Solar Potential in Barcelona through Earth Observation

class: center, middle, title-slide

# Solar Potential in Barcelona through Earth Observation

## Group RGB

### Christy Choi, Entao Wang, Jiaying Li, Yanxi Ren, Tianyi Chen

### CASA0023 Remotely Sensing Cities and Environments

### 2026/03/09 (updated: 23/03/2026)

### <svg aria-hidden="true" role="img" viewBox="0 0 496 512" style="height:1em;width:0.97em;vertical-align:-0.125em;margin-left:auto;margin-right:auto;font-size:inherit;fill:white;overflow:visible;position:relative;"><path d="M165.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6zm-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3zm44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9zM244.8 8C106.1 8 0 113.3 0 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C428.2 457.8 496 362.9 496 252 496 113.3 383.5 8 244.8 8zM97.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1zm-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7zm32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1zm-11.4-14.7c-1.6 1-1.6 3.6 0 5.9 1.6 2.3 4.3 3.3 5.6 2.3 1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2z"/></svg> [GitHub Repository](https://github.com/christychoicc/RSCE_GROUP)
---
# Barcelona
.pull-left-wide[
<img src="images/solar_irradiation.jpg" alt="" width="90%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 1: Spain Solar Irradiation (Adapted from:  <a href="https://www.sciencedirect.com/science/article/pii/S1364032111002620">Sumper et.al.(2011)</a>)
</p>
]
.pull-right-narrow[

- Located in northeast Spain, benefiting from a Mediterranean climate

- High solar irradiation of around 1,500–1,700 kWh/m²/year

- Average solar radiation of around 4.12-4.56 kWh/m²/day, offering excellent conditions for rooftop PV installation

- Highest solar resource cities in Europe, favourable for solar energy generation year-round
]
<p style="font-size: 0.6em; text-align: left; margin-top: 0;">
Source: <a href="https://pvgis.com/en/blog/solar-energy-spain-guide-installers"> PVGIS (2024)</a>, <a href="https://www.mdpi.com/2071-1050/9/2/247"> Jiménez-Torres (2017)</a> 
</p>

???
Barcelona has very favourable conditions for solar energy. Its Mediterranean climate and high solar irradiation make it one of the stronger urban environments in Europe for year round rooftop photovoltaic generation.

---
# Barcelona
### High Solar Potential, Low Energy Self-sufficiency
.pull-left-wide[
- **Potential**: 7.9M m² suitable roof area, where flat and slightly inclined terraces account for 67% of total roof surface; Generation potential of 1,191 GWh/year, around 60% of domestic electricity demand

- **Low realisation**: City produces just **1%** of its own energy, only **5%** of households have solar installations

- **Core Issue**: Barcelona has potential but the rooftops are under utilised for the city 
]
.pull-right-narrow[
<img src="images/solar_panels_barca.webp" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 2: Solar Panels (Source: <a href="https://www.theguardian.com/world/2022/jul/24/barcelona-school-residents-create-solar-energy-community">the Guardian (2022)</a>)
</p>
]

.clearfix[]
<div style="background: #722f37; color: white; padding: 10px 16px; border-radius: 6px; font-size: 0.9em;">
  Barcelona has strong rooftop potential for solar expansion, but current uptake remains far below this potential.
</div>

<p style="font-size: 0.6em; text-align: left; margin-top: 0;">
Source: <a href="https://www.energia.barcelona/en/generate-energy/generar-energia/map-how-much-energy-can-you-generate">Energia Barcelona(n.d.)</a>, <a href="https://netzerocities.app/resource-4178">Ajuntament de Barcelona(2024)</a> 
</p>

???
Despite this strong potential, Barcelona remains far from being energy self-sufficient. The city has about 7.9 million square metres of suitable roof area that could support solar panels, which is enough to generate roughly 60% of all domestic electricity demand. However, in reality Barcelona produces only about 1 percent of the energy it consumes, and only 5 percent of households have solar installations. So the core issue is not resource availability, but the gap between potential and implementation.

---
# Policy Goals and Agenda
.panelset[
.panel[.panel-name[Global Agenda]

.pull-left-3[
<img src="images/sdg7.jpg" alt="" width="100%" style="display: block; margin: auto 0 auto auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 4: SDG 7: increases access to affordable, reliable and clean energy; Source: <a href="https://www.un.org/sustainabledevelopment/news/communications-material/">United Nations(2024)</a>
</p>
]
.pull-middle-3[
<img src="images/sdg11.jpg" alt="" width="100%" style="display: block; margin: auto 0 auto auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 5: SDG 11: supports a more sustainable and resilient urban system; Source: <a href="https://www.un.org/sustainabledevelopment/news/communications-material/">United Nations(2024)</a>
</p>
]
.pull-right-3[
<img src="images/sdg13.jpg" alt="" width="100%" style="display: block; margin: auto 0 auto auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 6: SDG 13: reduces emissions and strengthens climate action in the built environment; Source: <a href="https://www.un.org/sustainabledevelopment/news/communications-material/">United Nations(2024)</a>
</p>
]
]
.panel[.panel-name[EU Policy]
- **The European Green Deal, Renewable Energy Directive**
  - Binding target of 42.5% renewable energy across the EU by 2030
  
- **EU Solar Energy Strategy**
  - Target of 600 GW of solar PV capacity across the EU by 2030
  
- **Renovation Wave Strategy**
  - Target to renovate 35 million buildings by 2030 (including improving energy performance)

