Electromagnetic Compatibility in Spacecraft Launches. Analysis and

Interpretation of the Phenomenon Using Statistical Correlations

Adriana Marcela Barrios Garc̀ıa

Degree work report to obtain the Title of Aerospace Engineer

Thesis Supervisor

Roy Stevenson Soler Chisabas, Specialist in Spacecraft Thermal Control and Space

Environment Simulation

Internal Advisor

Juan Francisco Puerta Ibarra, Msc in Space Systems Engineering

External Advisor

José Fernando Jiménez Vargas , HdR in Transdisciplinary Systems

Universidad de Antioquia

Engineering Faculty

Aerospace Engineering

Carmen de Vı̀boral, Antioquia, Colombia

2025



Cita Barrios Garcia, 2025 [1]

Referencia

Estilo IEEE (2020)

[1] Barrios Garćıa, A. M. “Electromagnetic Compatibility in Spa-

cecraft Launches. Analysis and Interpretation of the Phenome-

non Using Statistical Correlations ”, Bachelor’s degree project,

Aerospace Engineering, Universidad de Antioquia, Carmen de Vi-

boral, Antioquia, Colombia, 2025.

Internship University: Universidad de los Andes.

Campus El Carmen de Viboral

Repositorio Institucional: http://bibliotecadigital.udea.edu.co

Universidad de Antioquia - www.udea.edu.co

Chancellor: John Jairo Arboleda Céspedes.

Dean: Julio César Saldarriaga Molin.

Chair: Pedro León Simanca.

El contenido de esta obra corresponde al derecho de expresión de los autores y no compromete

el pensamiento institucional de la Universidad de Antioquia ni desata su responsabilidad

frente a terceros. Los autores asumen la responsabilidad por los derechos de autor y conexos.



Dedication

For my parents Felipe and Carmen,

my sister Laura,

my grandparents Aura and Carlos,

my family, my sisters by heart Paula and my angel Valentina,

and my friends,

For always supporting me throughout my studies and encouraging me to pursue my dreams.



Acknowledgments

I would like to express my deepest gratitude to Mr. Roy Stevenson Soler for his invaluable

support and guidance throughout the development of this thesis. His patience and

willingness to teach in the gaps I had, in such a complex subject, were essential for this

work to come to life. Far from growing impatient, he accompanied me with generosity and

clarity through every stage. Thanks to his guidance and expertise, this thesis is now a

reality. From Roy, I not only learned a lot but also found someone I truly admire, not only

is he incredibly skilled, but he is also a professional who lives his vocation with passion and

always works from a place of deep love for what he does.

I would also like to thank Professor Juan Francisco Puerta and José Fernando Jiménez,

who became the bridge that led me to complete my internship at Universidad de los Andes.

In that space, I not only learned deeply about the subject, but I also had the fortune of

receiving academic and human support that was invaluable in helping me define the focus

of my thesis.

To my family, for always being there, looking after my well-being, caring for me, and

encouraging me to do work that was not only rigorous but also fulfilling. Their love and

support were the pillars that sustained every step of this journey.

To Nidia, a lifelong friend, who has supported me unconditionally since i was little. She has

always looked out for me and believed in me, even during the times I doubted myself. Her

confidence and belief were essential in helping me trust my own abilities and believe that

one day I would become a professional.

To my friends, who throughout my entire university journey offered me their support

during the toughest times. From them, I learned not only academic knowledge but also

important life lessons. They protected me in moments of vulnerability and believed in me

far more than I believed in myself. Each of you played an essential role in this achievement.

From the bottom of my heart, I thank you for being there and for holding my hand when I

needed it most. This degree is not just mine... it is also yours. I hope you feel proud of what

we have accomplished together.

