AI Solar Output In The Baltimore Metropolitan Area

We are going to look at ground-mounted solar fields that were retrieved using Artificial Intelligence. Artificial intelligence is defined by accenture as a constellation of many different technologies working together to enable machines to sense, comprehend, act, and learn with human-like levels of intelligence. A ground-mounted solar panel system is solar panels that are mounted on the ground.

The questions we are trying to answer are:

• When and Where did the solar fields appear?
• How is solar energy changing in the Baltimore Metropolitan Area? What was the land cover before the solar hubs were installed?
• What explains the apparition of solar ground-mounted solar fields? What rules and regulations are already in place for solar energy?
• Where were solar fields implanted compared to what is deemed suitable for solar panels? How many solar fields have appeared in areas deemed suitable in Baltimore County?
• How does the solar density map compare with diversity, population density, median, and average income?

• Accenture: https://www.accenture.com/us-en
• Chesapeake Conservancy AI Ground-mounted Solar Output Data: https://cicgis.org/arcgis/rest/services/Chesapeake/SolarAIPoints_BaltimoreMetro/MapServer. https://cicgis.org/arcgis/rest/services/Chesapeake/SolarAIPolygon_BaltimoreMetro/MapServer.
• The Chesapeake Conservancy, Optimal Solar Siting for Maryland: https://www.chesapeakeconservancy.org/conservation-innovation-center/.
• Baltimore Metropolitan Area County and City Boundaries: https://cicgis.org/arcgis/rest/services/Chesapeake/BaltimoreMetropolitanArea/MapServer
• Chesapeake Conservancy Land Use in 2013/2014: https://cicgis.org/arcgis/rest/services/Land_Use_1m_CBP_2013/LU_1m_Mosaic_CBP_2013/ImageServer
• Esri's Wayback Imagery (2014-2021): https://services.arcgisonline.com/ArcGIS/rest/services/World_Imagery/MapServer
• Chesapeake Conservancy Land Use in 2013/2014: https://cicgis.org/arcgis/rest/services/NPS_Land_Cover/ChesapeakeBayWatershedLandCover/ImageServer
• Maryland Legislation: https://energy.maryland.gov/Pages/Info/renewable/solar.aspx.
• Chesapeake Conservancy Baltimore County Suitability Analysis: https://cicgis.org/arcgis/rest/services/_Final_Solar_Webmap_MIL1/MapServer
• Esri's Living Atlas Diversity Index: https://demographics5.arcgis.com/arcgis/rest/services/USA_Demographics_and_Boundaries_2020/MapServer
• Chesapeake Conservancy AI Ground-mounted Solar Density: https://services.arcgis.com/dqnSpuoMiocEiFNg/arcgis/rest/services/SolarAIPoints_BaltimoreMetro___copy_SolarAIPoints_BaltimoreMetro_Density/FeatureServer
• Esri's Living Atlas USA Census Populated Places: https://services.arcgis.com/P3ePLMYs2RVChkJx/arcgis/rest/services/USA_Census_Populated_Places/FeatureServer
• Esri's Living Atlas 2020 USA Average Household Income: https://demographics5.arcgis.com/arcgis/rest/services/USA_Demographics_and_Boundaries_2020/MapServer
• Esri's Living Atlas 2020 USA Median Household Income: https://demographics5.arcgis.com/arcgis/rest/services/USA_Demographics_and_Boundaries_2020/MapServer
• United States Department of Agriculture USDA:

https://www.nass.usda.gov/Statistics_by_State/Maryland/index.php.

• THE LEGEND FOR EVERY MAP IS LOCATED ON THE LOWER LEFT OF THE MAP.

The area defined by the Baltimore Metropolitan area highlighted in the map is Anne Arundel County, Baltimore City, Baltimore County, Carroll County, Harford County, Howard County and, Queen Anne's County. There are 36 ground-mounted solar fields in the area of interest. The distribution is as follows: 6 fields in Howard County, 4 fields in Anne Arundel County, 7 fields in Queen Anne's County, 6 fields in Carroll County, 7 fields in Baltimore County and, 6 in Harford County.