<p style="font-size: 0.6em; text-align: left; margin-top: 0;">
Source: <a href="https://energy.ec.europa.eu/topics/renewable-energy/renewable-energy-directive-targets-and-rules/renewable-energy-directive_en">Energy(2023)</a>, <a href="https://energy.ec.europa.eu/topics/energy-efficiency/energy-performance-buildings/renovation-wave_en#:~:text=Renovating%20both%20public%20and%20private,to%20regional%20and%20local%20actors 
">Energy(2020)</a>, <a href="https://energy.ec.europa.eu/topics/energy-efficiency/energy-performance-buildings/renovation-wave_en#:~:text=Renovating%20both%20public%20and%20private,to%20regional%20and%20local%20actors 
">European Parliament(2024)</a>
</p>

]

.panel[.panel-name[National Policy]
- Integrated National Energy and Climate Plan (NECP) 2023-2030

- Targets 81 % renewable energy in electricity generation and 48 % renewables on final energy use by 2030
  
  - Installation of 76 GW of photovoltaics (with 19 GW for self-consumption) by 2030
  
  - Lower household energy expenditure to 5.7% by 2030

<p style="font-size: 0.6em; text-align: left; margin-top: 0;">
Source: <a href="https://www.miteco.gob.es/es/prensa/ultimas-noticias/2024/septiembre/el-gobierno-aprueba-la-actualizacion-del-plan-nacional-integrado.html">Ministerio para la Transición Ecológica y el Reto Demográfico. (2023)</a>
</p>
]

.panel[.panel-name[Municipal Policy]
.pull-left-wide[
**Barcelona Climate Plan 2018-2030**
.pull-left-3[
<img src="images/bcp5.png" alt="" width="100%" style="display: block; margin: auto 0 auto auto;" />
<p style="font-size: 0.4em; text-align: center; margin-top: 0;">
Figure 7:Line of Action 5, recover terrace roofs and increase PV generation; Source: <a href="https://www.barcelona.cat/barcelona-pel-clima/sites/default/files/climate_plan_maig.pdf">Ajuntament de Barcelona(2018)</a>
</p>
]
.pull-middle-3[
<img src="images/bcp9.png" alt="" width="100%" style="display: block; margin: auto 0 auto auto;" />
<p style="font-size: 0.4em; text-align: center; margin-top: 0;">
Figure 8: Line of Action 9, renewables in public space and municipal installation; Source: <a href="https://www.barcelona.cat/barcelona-pel-clima/sites/default/files/climate_plan_maig.pdf">Ajuntament de Barcelona(2018)</a>
</p>
]
.pull-right-3[
<img src="images/bcp12.png" alt="" width="100%" style="display: block; margin: auto 0 auto auto;" />
<p style="font-size: 0.4em; text-align: center; margin-top: 0;">
Figure 9: Line of Action 12,green jobs, self-employment and climate-related economic transition; Source: <a href="https://www.barcelona.cat/barcelona-pel-clima/sites/default/files/climate_plan_maig.pdf">Ajuntament de Barcelona(2018)</a>
</p>
]
]
.pull-right-narrow[
**Strategy for Energy Transition**: increase renewable electricity generation by 10% and PV capacity fivefold, promote and financially support installations

**Tax Incentives**: up to 50% IBI on property, 95% ICIO on construction and installation and 50% IAE on business tax reduction for solar installations

<p style="font-size: 0.4em; text-align: left; margin-top: 0;">
Source: <a href="https://ajuntament.barcelona.cat/ecologiaurbana/en/what-we-do-and-why/energy-and-climate-change/strategy-energy-transition">Ajuntament de Barcelona(n.d.)</a>, <a href="https://www.barcelona.cat/barcelona-pel-clima/en/pla-clima/pla-transicio-energetica">Barcelona for Climate(2025)</a>, <a href="https://www.energia.barcelona/en/generate-energy/generate-energy/grants-and-allowances">Barcelona Energia(n.d.)</a>
</p>
]
]
]

???
To understand the wider drivers for improvement, we looked at the policy context across four levels. 
At the global level, the project supports SDG 7, SDG 11, and SDG 13 by promoting clean energy, more sustainable cities, and climate action. 
At the EU level, it aligns with the European Green Deal, the Renewable Energy Directive, the EU Solar Energy Strategy, and the Renovation Wave. 
At the national level, it supports Spain’s updated energy plan, which targets major PV expansion and much higher renewable electricity generation by 2030.
At the municipal level, it complements Barcelona’s own climate and energy transition strategies, which promote rooftop recovery, renewable installations, and tax incentives for solar installation.

---
# Key Problems and Issues

Despite strong policy frameworks, Barcelona still faces challenges...

- Renewable energy generation underdeveloped at both local and regional scale
- Current installations are concentrated in public and municipal buildings
- Ageing, insufficient housing stock, 81% of properties have a low energy performance rating (E, F, or G)
- Energy poverty concerns, energy bills exceed 10% of household income for many households
- Renewable transition is slow; energy system still relies on centralised, non-renewable sources
- The Iberian Peninsula Blackout in April 2025 exposed the fragility of Spain's energy system, underlining the urgency of strengthening energy sources and distributed solar installation

<div style="background: #722f37; color: white; padding: 10px 16px; border-radius: 6px; font-size: 0.9em; margin-top: 1em;">
The key challenges Barcelona faces are an ageing building stock, uneven solar installation, and energy vulnerability, highlighting the need for targeted implementation. 
</div>