4



TABLE OF CONTENTS

RESUMEN 12

ABSTRACT 13

I INTRODUCTION 14

II PROBLEM STATEMENT 17

IIIJUSTIFICATION 19

IVOBJECTIVES 20

A General Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

B Specific Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

V SCOPE 21

VI LITERATURE REVIEW 22

A Electromagnetic Compatibility (EMC) . . . . . . . . . . . . . . . . . . . . . 22

1 Types of EMC Phenomena: . . . . . . . . . . . . . . . . . . . . . . . 23

2 Sources of EMI: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

B Electromagnetic Compatibility in Space Launches . . . . . . . . . . . . . . . 25

1 Involved actors: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2 EMC interactions between launch actors: . . . . . . . . . . . . . . . . 25

3 EMC in the mission design process: . . . . . . . . . . . . . . . . . . . 27

C Applicable EMC Standards and Technical Regulations . . . . . . . . . . . . 28

1 Regulatory Standard: . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2 Technical Standards: . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3 Tailoring: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4 Handbook: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

5 Payload User’s Guide (PUG): . . . . . . . . . . . . . . . . . . . . . . 31



VIIMETHODOLOGY 32

A Part I: Statistical Analysis of the Electromagnetic Environment . . . . . . . 33

1 Structure and parameters of the database . . . . . . . . . . . . . . . 34

2 Analysis of the electromagnetic environment by actor type . . . . . . 36

3 Launch Vehicle maximum radiated emissions under worst-case scenario 41

4 Spacecraft maximum radiated emissions . . . . . . . . . . . . . . . . 46

5 Comparison of launch site emissions with launch vehicle and spacecraft

limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

a Launch site radiated emissions relative to spacecraft limits . 47

b Launch site radiated emissions relative to launch vehicle emis-

sions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

6 Environmental factors influence on launches . . . . . . . . . . . . . . 50

7 Pattern analysis and justification for emission data segmentation . . . 55

8 Explanation of the analysis sequence and methodology flow . . . . . . 56

9 Clustering Method and determination of the optimal number of clusters 57

10 Cluster analysis process . . . . . . . . . . . . . . . . . . . . . . . . . 58

11 Comparison with the general model . . . . . . . . . . . . . . . . . . . 59

B Part II: Analysis of Electromagnetic Compatibility (EMC) Requirements . . 60

1 EMC normative review context in the PUGs . . . . . . . . . . . . . . 60

2 Comparative Review of EMC-Related Requirements Across PUGs . . 61

VIIIRESULTS 63

A General Analysis of the Electromagnetic Environment . . . . . . . . . . . . . 63

1 Launch vehicle maximum radiated emissions . . . . . . . . . . . . . . 63

2 Spacecraft Maximum Radiated Emissions . . . . . . . . . . . . . . . . 67

3 Launch site radiated emissions relative to spacecraft limits . . . . . . 71

4 Launch site radiated emissions relative to launch vehicle emissions . . 72

5 Influence of environmental conditions on launch activity . . . . . . . 73

B Cluster-Based Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . 82



1 Cluster 0 – Light and slim launch vehicles with low electromagnetic

emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

2 Cluster 1 – Heavy launch vehicle with highest electromagnetic emissions 96

3 Cluster 2 - Robust launch vehicles with moderate electromagnetic emis-

sions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

C Analysis of Requirements and Referenced Standards in the PUGs . . . . . . 107

1 Comparative Requirement Analysis across PUGs . . . . . . . . . . . 107

2 Referenced, Suggested, or Absent Standards in the PUGs . . . . . . . 111

IXCONCLUSIONS 115

X RECOMMENDATIONS AND FUTURE WORK 117

REFERENCES 119

APPENDIX 123

APPENDIX A. Abstract accepted for presentation at the 76th IAC . . . . . . . . 123



LIST OF TABLES

Table I Summary of Launch Modules and Key Information . . . . . . . . . . . 34

Table II Environmental Conditions and Launch Activity of the Main Launch

Sites (2018–2025). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Table III Rocket Technical Parameters to Analyze LV Emissions . . . . . . . . . 64

Table IV General Analysis of LV Emissions for E-Field Range . . . . . . . . . . 65

Table V General Analysis of LV Emissions for Frequency Range . . . . . . . . . 66

Table VI Rocket Technical Parameters to Analyze S/C Emissions Limits . . . . 68

Table VII Analysis S/C Radiated Emissions Limit - Frequency Range . . . . . . 69

Table VIII Analysis S/C Radiated Emissions Limit - Frequency Range . . . . . . 69

Table IX Analysis S/C Radiated Emissions Limit – E-Field Range . . . . . . . . 70

Table X Environmental Conditions and Launch Activity of the Main Launch

Sites (2018–2025). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Table XI Correlations Between Environmental Variables and Launch Frequency

(2018–2025) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Table XII Average Environmental Conditions and Number of Launches by Köppen

Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Table XIII Average Characteristics by Cluster . . . . . . . . . . . . . . . . . . . . 82

Table XIV Statistical Summary of Rocket Parameters . . . . . . . . . . . . . . . . 84

Table XV Statistical Summary of S/C Radiated Emissions Limit by Cluster . . . 84

Table XVI LV Characteristics Summary for Launch Vehicle Emissions - Cluster 0 85

Table XVIILV Emission Analysis for Frequency Range - Cluter 0 . . . . . . . . . 87

Table XVIIILV Emission Analysis for E-Field Range – Cluster 0 . . . . . . . . . . 88

Table XIX LV Characteristics Summary for S/C Emissions Limit - Cluster 0 . . . 90

Table XX S/C Emissions Limit Analysis in Frequency Range – Cluster 0 . . . . 92

Table XXI S/C Emissions Limit Analysis for E-Field Range – Cluster 0 . . . . . . 93

Table XXIIAriane 6 – Technical and EMC Parameters Summary . . . . . . . . . 96

Table XXIIILV Characteristics Summary for Launch Vehicle Emissions - Cluster 2 98



Table XXIVLV Emission Analysis for Frequency Range – Cluster 2 . . . . . . . . . 99

Table XXVLV Emission Analysis for E-Field Range – Cluster 2 . . . . . . . . . . 100

Table XXVILV Characteristics Summary for S/C Emissions Limits - Cluster 2 . . 102

Table XXVIIS/C Emissions Limit Analysis for Frequency Range – Cluster 2 . . . . 104

Table XXVIIIS/C Emission Limit Analysis for E-Field Range – Cluster 2 . . . . . . 106

Table XXIXEMC Requirement Category Coverage per Launch Vehicle . . . . . . . 109

Table XXXSummary of Standards for Launches . . . . . . . . . . . . . . . . . . . 112

Table XXXISummary of Standards for Launches . . . . . . . . . . . . . . . . . . . 113



LIST OF FIGURES

Fig. 1 Elements of an EMC problem. Source: [1] . . . . . . . . . . . . . . . . . 23

Fig. 2 EMC phenomena. Source: [2] . . . . . . . . . . . . . . . . . . . . . . . . 24

Fig. 3 EMC actors in a space launch. Source: AI Generated . . . . . . . . . . . 26

Fig. 4 EMC process vs. mission design process. Source: [3] . . . . . . . . . . . 27

Fig. 5 Example of correlation methodology . . . . . . . . . . . . . . . . . . . . 38

Fig. 6 Launch vehicles maximum radiated emissions under worst-case scenario 63

Fig. 7 General analysis of LV emissions . . . . . . . . . . . . . . . . . . . . . . 65

Fig. 8 Spacecrafts maximum radiated emissions . . . . . . . . . . . . . . . . . 67

Fig. 9 General analysis of S/C emissions limit . . . . . . . . . . . . . . . . . . 69

Fig. 10 Launch site emissions relative to spacecraft limits . . . . . . . . . . . . . 71

Fig. 11 Launch site emissions relative to LV emissions . . . . . . . . . . . . . . 73

Fig. 12 Location of launch sites on a world map . . . . . . . . . . . . . . . . . . 75

Fig. 13 Launch frequency by köppen category . . . . . . . . . . . . . . . . . . . 79

Fig. 14 Annual launch trend by köppen category . . . . . . . . . . . . . . . . . 80

Fig. 15 Annual average environmental trends at launch sites (2018–2025) . . . . 81

Fig. 16 Average characteristics by cluster . . . . . . . . . . . . . . . . . . . . . 83

Fig. 17 Rocket emission analysis for frequency range – cluster 0 . . . . . . . . . 86

Fig. 18 Rocket emission analysis for E-Field range – cluster 0 . . . . . . . . . . 89

Fig. 19 Spacecraft emission limits analysis for frequency range – cluster 0 . . . . 91

Fig. 20 Spacecraft emissions limits analysis for E-Field range – cluster 0 . . . . 94

Fig. 21 Rocket emission analysis for frequency range – cluster 2 . . . . . . . . . 99

Fig. 22 Rocket emission analysis for E-Field range – cluster 2 . . . . . . . . . . 101

Fig. 23 Spacecraft limits emission analysis for E-Field range – cluster 0 . . . . . 103

Fig. 24 Rocket emission analysis for E-Field range – cluster 2 . . . . . . . . . . 105