Map of Ground-mounted AI Solar in the Baltimore Metro Area

Looking at the dashboard and comparing the prior land covers, it is noted that most solar fields appeared in either cropland or turf. According to a report published by the Chesapeake Conservancy, Optimal Solar Siting for Maryland, "without siting guidelines and incentives, the majority of future land used for solar energy development is likely to come from agriculture". The report further states that ground-mounted solar competes with desirable land uses for food production and environmental services. Conversion of prime farmland for solar energy development should be avoided because it removes the best land needed for food production. However, solar energy development is an opportunity to put degraded or contaminated lands and underutilized industrial sites to productive use.

Based on the table and the graph (illustration), most ground-mounted solar fields appeared between 2016 and 2018.

This map provides the land use back in 2013 of the different areas before most solar fields appeared. One can observe a clear pattern of solar fields being installed on either turf or cropland, as referenced in the dashboard above. (Click on the bottom left of the map to access the legend)

Map of Land Use in 2013 Before Solar

Example of One Solar Field with Prior Land Use vs Current Land Cover:

This feature provides a closer look at one of the solar fields. One can see the prior land cover which was turf land. With Wayback imagery, there is an image of what the land looks like now, with ground-mounted solar.

New legislation and incentives could explain the growing number of ground-mounted solar fields in the state. Those incentives include:

• The Federal Business Energy Investment Tax Credit allows 30% tax credit for installed solar energy systems placed in service after December 31, 2019.

• The Modified Accelerated Cost-Recovery System (MACRS) allows businesses to recover investments in solar technologies through depreciation deductionss.

• The Federal Residential Renewable Energy Tax Credit allows taxpayers to claim a credit of 30% of qualified expenditures for a solar energy system that serves a residence. Tax credit applicable to systems in service by December 31, 2019.

• The Maryland Energy Association (MEA) offers grants for clean energy systems installed at primary residences through its Residential Clean Energy Grant Program
• MEA offers grants for clean energy systems installed at businesses, local governments, and non-profits through its Commercial Clean Energy Grant Program​
• For systems operating before December 31, 2015, MEA’s Clean Energy Production Tax Credit offers a state income tax credit for electricity generated by solar photovoltaic systems (over 20 kW) of 0.85 cents per kWh.
• These credits can be claimed over a period of five years. (citation)

The yellow dots represent the different areas where there is a permit to build solar fields in Baltimore County. They mean that in the county's database, there was a permit for solar panels issued for that house. Most likely, rooftop solar will exist in these locations. It's possible they got a permit but have not yet built the solar panels. The optimal sites according to the map are Baltimore County landfills (Hernwood and Parkton), parking garages, identified under-utilized industrial parcels, degraded land solar potential parcels. The preferred sites are ground-mounted, rooftop, or canopy solar. Of the 7 solar fields present in the county, the suitability map shows only 2 built on preferred sites. No solar fields are present on optimal sites.

Map of Solar Suitability Analysis in Baltimore County

The density map shows the area with the biggest concentration of ground-mounted solar fields within the Baltimore Metropolitan Area. Looking at the map, the highest concentration is shared between Harford and Baltimore County. The distribution in Queen Anne's is evenly spread within county boundaries. Anne Arundel County has the least amount of solar fields even though they are concentrated around the same area.

Solar Density Map

The map below shows the diversity distribution cross-referenced with the density map. Ground-mounted solar fields are implanted for the most part in areas with the highest diversity index. Of the study area, Carroll County is the least diverse and, Harford County is the most diverse.

Diversity vs Solar Density Map

From the population density map, there is an even distribution regarding the location of ground-mounted solar fields. Carroll and Queen Anne's county have the lowest population density (0 - 1,000 people per sq mi). Anne Arundel, Harford, Howard, and Baltimore County are more diverse (about 1,000 - 8,400 people per sq mi).