<p style="font-size: 0.6em; text-align: left; margin-top: 0.5em;">
Sources: <a href="https://ajuntament.barcelona.cat/barcelonallibres/en/noticia/ageing-housing-with-poor-energy-efficiency-but-a-huge-potential-for-improvement-1188771">Ajuntament de Barcelona (n.d.)</a>, <a href="https://bcnroc.ajuntament.barcelona.cat/jspui/bitstream/11703/134731/2/Climate%20City%20Contracte%20of%20Barcelona%20%28Annexes%29.pdf">Ajuntament de Barcelona(n.d.)</a>, <a href="https://www.energia.barcelona/en/generate-energy/generar-energia/map-how-much-energy-can-you-generate">Barcelona Energia(n.d.)</a>, <a href="https://ajuntament.barcelona.cat/barcelonallibres/en/noticia/ageing-housing-with-poor-energy-efficiency-but-a-huge-potential-for-improvement-1188771">Ajuntament de Barcelona(n.d.)</a>
</p>

???
Despite strong policy support, major barriers still exists. Renewable generation is still underdeveloped, and existing solar installations are concentrated mainly on public and municipal buildings rather than being widely distributed across the city. 
At the same time, Barcelona has an ageing and energy-inefficient housing stock, while energy poverty remains a serious social concern. 
This matters because more than two thirds of buildings still meet the structural conditions needed for solar installation, particularly through flat or gently sloped rooftops. 
So the challenge is not just about technology, but also about where need is greatest and who benefits most.

---
class: inverse, center, middle
# What is missing?
### Barcelona has a strong policy commitment, but lack of a city-scale operational system to identify where rooftop solar should be prioritised to maximum social, economic, and environmental benefit!

.clearfix[]

<div style="background: #F1BF00; color: #722f37; padding: 10px 16px; border-radius: 6px; font-size: 0.9em; text-shadow: none;">
Therefore, our project aims to address this implementation gap by using Earth Observation to identify and prioritise rooftop solar opportunities across Barcelona.
</div>

???
So, what is missing from current policy is an operational system at city scale. Barcelona has strong commitments, but lacks a practical method to identify where PV should be prioritised for the greatest social, economic, and environmental benefit. 
Our project objective is therefore to further develop the existing dashboard by integrating an Earth Observation workflow to identify and prioritise rooftop solar opportunities across Barcelona. By doing so, the project would support more targeted decision-making, accelerate solar installments, helping the city to progress towards its wider energy, climate, and equity goals.

---
class: inverse, center, middle
## What is missing from the operational system?
### Analysis of Barcelona’s Current Solar Dashboard

--
Do you want to know how much energy you can generate in your home?

Search for your building and discover its energy potential:

???
ICY:

Do you want to know how much energy you can generate in your home?

That is exactly the question Barcelona’s official portal answers today.

BUT, let’s bring a critical eye to this:

---

/* slider panel */
.scroll-box {
  height: 400px;
  overflow-y: auto;
  padding: 6px;
  border-left: 1px solid #802222;
}
/* Markdown in div */
.scroll-box ul { display: block; }
</style>

# Analysis of Barcelona’s Current Solar Dashboard

.left-70[
### .center[Dashboard Overview]

.footnote[Source: [Barcelona Energia Mapa](https://www.energia.barcelona/ca/generar-energia/generar-energia/mapa-quanta-energia-podeu-generar)]
]

.right-30[
### .center[Key Metrics]

.scroll-box[
#### ☀️ Solar Resource Potential
- **Incident Solar Radiation Radiation:** Qualitative rating  ***(e.g., "Very Good")***

- **Solar Irradiated Surface Area:** Total roof area (`$m^2$`)

#### ⚡ Electric Performance Data
- **Usable Area:** Net area
- **Capacity:** Peak power (`$kW$`)
- **Generation:** `$kWh/year$`
- **Coverage:** `$>100\%$` needs

#### 🌍 Environmental Impact
- **GHG Savings:** `$kgCO_2eq/year$`

#### 💰 Economic Assessment
- **Investment:** CapEx (€)
- **Maintenance:** €/year
- **Financial Savings:** €/year
]
]

???
ICY:

On the left is the current dashboard.

It answers the questions of

How is the soalr potential What is my Capacity of my roof?

How many Energy will I get?

how much we can save On carbon footprint.

The problem is Its only show the potential estimation  //

---
class: inverse, center, middle
# Research Question: 4 GAPs

--
### 1 The "Static" Gap: 
Current data dashboard provides a 'snapshot' of theoretical potential but cannot monitor the actual progress of solar installations in real-time.

(Without knowing what is already "on the roof," the city cannot track its progress toward Global Development Goals (e.g., SDG 7).) 
--

### 2 The "Precision" Gap: 
Small obstacles (chimneys, HVAC) are often ignored, leading to a 20% overestimation of usable area. (resulting in unrealistic financial expectations for citizens and flawed energy planning for the city.)

???
We’ve identified four critical GAPs:

fail to track real progress.

ignores rooftop obstacles

and we are **missing the 'Social' and 'Future' context**
---
class: inverse, center, middle
# Research Question: 4 GAPs

### 3 The "Integration" Gap:  
The dashboard does not place Barcelona in a broader background context (e.g., comparable spatial and policy indicators).
--

### 4 The "Future" Gap:
The dashboard provides limited support for understanding future development potential.

???
We’ve identified four critical GAPs:

fail to track real progress.

ignores rooftop obstacles

and we are **missing the 'Social' and 'Future' context**

---
# Our Solution

1. Use remote sensing data to reduce the cost and turnaround time of data updates.
2. Use Machine Learning methods to better understand the current solar panel distribution.
3. Fit Barcelona into a broader policy context by providing additional policy-relevant layers.
4. Add more ways to calculate and prioritise the huge potential rooftop area.