Fig. 25 Abstract accepted for IAC 2025 – Interactive Presentation . . . . . . . . 123



ACRONYMS AND ABBREVIATIONS

BC Baikonur Cosmodrome

CASC China Aerospace Science and Technology Corporation

CCSFS Cape Canaveral Space Force Station

CNSA China National Space Administration

CSG Guiana Space Centre

dBµV/m Decibel per microvolt per meter

IEEE Institute of Electrical and Electronics Engineers

ISS International Space Station

KSC Kennedy Space Center

LC Launch Complex

MHz Megahertz

MHI Mitsubishi Heavy Industries

PC Plesetsk Cosmodrome

PSC Pacific Spaceport Complex

RKTs Progress Progress Rocket Space Centre

TNSC Tanegashima Space Center

UdeA Universidad de Antioquia

VSFB Vandenberg Space Force Base

XSLC Xichang Satellite Launch Center



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino12

RESUMEN

Esta tesis realiza un análisis exploratorio del entorno electromagnético que enfrentan las

cargas ùtiles en lanzamientos comerciales a órbita baja terrestre (LEO), considerando tan-

to los niveles de emisión radiada como los requisitos de compatibilidad electromagnética

(EMC) descritos en los Payload User’s Guides (PUGs) de distintos lanzadores. Se aplica-

ron correlaciones y segmentaciones por clúster para identificar qué variables estructurales

y operacionales influyen en las emisiones y restricciones. A nivel general, se observó que el

tipo de propulsión (ρ = –0,81) y la relación L/D del cohete (ρ = –0,46) se correlacionan

con las emisiones generadas por el lanzador, mientras que el tipo de misión (ρ = –0,55) y

la capacidad de carga útil (ρ = –0,33) influyen sobre las restricciones impuestas a la carga

útil. En clústeres espećıficos, otras variables como la masa (ρ = 0,81), el número de etapas

(ρ = –0,81) y la relacion L/D del carenado (ρ = 0,94) también mostraron asociación con

el rango de frecuencia o campo eléctrico permitido. Respecto al entorno del sitio de lanza-

miento, se identificó que LC-36 tiende a generar emisiones más altas relativas al vehiculo de

lanzamiento, y que en algunos casos estas superan también las restricciones impuestas a la

carga útil. Aunque no se encontró una relación cuantitativa entre condiciones ambientales y

frecuencia de lanzamientos, se evidenció una alta concentración en climas tipo CFA y CSB, lo

que responde más a decisiones loǵısticas e infraestructura disponible. En cuanto a los requisi-

tos EMC, se identificaron 52 en total, distribuidos en 11 categoŕıas. La mayoŕıa corresponde

a lineamientos generales o sugerencias sin criterios de prueba definidos, y con baja presencia

de referencias normativas, lo que pone de manifiesto la necesidad de mayor estandarización,

especialmente para validar sistemas integrados.

Palabras clave — Compatibilidad Electromagnética (EMC), CubeSats,

Emisiones del Veh́ıculo Lanzador, Requisitos de la Carga Útil, Estandarización



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino13

ABSTRACT

This thesis presents an exploratory analysis of the electromagnetic environment faced by spa-

cecrafts during commercial launches to Low Earth Orbit (LEO), considering both radiated

emission levels and electromagnetic compatibility (EMC) requirements as described in va-

rious launch providers’ Payload User’s Guides (PUGs). Correlation analyses and cluster-based

segmentation were applied to identify which structural and operational variables influence

emissions and constraints. Globally, the type of propulsion system (ρ = –0,81) and rocket

L/D dimensions ratio (ρ = –0,46) were found to correlate with launch vehicle emissions, whi-

le payload constraints were mainly influenced by the mission type (ρ = –0,55) and payload

capacity (ρ = –0,33). In specific clusters, other variables such as vehicle mass (ρ = 0,81),

number of stages (ρ = –0,81), and fairing L/D ratio (ρ = 0,94) also showed associations with

frequency range or electric field intensity. Regarding launch sites, it was found that LC-36

tends to generate higher emissions relative the launch vehicle itself and in some cases, these

emissions also exceed the maximum allowed for the spacecrafts. Although no quantitative

relationship was found between environmental conditions and launch frequency, most launch

sites are located in Köppen CFA and CSB climates, due more to logistical and infrastruc-

ture factors than environmental suitability. In terms of EMC requirements, 52 entries were

identified across 11 categories. Most are general guidelines or non-binding recommendations,

with limited normative references and test criteria, highlighting the urgent need for clearer

standardization—particularly for the validation of fully integrated systems.