The percent of population living below the Federal Poverty Line in our area of study is located in Baltimore City.

Population Density vs Solar Density Map

All the 36 ground-mounted solar fields are implanted within these three mean income levels:

• $62,200 - 109,600
• $109,600 - 156,900
• $156,900 - 394,000

Solar fields are installed in the three highest income brackets.

Average Income vs Solar Density Map

All 36 ground-mounted solar fields are implanted within these three median income levels:

• $48,600 - 84,800
• $84,800 - 121,000
• $121,000 - 200,100

• In the State of Maryland, the median income for households is $76,067 (Census Bureau, American Community Survey 5-Year Estimates 2012-2016). In the ranking of counties from highest to lowest figures, Howard County is ranked first, Anne Arundel fourth, Carroll County sixth, Queen Anne's eighth and, Harford County 10th. All of these counties are above the State of Maryland median income level. Baltimore County is ranked 12th and Baltimore City 22th (out of 24).

Median Income vs Solar Density Map

There are 36 ground-mounted solar fields in the Baltimore Metropolitan Area. Most of these solar fields were built on cropland and turf between 2016 to 2018. The State of Maryland has made many efforts to increase solar energy production along with other renewable energy. Many of the State's incentives could explain the increase of solar fields in that time period. From the suitability analysis done on Baltimore County, it was established that only two of the solar fields were built on preferable sites. From the demographics analysis, the majority of the solar fields are present in higher-income areas but less populated.

Living in this information and data age, every individual must develop basic data analytics skills. I came into the internship willing to learn because most of these concepts were new to me. The job duties consisted of analyzing databases, creating web maps and story maps, and reviewing state legislation. We received training on data processing best practices for analysis, geographic information systems, effective and ethical data visualization techniques for applications that help communities in Baltimore. I learned more in-depth about Geographic Information Systems (GIS). We analyzed spatial data, learned how to create maps, learned about the different software available to create maps, and navigated on official government sites to retrieve the data. A big part of data visualization techniques is to analyze. I learned to distinguish patterns just by looking at data. I also learned how to properly analyze spatial data using ArcGIS online and Artificial Intelligence (machine learning and deep learning). All the interns learned coding languages, especially python, Jupiter notebook, Google collaborative, and GitHub. Some aspects of data analysis are very hands-on. Being remote brought its own set of challenges because observing someone do the process in person is easier for learning. I had to adapt and understand how to use complex programs from home. I also struggled with staying motivated. Working remotely gives much independence that could be mistakenly taken for “free time.” Students were expected to work a maximum of 8 hours a week. Juggling between school, work, and my internship was very challenging. The experience taught me how to stay focused and productive. I had to organize myself to follow my schedule, work on my blog, and attend the weekly meetings and training. In conclusion, this experience has been an advantageous one for me. I can safely say that my analytical skills have improved tremendously. Even though I made incredible progress, I still have much to learn in data science. Nevertheless, I am more equipped than I was at the start of the semester.

• How to copy data in AGOL: https://doc.arcgis.com/en/arcgis-online/create-maps/copy-save-layers.htm#:~:text=Open%20the%20map%20in%20Map,click%20Details%2C%20and%20click%20Contents.&text=%2C%20and%20click%20Copy.,layer%20as%20a%20new%20item
• How to make a web map on AGOL: https://learn.arcgis.com/en/projects/create-a-map/
• How to make an application from a web map in AGOL: https://doc.arcgis.com/en/arcgis-online/create-maps/create-map-apps.htm
• How to add data to AGOL web map: https://doc.arcgis.com/en/arcgis-online/create-maps/add-layers.htm#:~:text=To%20get%20started%20adding%20layers,files%2C%20and%20add%20map%20notes.
• How to view older imagery: https://www.esri.com/arcgis-blog/products/arcgis-online/mapping/using-world-imagery-wayback/