???
BY fusing Remote Sensing with Machine Learning, we can get across the gaps and   
 
Turn it to multi-layered policy tool that **connects rooftop potential to urban reality**

---
class: reverse, center, middle
# Dashboard Redesign   
## How can we improve the operational system?

???
ICY:  
 
SO We’ve redesigned the operational system based on three fundamental shifts:

### 1. Static---Dynamic

### 2. Theoretical---Practical

### 3. Isolated---Intergrated

---
# Workflow
<img src="images/workflow_final.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 10: Model Workflow
</p>

---
# Methodology - Step 1.1: U-net

### **1. Symmetric Encoder-Decoder Structure**:

<ul>
  <li><b>Contracting Path (Encoder)</b>: Captures multi-scale spectral and geometric features of PV panels through deep feature extraction.
  
  <li><b>Extensive Path (Decoder)</b>: Recovers feature maps to original spatial resolution, enabling centimeter-level precision in localization.
</ul>

### **2. Cross-Scale Skip Connections**
        
### **3. Robustness in Complex Urban Environments**

### **4. End-to-End High-Efficiency Mapping**

---
# Methodology - Step 1.1: U-net
<img src="images/U-net.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 11: Results of U-net model (Source: <a href="https://www.mdpi.com/2076-3417/11/14/6524"> Pérez-González et.al.,(2021) </a>) 
</p>
---
# Methodology - Step 1.1: U-net
<img src="images/U-net1.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 12: Results of U-net model (Source: <a href="https://www.mdpi.com/2076-3417/11/14/6524"> Pérez-González et.al.,(2021) </a>) 
</p>

---
# Methodology - Step 1.2: OBIA
**1. Multiresolution Segmentation** 
  - Aggregates spectrally similar pixels into geographically meaningful objects to mitigate "salt-and-pepper" noise.
        
**2. Geometric Constraints**
  - Leverages physical attributes—including aspect ratio, rectangularity, and area—to filter out non-PV entities (e.g., circular water tanks).
        
**3. Spectral-Spatial Integration**
  - Evaluates object regularity alongside spectral signatures to significantly reduce false-positive rates.
        
**4. Vectorized Output** 
  - Facilitates the direct export of GIS-compliant polygonal boundaries, streamlining precise surface area calculations.

---
# Methodology- Step 1.2: OBIA

.pull-left[
<img src="images/OBIA.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 13: Workflow of OBIA model
</p>]

.pull-right[
<img src="images/OBIA_4.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 14: Result of SLIC (Source: <a href="https://jakubnowosad.com/ogh2021/#10"> Ladisa et.al.,(2022) </a>)
</p>]

---
# Methodology - Step2: Data
<div style="height: 30px;"></div> 
<img src="images/lidar5.png" alt="" width="90%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure15: Models (DSM and DTM) from ICGC  (Source: 
  <a href=https://www.icgc.cat/en/Geoinformation-and-Maps/Data-and-products/Elevations/Territorial-elevations/Territorial-LiDAR> Institut Cartogràfic i Geològic de Catalunya (ICGC) </a>)
</p>

???
Now, we move on to **Step 2**, which focuses on the LiDAR workflow.

we **downloaded** the DEM data from the ICGC platform.

---
# Methodology - Step2: Lidar Workflow
<img src="images/lidar1.png" alt="" width="90%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure16: LiDAR Workflow (Source: <a href="https://www.sciencedirect.com/science/article/abs/pii/S0960148117303324"> Suomalainen et.al.,(2017) </a>)
</p>

???
**This step has two main parts: analyzing the rooftops and calculating the** solar radiation

Then, we clipped the data to **cut out** streets and trees, **leaving just** the rooftops.

---
# Methodology- Step 2.1: Slope

<img src="images/slop3.png" alt="" width="90%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure17: Calculate the slope (in degrees)  (Source: <a href="https://www.sciencedirect.com/science/article/abs/pii/S0960148117303324"> Suomalainen et.al.,(2017) </a>)
</p>

???
For these clipped rooftops, we calculate the slope and aspect for each location. and fin out the rooftops that are physically unsuitable for installation.

---
# Methodology - Step2: Lidar Workflow
<img src="images/lidar4.png" alt="" width="90%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure16: LiDAR Workflow (Source: <a href="https://www.sciencedirect.com/science/article/abs/pii/S0960148117303324"> Suomalainen et.al.,(2017) </a>)
</p>

???
Next, we turn the rooftops into points.  Using the Point Solar Radiation tool in ArcGIS,  calculate the exactly annual solar radiation for each spot.

---
# Methodology -Step2.2: Solar Radiation
<img src="images/lidar2.png" alt="" width="90%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure18: LiDAR Solar Detection (Source: <a href="https://ars.els-cdn.com/content/image/1-s2.0-S0960148117303324-gr7.jpg"> Els-cdn.com. (2017) </a>)
</p>

???
This picture easily shows us the areas shaded by trees or taller buildings.

---

???
Finally, we subtract existing panels, shaded areas, and unsuitable slopes from the total rooftop area. This gives us the rooftops with true solar potential.

---
# Methodology - Step 3: Multispectral Indices
*Assessing Urban Resilience and Environmental Context using Sentinel-2*

.panelset[
.panel[.panel-name[🌿 NDVI (Vegetation)]
**Definition:** The Normalized Difference Vegetation Index quantifies vegetation health and density by measuring the difference between near-infrared (which vegetation strongly reflects) and red light.  
**Formula:** `$NDVI = \frac{NIR - Red}{NIR + Red}$`

**Purpose in Project:** Identifies existing green roofs and street-level tree canopies. It helps exclude tree-shaded areas from PV potential and flags vegetation-poor neighborhoods (low NDVI) as high-priority zones for **"Biosolar" (PV + Green Roof) subsidies** to mitigate heat risk.