Keywords — Electromagnetic Compatibility (EMC), Launch Vehicle Emis-

sions, Radiated Emissions, Payload Requirements, Standardization



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino14

I. INTRODUCTION

An electromagnetic emission is any form of energy in electromagnetic waves that

is generated and released by an electrical or electronic system during its operation. These

emissions can propagate through space (radiated) or via cables and conductive structures

(conducted), and may interfere with the operation of nearby systems if not adequately con-

trolled [4]. The greater the system’s complexity, the harder it becomes to mitigate such

interference, increasing the need to ensure its compatibility. One of the sectors where this

becomes especially critical is space launches, due to the high number of electronic systems

operating simultaneously. In this context, it is first necessary to ensure that each system

does not generate electromagnetic interference (EMI) towards itself [4], and then ensure its

compatibility with larger systems, such as the launch vehicle (rocket), the launch site, and

the onboard payload, all of which generate their own emissions. Likewise, it is necessary to

avoid interference between these systems, particularly regarding radiofrequency characteris-

tics, and ensure that each one can successfully fulfill the function for which it was designed

[5].

With the increase in LEO orbit launches, driven by the rideshare modality—which

allows multiple satellites to be sent in a single mission, reducing costs and minimizing envi-

ronmental impact—the launch frequency has grown significantly [6], [7], [8]. This trend has

promoted commercialization in the sector, with multiple companies offering launch services

through planning documents called Payload User’s Guides (PUG), designed to help various

stakeholders understand the standard services prior to contracting [9] and access minimum

requirements and recommendations. However, despite this practice being several years old,

deficiencies remain in these documents, which do not always contain the full information a

potential client would need to make appropriate decisions. Often, full mission details must be

disclosed or confidentiality clauses signed to access critical information, posing a significant

barrier.

The section related to electromagnetic compatibility (EMC) is, by itself, one of the

most complex, as it involves considering multiple factors simultaneously. Electromagnetic



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino15

emissions are not constant throughout a mission: they vary during launch, transition to

the orbital environment, and in the final environment where the payload will operate [10],

[11], [12]. To ensure a viable launch from the EMC perspective, it is necessary to guarantee

that the radiated emissions generated by the payload do not exceed the limits imposed by

the launch vehicle, that it can withstand the maximum emissions generated by the rocket

and launch site in the worst-case scenario, that there is no interference in radiofrequency

systems, and that the internal EMI of the payload itself is controlled before considering the

complete environment [5]. Additionally, the environment and conditions present in integration

laboratories, transport, the launch site, and the orbital environment directly influence final

compatibility [13].

However, there is little documentation explaining how radiated emissions behave du-

ring a launch, what factors influence them, or how they relate to the technical, structural,

regulatory, or environmental characteristics –of the various actors involved (rockets, launch si-

tes, and payloads). This is compounded by the significant shortcomings found in many PUGs,

which—despite being the first technical point of contact between provider and client—do not

always serve as comprehensive tools for any stakeholder, whether developer, operator, or tes-

ting infrastructure, aiming to guide the design and planning of a mission from the EMC or

other relevant perspectives.

Given the lack of a clear tool—academic, technical, or commercial—to anticipate which

launch vehicle impose stricter or more permissive conditions in terms of electromagnetic com-

patibility, the aim of this work is to characterize the electromagnetic environment present

in commercial launches to LEO orbit from the payload’s perspective, using the PUGs and

through an exploratory statistical analysis of the reported radiated emissions. The analysis

includes the maximum emissions generated by the rocket, the launch site, and the emission

limits imposed on the payloads, with the goal of identifying common behaviors and deriving

technical criteria. This analysis involves the use of correlations to identify relationships bet-

ween variables associated with each type of emission, as well as cluster-based segmentation

algorithms to group launchers according to similarities in their electromagnetic profile. This

will allow for the development of a tool or guide to help any stakeholder more efficiently



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino16

select a launch vehicle compatible from the EMC perspective.

