), and other CAC member-agency staff, the Center for Climate Strategies (CCS) and other stakeholders. This Interim Report is presented by sections and chapters at the DEC website.

First Climate Action Council Plan

For my purposes, Chapter 4: Envisioning a Low-Carbon Future is of most interest. This effort is based in large part upon a Brookhaven National Lab white paper entitled Envisioning a Low-Carbon Clean Energy Economy in New York. The ultimate question is whether the earlier New York State 80 by 50 goal is feasible not only based on cost but on technical considerations. I had originally intended to dissect this vision of the future to address those points but I think the following Important Note to Readers from the white paper speaks to my concerns. I have highlighted the critical point.

Important note to readers:

This is the first complete draft of a paper designed to inform the NYS Climate Action Council’s work to develop a State Climate Action Plan.

The Council’s mandate is uncommonly broad in scope. It has a planning horizon far longer than what most planners address. It entails large uncertainties. No clear precedent for an enterprise of this scope exists.

Consequently, this draft paper is necessarily provisional. As the planning process proceeds, the paper will be revised, and it will steadily gain in value as fresh insights are acquired and the knowledge base it draws from expands.

One feature of this paper is a description of three scenarios that illustrate different versions of a low-carbon 2050 future for the state. It’s important that readers understand that these scenarios are offered for illustrative purposes only. In no sense do they constitute the elements of a plan, and indeed even a casual review of them reveals that there is no way in which they could be fashioned into a plan. Rather, they’re intended to facilitate and provoke thinking about the future.

We hope other parties will generate their own 80×50 scenarios and share them. The ability to imagine a sustainable future, model it rigorously, and explore it is as vital to achieving that future as the clean-energy technologies, best management practices, and behavioral changes that must be developed, advanced, and adopted.

Conclusion

The Brookhaven White Paper developed three future scenarios. One scenario expanded on existing programs to make the most obvious emission reductions. Although it assumed “significant changes to current practices, this scenario falls far short of achieving 80 percent emissions reduction by 2050.” The second scenario assumed electrification of the entire light-duty vehicle fleet to hydrogen fuel produced with nuclear or other low-carbon electricity, elimination of fossil fuel combustion in the residential, commercial, and industrial sectors and significant use of locally-sourced biofuels for trucks and aircraft but was only able to make a 79% reduction. In order to get to an 80% reduction the final scenario assumes 95% of all vehicle miles are all-electric miles, eliminates fossil fuel combustion in the residential/commercial/industrial sector with “part of the resultant increase in electricity demand met through local, point-of-use solar and much of the remainder with low-carbon generation and the wide-spread use of carbon-capture and sequestration”.

It does not take much effort to come to the same conclusion as Brookhaven that there is no way that these scenarios could be fashioned into a plan. Ultimately, the question is whether there is any possible plan to meet the ambitious goals of New York’s Green New Deal.

RGGI in the Weeds

Tom Shepstone at Natural Gas Now has graciously re-posted several of my posts including this post Regional Greenhouse Gas Initiative on the Fast Track to Nowhere based on this post. Unfortunately, the arcane world of pollution control programs is difficult to understand without a lot of background and my posts presume more than a little background. As a result there are some things that need to be clarified with respect to Tom’s conclusions from my post.

Tom made the following four conclusions. My indented and italicized comments follow.

First a bit of background. The Regional Greenhouse Gas Initiative (RGGI) is a cap and trade program. In order to understand the point I was trying to make you need to understand the fundamentals of cap and trade. What you need to know about this pollution control approach is that there are two components: the cap and tradable allowances for the pollutant covered. The cap sets a limit on the total regional emissions that must be met over a trading season. The cap is set at a level such that the pollutant of interest will be reduced to levels that are supposed to improve air quality to the appropriate standard. Setting the cap level correctly is critically important: too high and the environmental objectives won’t get met and too low and the market mechanism won’t work.

 There is a wrinkle for RGGI. Instead of a traditional cap and trade program it is a cap and auction program. Normally allowances are allocated to the affected sources based on some past historical performance metric. In RGGI allowances are sold off in quarterly auctions. The affected sources universally consider this a tax inasmuch as they have to pay for the allowances they need to operate. Seriously, no one is claiming that RGGI is going to have any impact on global warming but proponents can claim that they use the auction proceeds to fund all sorts of feel-good initiatives that in some cases actually do reduce CO2 emissions. I described my thoughts whether RGGI was a success here, here, here and here.