**Literature Support:** [Weng et al. (2004)](https://doi.org/10.1016/j.rse.2003.11.005) demonstrated a strong negative correlation between NDVI and Land Surface Temperature (LST), validating NDVI as a crucial metric for mapping urban heat vulnerability.
]

.panel[.panel-name[💧 NDWI (Water)]
**Definition:** The Normalized Difference Water Index enhances water features while eliminating soil and terrestrial vegetation features.  
**Formula:** `$NDWI = \frac{Green - NIR}{Green + NIR}$`

**Purpose in Project:** Acts as a strict **spatial mask**. Barcelona is a coastal city; NDWI automatically delineates the Mediterranean Sea, inland water bodies, and large rooftop swimming pools, ensuring these flat surfaces are not falsely identified as viable solar potential areas by the algorithms.

**Literature Support:** [McFeeters (1996)](https://doi.org/10.1080/01431169608948714) introduced NDWI to successfully delineate open water features and suppress non-water noise in urban and environmental remote sensing.
]

.panel[.panel-name[🏗️ NDBI (Built-up)]
**Definition:** The Normalized Difference Built-up Index highlights urban areas of barren soil, concrete, and asphalt, which reflect shortwave infrared (SWIR) stronger than NIR.  
**Formula:** `$NDBI = \frac{SWIR - NIR}{SWIR + NIR}$`

**Purpose in Project:** Maps the density of the built environment (e.g., the Eixample district). High NDBI strongly correlates with **severe Urban Heat Islands (UHI)**. Overlapping high NDBI with our "Solar Gap" allows the city to target rooftop PV as a physical shading mechanism to cool down the most exposed concrete blocks.

**Literature Support:** [Guo et al. (2015)](https://doi.org/10.1016/j.landurbplan.2014.11.007) explored the quantitative relationship between urban biophysical composition and heat, proving that NDBI is the primary driver of increased Land Surface Temperature in dense urban clusters.
]
]

---
class: inverse, center, middle
# NEW Dashboard!

???
ICY:  
  
Let's  took all those methodologies and we fused them into one single, powerful interface.

/* SLIDER */
.scroll-box-dl {
  height: 450px;
  overflow-y: auto;
  padding: 10px;
  border-left: 1px solid #802222;
  background: rgba(128, 34, 34, 0.03);
}

/* PIXEL */
.img-container-dl {
  width: 100%;
  height: 340px;
  background: #f0f0f0;
  display: flex;
  align-items: center;
  justify-content: center;
  overflow: hidden;
  border-radius: 4px;
  border: 1px solid #ddd;
}

.img-container-dl img { transition: transform 0.3s ease; }

/* BUTTON */
.nav-btns-dl {
  display: flex;
  justify-content: center;
  gap: 8px;
  margin-top: 10px;
  flex-wrap: wrap;
}

.nav-btns-dl button {
  font-size: 11px;
  padding: 5px 10px;
  cursor: pointer;
  border: 1px solid #802222;
  background: white;
  border-radius: 4px;
}

.active-dl { background: #802222 !important; color: white; }

/* FOLDER */
.dl-details {
  background: white;
  padding: 8px;
  margin-bottom: 8px;
  border-radius: 4px;
  border: 1px solid #eee;
  font-size: 0.8em;
}
.dl-details summary { font-weight: bold; cursor: pointer; color: #802222; }
</style>

---

# Step 1: DL Existing Asset Detection

.left-70[
### .center[DL Segmentation Layers]

<div class="nav-btns-dl">
  <button class="active-dl" onclick="swapDL('images/01StatusLayerML.png', 'Layer 1: Input High Resolution Image', this)">01. Input</button>
  <button onclick="swapDL('images/02StatusLayerMLOutcome.png', 'Layer 2: DL Segmentation', this)">02. Outcome</button>
  <button onclick="swapDL('images/023StatusLayerMLOutcome.png', 'Layer 3: Final Detection- Results of U-net/OBIA', this)">03. Final</button>
</div>

<div class="nav-btns-dl" style="margin-top:10px;">
  <button onclick="zoomDL(1.2)">🔍 Zoom In</button>
  <button onclick="zoomDL(0.8)">🔍 Zoom Out</button>
  <button onclick="resetDL()">🔄 Reset</button>
</div>

<p id="dl-caption" style="text-align:center; font-size:0.8em; color:#802222; margin-top:5px; font-weight:bold;">Current: Layer 1: Input Status</p>
]

.right-30[
### .center[Key Metrics]

.scroll-box-dl[

<details open class="dl-details">
<summary>🔍 Asset Detection</summary>

- Existing PV Area: PV Area Detected(`$m^2$`)

<br>

- Data Source: VHR Satellite Imagery

<br>

- Input Resolution: 0.3m per pixel

<br>

- Last Updated: 2024-09-20

</details>

<details class="dl-details">
<summary>⚡ Capacity Info</summary>

- Installed Capacity: 500 kW

<br>

- Generation: 1,200 kWh/year

<br>

- Coverage:> 100% needs

</details>

<details class="dl-details">
<summary>🛠️ Model Config</summary>

- CNN: ResNet-50 / Swin-T

</details>

]
]

function swapDL(path, text, btn) {
  dlImg.src = path;
  dlCap.innerText = "Current: " + text;
  resetDL();
  document.querySelectorAll('.nav-btns-dl button').forEach(b => b.classList.remove('active-dl'));
  btn.classList.add('active-dl');
}

function zoomDL(f) {
  dlScale *= f;
  dlImg.style.transform = "scale(" + dlScale + ")";
}

function resetDL() {
  dlScale = 1;
  dlImg.style.transform = "scale(1)";
}
</script>

???