Unlike other approaches that begin with experimental validation in anechoic chambers

[4], this thesis is developed exclusively through the analysis of data available in PUGs, without

physical intervention on the payloads or their testing environment. Likewise, it proposes to

normatively contextualize which standards are recommended or required for verification and

validation processes, and to understand how relationships are structured among the various

actors involved in the launch process. This work not only identifies differences between launch

vehicles and their environments, but also offers an initial technical analysis framework useful

for strategically planning design, integration, and test preparation. It represents one of the

first systematic attempts to democratize EMC preparation in emerging space missions, and

can be further strengthened over time as more launch service providerss include relevant data

in their PUGs. Even if all EMC-related information is not detailed, knowledge of the launch

vehicle’s technical characteristics could allow for the inference of its expected electromagnetic

behavior and open the door to the development of specialized tools or supporting software.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino17

II. PROBLEM STATEMENT

During spacecraft launches to LEO orbit, it is essential to ensure electromagnetic

compatibility (EMC) between the rocket, the launch site, and the payload. This involves

guaranteeing that radiated emissions do not generate interference, that the limits imposed

by the provider are respected, and that the environment does not negatively affect system

performance. All these requirements are usually stated in the Payload User’s Guides (PUGs)

of each launch service provider.

Standards such as MIL-STD-461G, ECSS-E-ST-20-07C, or NASA-HDBK-4002A [10]

[11] [12] address verification and validation processes from the perspective of payload design

or development. However, they are not specifically focused on the launch stage, and not all

of them are applicable or accessible to small-scale programs. These standards tend to focus

on general testing procedures without addressing how radiated emissions behave under real

launch conditions, which technical or environmental factors affect them, or how operational

limits are established.

Although some research has been conducted on EMC in this context — such as Erik

Lidman’s master’s project at Lule̊a University of Technology [4] — it focuses on the execu-

tion of tests adapted to laboratory environments, based on the aforementioned technical

standards. While such studies are useful from the perspective of testing infrastructure, they

do not address the problem from the payload’s point of view and do not use PUGs as a

source of analysis.

With the growing availability of commercial launch services and services, there is

currently no tool that enables comparison of their electromagnetic requirements or prediction

of how demanding they may be in this regard. There is also no clear information on how

the technical characteristics of the launch vehicle or the environment relate to the behavior

of radiated emissions, making it difficult to anticipate restrictions, make informed decisions,

or adapt spacecraft design in early development stages. Even when technical data from the

launch vehicle is available, it remains unclear how such data translates into concrete EMC

constraints.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino18

In response to this methodological and documentary gap, this research proposes to

characterize the electromagnetic environment in commercial launches through an explora-

tory statistical analysis of the data reported in PUGs. The objective is to derive technical

criteria that enable the classification and comparison of launch vehicles from the EMC pers-

pective, thereby facilitating early-stage decision-making in mission development. This study

represents an initial approach and is limited to an exploratory scope; it does not involve

experimental validation.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino19

III. JUSTIFICATION

With the increase in orbital launches and the reduction of their costs, selecting a

launch service provider in a technical and efficient manner has become essential. To achieve

this, it is crucial to understand the electromagnetic behavior during launches and the factors

that influence it, especially based on the information available in the Payload User’s Guides

(PUG). While existing standards offer recommendations for verifying electromagnetic com-

patibility (EMC) throughout a spacecraft’s lifecycle, these guidelines tend to be general and

focused on testing execution, not on how electromagnetic emissions behave during a launch.

This research proposes an exploratory statistical analysis of the data available in

PUGs, with the aim of deriving technical criteria to classify and compare launch vehicles from

the EMC perspective. Through correlations and cluster-based segmentation, this study seeks

to build a preliminary guide to support informed decision-making during the early planning

stages of missions, minimizing risks such as launch provider changes or cost overruns due to

compatibility issues.

It also emphasizes a key practical contribution: while large companies may have inter-

nal tools or even direct agreements with launch providers to access more detailed technical

information, small-scale programs—such as those developed by universities, research centers,

or startups—depend heavily on publicly available data. This proposal addresses that gap by

offering a technically grounded alternative for interpreting and comparing EMC constraints

from the payload perspective.