Natural gas has reduced emissions faster than anyone thought possible, making it necessary to actually increase emission allowances in 2017 for the obvious purpose of giving renewables at least a chance to catch up.

When the RGGI program was being implemented the forecasts of future generation and emissions assumed much higher gas prices which resulted in high coal unit usage and high CO2 emissions. As a result the cap was set high but the natural gas revolution made those estimates inappropriate. As a result when RGGI started the auction price of allowances was so low that proponents of the program were not getting as much money as they wanted.

There is a scheduled program review component in RGGI and during that process the existing caps were lowered significantly and future reductions were incorporated that are more ambitious than I believe is warranted. The RGGI states and environmental organizations believe that RGGI was the reason for most of the reductions and argued that because reductions had been so significant to date that lower caps were appropriate. However, they missed the point that the reductions were mostly due to reduced operations in the RGGI states and fuel switching from coal and residual oil to natural gas. RGGI had very little to do with it.

Renewables are not catching up much, if at all, because investments in them are dependent on Federal subsidies and, therefore, their potential is limited.

RGGI auction proceeds are supposed to be used to reduce emissions or provide ratepayer relief. The fact of the matter is that the record of RGGI investments actually reducing emissions is poor.   As Tom notes the potential is limited for further reductions based on RGGI’s own data.

Because of these facts, the opportunities to achieve meaningful reductions in greenhouse gases by rewarding investment in “compliance” entities are dissipating like a sunset and faster than the wind dies down in severe cold.

I did not adequately describe the terms “compliance entities” and “non-compliance” entities. Compliance entities are those fossil-fired generating units that have the compliance obligation to surrender a RGGI allowance for every ton of CO2 emitted. Non-compliance entities are those organizations that have purchased RGGI allowances as investments.

This means the Regional Greenhouse Gas Initiative is headed nowhere in terms of the strategy the public has been sold by the politicians; it is approaching a situation (if not already there) where it will have to reward investment in non-compliance entities such as natural gas fired power plants or fine these entities, which will then pass the costs onto consumers who will never know what hit them.

I agree that RGGI is headed nowhere but the problem is different than Tom described. The problem is that the non-compliance entities (think Morgan Stanley and other investment companies) now hold the majority of the RGGI allowances. The RGGI states reduced future allowances allocated to the auctions and their cap presumes that further reductions are possible when the fact is that most of the fuel switching has already occurred. As a result, there are not enough allowances for compliance entities to purchase at upcoming auctions in order to operate. Therefore they will have to go to the non-compliance entities and purchase their allowances if they want to run. This shortage will increase the price and the non-compliance entities will profit. However, the public will not get any benefit from the increased price of the non-compliance entity allowance sales because they only get benefits from auction proceeds.  In other words, the non compliance entities have already purchased the allowances so the higher price of the allowances due to profiteering will simply be passed on to consumers. Worse if the compliance entities are not able to get the allowances they need to run their only compliance option is to not run which could lead to reliability issues.

My final point is that this is uncharted territory for RGGI. No one knows how the market will react or what the prices on the market relative to auctions will be when this allowance shortage hits. Consumers in the RGGI states will be the guinea pigs for this experiment.

RGGI Emission and Allowance 2018 Status

This is a post on the status of emissions and allowances in the Regional Greenhouse Gas Initiative (RGGI). It is another in a series of posts on RGGI that discusses how RGGI has fared so far and what might happen in the future. The fact is that RGGI is edging towards uncharted territory where affected sources that have to comply with the regulations are going to have to get the allowances they need for compliance from investors.

I have been involved in the RGGI program process since its inception. Before retirement from a non-regulated generating company, I was actively analyzing air quality regulations that could affect company operations and was responsible for the emissions data used for compliance. After years dealing with RGGI I worry that whether due to boredom or frustration, that there is very little dissent to the program. It may be because, contrary to EPA and State agency rulemakings, RGGI does not respond to critical comments and rebut concerns raised by stakeholders. After years of making comments that disappear into a void, industry does not seem to think there is value to making comments. The opinions expressed in this post do not reflect the position of any of my previous employers or any other company I have been associated with, these comments are mine alone.

Emissions

RGGI Annual CO2 Emissions lists the total CO2 emissions from the states currently in RGGI. After pretty consistent reductions over time last year there was a six million ton increase from 2017. Stay tuned to another post that looks into the emissions data in more detail. However, note that I have previously posted on the reductions to date which suggest that further reductions will be much more difficult than in the past.