Here is the fix of Static Gap.

We input Very High Resolution Satellite Imagery and our model's outcome to see **how the AI highlights existing** solar panels **in red**.

and updated the system with three critical variable groups:

---

/* SLIDER - LiDAR */
.scroll-box-lidar {
  height: 450px;
  overflow-y: auto;
  padding: 10px;
  border-left: 1px solid #2980b9;
  background: rgba(41, 128, 185, 0.03);
}

/* PIXEL */
.img-container-lidar {
  width: 100%;
  height: 340px;
  background: #f0f0f0;
  display: flex;
  align-items: center;
  justify-content: center;
  overflow: hidden;
  border-radius: 4px;
  border: 1px solid #ddd;
}

/* BUTTON */
.nav-btns-lidar {
  display: grid;
  grid-template-columns: repeat(3, 1fr);
  gap: 5px;
  margin-top: 10px;
}

.nav-btns-lidar button {
  font-size: 11px;
  padding: 5px;
  cursor: pointer;
  border: 1px solid #2980b9;
  background: white;
  border-radius: 4px;
}

.active-lidar { background: #2980b9 !important; color: white; }

/* FOLDER */
.lidar-details {
  background: white;
  padding: 8px;
  margin-bottom: 8px;
  border-radius: 4px;
  border: 1px solid #eee;
  font-size: 0.8em;
}
.lidar-details summary { font-weight: bold; cursor: pointer; color: #2980b9; }
</style>

# Step 2: LiDAR Precision Refinement

.left-70[
### .center[3D Geometric Analysis]

<div class="nav-btns-lidar">
  <button onclick="swapL2('images/03ShadeMor.png', 'Morning Shade', this)">🌅 Morning</button>
  <button onclick="swapL2('images/031Shadenoon.png', 'Noon Shade', this)">☀️ Noon</button>
  <button class="active-lidar" onclick="swapL2('images/032Shadeeve.png', 'Evening Shade', this)">🌆 Evening</button>
  <button onclick="swapL2('images/04Slope.png', 'Slope Analysis', this)">📐 Slope</button>
  <button onclick="swapL2('images/05Aspect.png', 'Aspect Analysis', this)">🧭 Aspect</button>
  <button onclick="swapL2('images/06Elevation.png', 'Elevation Analysis', this)">⛰️ Elevation</button>
</div>

<p id="lidar-caption-s2" style="text-align:center; font-size:0.8em; color:#2980b9; margin-top:5px; font-weight:bold;">Current: Evening Shade</p>
]

.right-30[
### .center[Key Metrics]

.scroll-box-lidar[

<details open class="lidar-details">
<summary>📐 LiDAR Analysis</summary>
- Average Tilt: `$13.5^\circ$`
<br>
- Azimuth: `$155^\circ$` (180°=South)
<br>
- Shadows: Dynamic ray-tracing.
<br>
- Detection: 10cm+ rooftop obstacles.
</details>

<details class="lidar-details">
<summary>☀️ Effective Potential</summary>
- Solar Hours: 10h (0-24)
<br>
- Obstacle Ratio: 0.36
<br>
- Net Area: 12 `$m^2$`
</details>

] ] <script>

function swapL2(path, displayName, btn) {
  var img = document.getElementById('lidar-display-s2');
  var caption = document.getElementById('lidar-caption-s2');
  
  if(img) img.src = path;
  if(caption) caption.innerText = "Current: " + displayName;

var buttons = btn.parentElement.querySelectorAll('button');
  buttons.forEach(b => b.classList.remove('active-lidar'));
  btn.classList.add('active-lidar');
}
</script>

???
ICY:

Using Ray-tracing, that filters out shadows and identifies Slope to subtract 3D obstacles like chimneys/HVAC UNITS.

We focus on four key dimensions:
shade  
slope  
aspect  
elevation

---

/* SLIDER - API */
.scroll-box-api {
  height: 450px;
  overflow-y: auto;
  padding: 10px;
  border-left: 1px solid #27ae60;
  background: rgba(39, 174, 96, 0.03);
}

/* PIXEL */
.img-container-api {
  width: 100%;
  height: 340px;
  background: #f0f0f0;
  display: flex;
  align-items: center;
  justify-content: center;
  overflow: hidden;
  border-radius: 4px;
  border: 1px solid #ddd;
}

/* BUTTON */
.nav-btns-api {
  display: grid;
  grid-template-columns: repeat(4, 1fr);
  gap: 5px;
  margin-top: 10px;
}

.nav-btns-api button {
  font-size: 10px;
  padding: 5px;
  cursor: pointer;
  border: 1px solid #27ae60;
  background: white;
  border-radius: 4px;
}

.active-api { background: #27ae60 !important; color: white; }

/* FOLDER */
.api-details {
  background: white;
  padding: 8px;
  margin-bottom: 8px;
  border-radius: 4px;
  border: 1px solid #eee;
  font-size: 0.8em;
}
.api-details summary { font-weight: bold; cursor: pointer; color: #27ae60; }
</style>

# Step 3: Syn API & Spectral Analysis

.left-70[
### .center[Multispectral Index & Land Use]

<p id="api-caption" style="text-align:center; font-size:0.8em; color:#27ae60; margin-top:5px; font-weight:bold;">Current: Layer: NDVI (Vegetation Index)</p>
]