Initially, the project considered adapting an anechoic chamber to perform experimental

validation. However, infrastructure, time, and documentation limitations led to a shift in

focus. This work is presented as a first systematic step that lays the foundation for more

complex future studies and enables progress—despite limited information—towards a more

accessible, scalable, and evidence-based EMC preparation approach.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino20

IV. OBJECTIVES

A. General Objective

To characterize the electromagnetic environment present in commercial spacecraft

launches to Low Earth Orbit (LEO) through an exploratory statistical analysis of radiated

emissions generated by the launch vehicle, the launch site, and the emission limits imposed

on payloads, in order to derive technical criteria to guide launch provider selection from the

perspective of electromagnetic compatibility, without including experimental validation.

B. Specific Objectives

• To collect and systematize parameters related to electromagnetic emissions in the launch

environment based on documentary sources such as the Payload User’s Guides (PUG).

• To statistically analyze the electromagnetic emissions generated by the launch vehicle and

the launch site, as well as the permissible limits for spacecrafts, identifying patterns and

applying clustering segmentation techniques.

• To compile and compare electromagnetic compatibility requirements established for spa-

cecrafts in the Payload User’s Guides of different rockets, relating them to applicable

technical standards and to the criteria defined by each provider for their interpretation

and compliance.

• To establish a set of technical criteria derived from statistical analysis and data segmen-

tation, to serve as a reference tool for guiding the selection of commercial launch services

from the perspective of electromagnetic compatibility.

• To generate technical guidelines applicable to the preliminary design of electromagnetic

compatibility verification processes, particularly useful in environments lacking advanced

testing infrastructure, thereby contributing to the development of future validation stra-

tegies.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino21

V. SCOPE

This research focuses on characterizing the electromagnetic environment associated

with commercial spacecraft launches to LEO orbit, from the perspective of the payload. The

study is limited to an exploratory and statistical analysis of radiated emissions reported in

publicly available Payload User’s Guides (PUGs).

Three main sources of emission are analyzed: the launch vehicle, the launch site, and

the emission limits imposed on payloads. Additionally, the influence of the environmental

conditions and the geographic location of the launch site is considered. To this end, data

on temperature, humidity, pressure, precipitation, and Köppen climate classification were

collected from platforms such as Time and Date and WeatherSpark [14] [15]. Monthly launch

activity statistics for each site were also integrated from sources such as Gunter’s Space Page,

Launch Report, and Our World in Data [16] [17] [18].

Based on these data, common patterns are identified and relationships between varia-

bles are established through correlation analysis and cluster-based segmentation. The charac-

terization is conducted exclusively in terms of the frequency range in which emissions occur

and the amplitude of the radiated electric field intensity.

The study does not include experimental validation, physical modeling of the envi-

ronment, or direct verification of susceptibility or immunity tests. It also excludes missions

beyond LEO orbit and launch providers that do not provide publicly accessible information

in their PUGs.

The scope of this work is restricted to the development of a preliminary tool to classify

launch vehicles according to their electromagnetic and environmental conditions, serving as

a basis to support early-stage decision-making in spacecraft design, integration, and test

planning.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino22

VI. LITERATURE REVIEW

A. Electromagnetic Compatibility (EMC)

It is the capacity of a system, device, or equipment to operate correctly within an elec-

tromagnetic environment, without causing interference to other devices and without being

affected by surrounding radiated emissions. In other words, an electromagnetically compa-

tible system neither causes nor suffers from interference [1]. EMC problems usually arise

when emissions (EMI) exceed the system’s immunity capacity (EMS). This relationship is

commonly represented by equation (1):

EMC = EMI + EMS (1)

Where:

• Electromagnetic Interference (EMI) is the unwanted electromagnetic energy emitted by a

device, which can alter or degrade the performance of nearby systems. These emissions can

cause failures, malfunction, temporary or permanent errors in the affected system.

• Electromagnetic Susceptibility (EMS) is the ability of a system or device to continue

operating properly in the presence of an electromagnetic field. In other words, it measures

how immune a device is to external disturbances that could affect its performance.

Emissions and susceptibility can be classified according to the type of coupling such

as:

• Radiated: transmitted through space as electromagnetic waves.

• Conducted: transmitted through cables or conductive structures.

For an electromagnetic compatibility problem to occur, three fundamental elements

must coincide, as shown schematically in Figure 1. If any of these three elements is eliminated

or mitigated, the EMC issue is reduced or disappears.

• Interference source: device, system, or phenomenon that generates the emission.