Allowance Status

RGGI is a cap and auction program. Allowances are sold in quarterly auctions to anyone and the proceeds are supposed to be invested in programs that reduce CO2 or ratepayer impacts. RGGI states have modified the original auction allocations to reduce the number or bank of the allowances that have been purchased but not used yet because the original estimate did not account for the possibility that natural gas would supplant coal to the extent observed. Up to this point the affected sources or compliance entities have been able to purchase the allowances needed to cover emissions from auctions. This is going to change in the next couple of years.

RGGI relies on Potomac Economics to provide technical analyses. Frankly, my impression is that the purpose of those reports is to obfuscate and confuse rather than clearly show the status of the program. One of my biggest frustrations is that there is no summary status report and I have inconsistencies in my summary estimates. I have made a couple of guesses at the status of the number of allowances that are held for compliance purposes after the 2018 RGGI emissions are fully surrendered.

According to the Potomac Economics Secondary Monitoring Report for the 3rd quarter of 2018, there were 155 million allowances in circulation and 72 million were held for “compliance purposes”. According to the Potomac Economics Market Monitor Report for the fourth quarter of 2018 entities purchasing allowances for “compliance purposes” bought 77% of the 13,360,649 allowances sold. In RGGI Allowance Allocation Status End of 2018 I added the fourth quarter allowances to get 168 million allowances in circulation and 82 million allowances held for “compliance purposes”. RGGI-wide CO2 emissions were 72 million tons in 2018. Ultimately only 10 million allowances will be in the “compliance purpose” bank. In all of 2018 54 million allowances were auctioned and barring a major reduction in emissions the 10 million ton “compliance purpose” bank will be gone next year.

The Potomac numbers do not include other transfers to the allowance banks. I tried to calculate the allowance bank based on the RGGI allowance distribution reports. (Note that RGGI compliance periods are three years long.) In RGGI Compliance Period Allowance Allocations and Compliance Period Emissions I list the allowances in circulation at the end of 2017 (total allowances released less total emissions). Adding the allowances added this year gives a bank of 140.7 million. I used Potomac Economic’s estimate that 50% of allowances were for compliance purposes to get 70.4 million allowances. In 2018 CO2 emissions were 72.3 million tons so according to this approach compliance entities are already in debt to the non-compliance entities.

Conclusion

I cannot emphasize enough that RGGI is headed towards a situation where the affected sources will have to go to the non-compliance entities to get enough allowances to cover their emissions. If you recall the proceeds that RGGI receives from the auctions are supposed to be used to reduce emissions and provide ratepayer relief. Ideally, the added costs of this carbon tax are offset by those investments. Now, however, the investors will be able to charge whatever they want for the allowances and their profit will be covered by increased costs to the consumer. (In the interests of full disclosure I bought 11,000 allowances in 2018 and will profit from this situation.) In my opinion, affected sources should buy allowances as needed and never run without enough to cover current emissions.

 

If an affected source does not have enough allowances on hand to cover their current emissions they are faced with two issues. When a source bids into the market they prefer to know the price of allowances so they can price their bid appropriately but if they don’t have them they don’t know the cost. Worse would be the case where a facility assumes that they can get allowances but eventually find out none are available at any cost. In that case then they would be out of compliance and would face significant fines. The worst case scenario is that a facility does not have allowances in hand, cannot purchase what is needed and then declines to bid. While unlikely, that could lead to reliability issues because you cannot force an owner/operator to run knowing they are out of compliance without a whole lot of histrionics.

Finally, note that RGGI has closely guarded the ownership of allowances. The market monitoring reports name who has bid but does not list who owns what. Instead they list ownership by three categories:

  • Compliance-oriented entities are compliance entities that appear to acquire and hold allowances primarily to satisfy their compliance obligations.
  • Investors with Compliance Obligations are firms that have compliance obligations but which hold a number of allowances that exceeds their estimated compliance obligations by a margin suggesting they also buy for re-sale or some other investment purpose. These firms often transfer significant quantities of allowances to unaffiliated firms.
  • Investors without Compliance Obligations are firms without any compliance obligations.

In my opinion those categories are pretty broad. In a transparent program there would be examples of which company is in which category but we are left in a position where we have to hope they got the definitions right. Finally, note that investors without compliance obligations could also include those who want to hold allowances to prevent emissions. If that is the case for a significant fraction of investors, then the market is in trouble.