.right-30[
### .center[Key Fields]

.scroll-box-api[

<details open class="api-details">
  <summary>🛰️ Spectral Reflectance</summary>
  - NDVI: Identifies green roof potential.
<br>

- NDBI: Distinguishes roof materials vs. soil.
<br>

- NDWI: Detects ponding or water features.
</details>

<details class="api-details">
  <summary>🏠 Context</summary>
  - Roof_Material: Concrete, metal, or tile.
<br>
  - Land_Use_Class: Residential, Industrial, or Municipal classification.
</details>

<details class="api-details">
  <summary>🏛️ Regulatory Constraints</summary>
  - Heritage_Status: 0/1 (Restricted zones)
<br>
  - Data_Confidence: API cross-validation score
</details>

]
]

---
# Smart Solar Analytical Engine

/* SLIDER */
.summary-scroll {
  height: 500px;
  overflow-y: auto;
  padding: 5px;
  border-left: 2px solid #2c3e50;
}

.summary-details {
  background: #f8f9fa;
  padding: 8px;
  margin-bottom: 8px;
  border-radius: 4px;
  border: 1px solid #eee;
  font-size: 0.75em;
}

.summary-details summary {
  font-weight: bold;
  cursor: pointer;
  color: #2c3e50;
  list-style: none;
}

.summary-details summary::-webkit-details-marker { display: none; }

.summary-details[open] { background: #fff; border-left: 4px solid #e67e22; }

.field-tag {
  color: #e67e22;
  font-weight: bold;
  font-family: monospace;
}
</style>

.left-70[
### .center[Integrated Solar Potential Map]

.footnote[**Integration:** Fusion of RS, LiDAR, and Municipal API datasets.]
]

.right-30[
### .center[Data Schema]

.summary-scroll[

<details open class="summary-details">
  <summary>🆔 Metadata</summary>
  - <span class="field-tag">Building_ID:</span> Municipal cadastre link.
<br>
  - <span class="field-tag">Global_ID:</span> Spanish national parcel code.
<br>
  - <span class="field-tag">Confidence:</span> RS reliability (0.0-1.0).
</details>

<details class="summary-details">
  <summary>📏 Detection</summary>
  - <span class="field-tag">PV_Area_Detected:</span> Current panels (`$m^2$`).
  <br>
  - <span class="field-tag">Effective_Potential:</span> Net gap (`$m^2$`).
  <br>
  - <span class="field-tag">Obstacle_Ratio:</span> LiDAR point cloud filter.
</details>

<details class="summary-details">
  <summary>📐 Attributes</summary>
  - <span class="field-tag">Average_Tilt:</span> Slope (`$0.0-90.0^\circ$`).
  <br>
  - <span class="field-tag">Azimuth:</span> Orientation (`$0-359^\circ$`).
  <br>
  - <span class="field-tag">Solar_Hours:</span> Ray-tracing (0-8,760h).
</details>

<details class="summary-details">
  <summary>⚖️ Assessmnet </summary>
  - <span class="field-tag">Solar_Gap_Score:</span> Ranking priority.
  <br>
  - <span class="field-tag">Social_Equity:</span> Income + Poverty data.
    <br>
  - <span class="field-tag"> CO2_Offset_Annual:</span> Estimated carbon reduction potential.
      <br>
  - <span class="field-tag">Economic_ROI:</span> Dynamic return on investment period.

</details>

<details class="summary-details">
  <summary>🧱  Context</summary>
  - <span class="field-tag">Roof_Material:</span> Spectral reflectance.
  <br>
  - <span class="field-tag">Heritage_Status:</span> Planning constraint.
  <br>
  - <span class="field-tag">Land_Use:</span> Million-AID classification.
</details>

<details class="summary-details">
  <summary>📈 Monitoring & Yield</summary>
  - <span class="field-tag">Self_Sufficiency:</span> IoT + RS fusion.
  <br>
  - <span class="field-tag">Current_Gen_kW:</span> Real-time output.
</details>

]
]

???
ICY:  
Here is the integrated final outcome

We recognizes existing assets and  
precisely calculates the Net solar Gap and   
most importantly point exactly where **High-Potential zones meet High-Priority needs**.

To make this operational system for policy, we move forward to Assessment. 
---
<style type="text/css">
/* local arrangemnt */
.formula-accordion { width: 100%; font-family: 'Arial', sans-serif; }

.formula-accordion details {
  margin-bottom: 12px;
  padding: 15px;
  border-radius: 10px;
  background: #ffffff;
  border: 1px solid #eee;
  transition: all 0.3s ease;
}

.formula-accordion summary {
  font-weight: bold;
  cursor: pointer;
  outline: none;
  font-size: 1.1em;
}

/* color bar */
.orange-panel { border-left: 8px solid #e67e22 !important; color: #e67e22; }
.green-panel  { border-left: 8px solid #27ae60 !important; color: #27ae60; }
.pink-panel   { border-left: 8px solid #e91e63 !important; color: #e91e63; }
.blue-panel   { border-left: 8px solid #2980b9 !important; color: #2980b9; }

/* bg */
.formula-accordion details[open] { background: #fcfcfc; box-shadow: 0 4px 10px rgba(0,0,0,0.05); }

/* overflow */
.formula-content {
  padding: 10px 0;
  color: #333; 
}
</style>

# Multiple Index Synthesis

.formula-accordion[

<summary>1. Precision: Net Solar Potential Gap (`$E_{gap}$`)</summary>
<div class="formula-content">
`$$E_{gap} = [ (A_{roof} \times f_{obs}) - A_{exist} ] \times \left( \sum_{h=1}^{8760} G_{h} \times \text{Shade}_h \right) \times \eta_{sys} \times PR$$`
</div>
</details>

<details class="green-panel"><summary>2. Climate: Carbon Gap Tracking (`$\Delta CO_2$`)</summary>

<div class="formula-content">
`$$\Delta CO_2(t) = \text{Target}_{2030} - [ \sum (\text{Stock}_{detected} + \text{Potential}_{Egap}) \times EF_{grid} ]$$`
</div>
</details>

<details class="pink-panel"><summary>3. Economy: Dynamic ROI (`$T_{payback}$`)</summary>

`$$T_{payback} = \frac{Cost_{invest}}{\sum_{y=1}^{n} [ (Yield \times SSR \times P_{buy}) + (Yield \times (1-SSR) \times P_{sell}) - Cost_{maint} ]}$$`
</div>
</details>

<details class="blue-panel"><summary>4. Equity: Social Equity Index (`$SEI$`)</summary>
<div class="formula-content">
`$$SEI = \text{Norm}(Score_{gap}) \times \left( 1 - \frac{\text{District\_Income}}{\text{City\_Avg\_Income}} \right)$$`
</div>
</details>]

.center[<img src="images/EEE.png" width="70%" style="border-radius:12px; border: 1px solid #ddd;">]

???
ICY:  
Here is the underlying Formula that powers this prioritization.

It shows the financial feasibility   , help tracking the carbon target,  Ensuring energy justice

---
# Data Feasibility Analysis
*Evaluating the strengths and limitations of our Earth Observation data sources.*

### 🤖 Step 1: PV Mapping

**Main data source: Airbus Pléiades Neo (30 cm)**

**Advantage:** 30 cm RGB imagery plus infrared and Near Infrared (NIR) bands for PV vs. non-PV discrimination.

**Disadvantage:** Higher acquisition cost.

**Supplementary data source: Cartographic and Geological Institute of Catalonia (ICGC)**

**Advantage:** 25 cm RGB plus infrared imagery to improve classification detail and robustness.

**Disadvantage:** Does not provide NIR, so generalisation may require careful transfer learning.

---
# Data Feasibility Analysis
*Evaluating the strengths and limitations of our Earth Observation data sources.*

### ⛰️ Step 2: Potential Layers

**Main data source: ICGC LiDAR (Terrain & Shadows)**

**Advantage:** 
1. Very high resolution surface elevation (DEM/DSM) and robust derivation of slope and aspect.
2. Supports shadow modelling and incidence-geometry correction for solar potential calculations.

**Disadvantage (Assumption):** LiDAR-derived features may be insufficient alone for complex rooftop cases, so ML + rule-based corrections are needed for final obstacle handling and PV mapping.

---
# Data Feasibility Analysis
*Evaluating the strengths and limitations of our Earth Observation data sources.*

### 🌍 Step 3: Context & Indicators

**Main data source: Sentinel-2**

**Advantage:** 10 m imagery and derived indices (RGB, NDVI, NDBI, NDWI) available since 2015.

**Disadvantage:** Limited spatial detail for rooftop-scale signatures; cloud availability issues.

**Supplementary data source: Sentinel-3**

**Advantage:** Complementary thermal/heat-map style context at low cost (useful for broader environmental interpretation).

**Disadvantage:** Coarser resolution; used for city-scale context rather than rooftop-level detection.

---
class: inverse, center, middle
# Project General Plan
---
# Budget (£500,000 total)

<img src="images/budget.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure 19: Sankey Diagram of Project Budget Allocation

???
Over half of our budget supports the data infrastructure and our core algorithms. **The rest goes into building the final platform and daily management.** To ensure we deliver the product, but still keep a safe buffer for any risks.

---
# Project Timeline

<img src="images/Timeline1.png" alt="" width="100%" style="display: block; margin: auto;" />
<p style="font-size: 0.6em; text-align: center; margin-top: 0;">
Figure20: Timeline
</p>

PM: Project Manager

GA: Geospatial Analyst

SDA: Senior Data Analyst

EC: Environmental Consultant

BCN: Barcelona City Council

IT/Dev: Web Developer

---
# Risk, Ethics and Limitations
.panelset[
.panel[.panel-name[Risks and Ethics]
.pull-left[
**Risks**
- Budget Constraints and Cost Exceedance
- Data Accuracy and Uncertainty
- Technical and Implementation Barriers
- Policy, Regulatory and Institutional Barriers
- Timeline Sensitivity
]
.pull-right[
**Ethics**
- Transparency and Accountability
- Data Usage, Consent and Privacy
- Social Equity Considerations
- Safety Standards and Professional Responsibility
]
]

.panel[.panel-name[Limitations]
- Identified potential does not reflect economic or market viability
  - Rooftops identified suitable for solar installation are based on size, shading, orientation, and location but does not reflect any financial considerations.
  - Costs for installation, grid connection, and prevailing market conditions remain outside the scope.

- Structural Stability and Load-Bearing Capacity
  - Structural integrity of existing building roofs is sufficient to withstand the additional weight and wind loads imposed by solar panels and their mounting structures were not assessed
  - A rooftop identified as suitable by the dashboard may still be rejected upon physical inspection. 
  - Actual installation would require integration with structural engineering assessments and consultation with qualified professionals.

<p style="font-size: 0.6em; text-align: left; margin-top: 0;">
References: <a href="https://www.energy.gov/eere/solar/solar-rooftop-potential"> Energy.gov. (n.d.)</a>
</p>

]
]
---
class: inverse, center, middle
# Thank you !