• Susceptible circuit: device or system affected by the disturbance.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino23

Fig. 1. Elements of an EMC problem. Source: [1]

• Coupling path: the route through which the disturbance reaches the victim (it may be

conduction, electric field, or magnetic field). This coupling can occur through different

mechanisms:

1. Capacitive coupling: via electric fields.

2. Inductive coupling: via magnetic fields.

3. Electromagnetic coupling: via radiation in space.

4. Common impedance coupling: when the source and the victim share the same con-

ductive path or electrical return.

1) Types of EMC Phenomena: The disturbances generated are grouped into four main

types, as shown in Figure 2:

• Conducted Emission (CE): conducted emissions generated by a device.

• Radiated Emission (RE): radiated emissions emitted from the device into its environment.

• Conducted Susceptibility (CS): device sensitivity to conducted energy.

• Radiated Susceptibility (RS): device sensitivity to radiated fields.

When a system can operate without suffering interference caused by any of the four

electromagnetic disturbances (CE, RE, CS, RS), and also does not interfere with itself, it

is considered to have electromagnetic compatibility (EMC). Solving an EMC problem can

be either simple or complex, depending on the system’s design, its level of integration, and

whether it acts as the source or the victim of the disturbance. If the system is the source of



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino24

Fig. 2. EMC phenomena. Source: [2]

the problem, a common strategy is to decouple the components that are mutually affecting

each other or to increase the physical separation between them to allow greater attenuation

of the electric and magnetic fields generated during operation.

2) Sources of EMI: There are various sources of electromagnetic interference, which can

be found all around us at any time. These sources can be:

• Natural: lightning, solar storms, volcanic eruptions.

• Intentional artificial: devices specifically designed to emit electromagnetic waves, such as

antennas, phones, routers, radars.

• Artificial non intentional: computers, motors, switches, fluorescent lights, chargers, which

emit EMI as a result of their operation.

Regardless of the nature of the interference source, it can degrade the performance of

a sensitive system or even compromise critical systems.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino25

B. Electromagnetic Compatibility in Space Launches

1) Involved actors: A typical space launch involves four main actors that must be consi-

dered from the mission design stage to avoid electromagnetic interference between them:

1. Launch vehicle (rocket): this is the system responsible for transporting the payload to

the desired orbit. The payload may include satellites, humans, or other spacecraft such

as probes, rovers, or observatories; everything being transported is referred to as the

payload [19].

2. Payload: the device intended to operate. In the case of a rocket, it depends on its

mission—it can be a satellite, humans, or even fireworks [20].

3. Launch site: this includes all the ground infrastructure necessary to prepare and execu-

te a space launch. It encompasses facilities for vehicle assembly, payload integration,

verification and control systems, communication infrastructure, and safety protocols.

For optimal performance, launch sites are usually located in geographically strategic

areas—such as near the equator—to take advantage of Earth’s rotation and maximize

launch energy efficiency [21].

4. The external environment: this includes natural conditions such as atmosphere, tem-

perature, humidity, or electrical phenomena present at the launch site. This will be

discussed in more detail in a later section.

Each of these actors can be a source or victim of electromagnetic emissions, so their

interaction must be carefully evaluated to ensure electromagnetic compatibility (EMC).

2) EMC interactions between launch actors: During the launch phase, as shown in Figure

3, the typical electromagnetic compatibility (EMC) relationships between key actors in a

space mission are illustrated. In this scenario, four main actors operate simultaneously: the

rocket, the launch site, the satellite (payload), and the environment. For a launch vehicle to

successfully complete its objective of transporting a payload to its destination, it is essential

that all these elements comply with EMC requirements.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino26

Fig. 3. EMC actors in a space launch. Source: AI Generated

The figure presents a simplified representation of these actors and their interactions.

Each can both generate and receive electromagnetic emissions, which may result in inter-

ference if not properly managed. For example, the rocket may generate radiated emissions

(RE) that interfere with sensitive systems aboard the spacecraft during ascent, or generate

conducted emissions (CE) toward the launch site during the pre-launch phase, when both

systems are connected via electrical umbilicals for telemetry, power, or control. Similarly,

the environment (atmosphere, humidity, pressure, etc.) may influence the electromagnetic

behavior of other components, altering the attenuation or propagation of signals.

The bidirectional arrows connecting the various actors represent the possibility of elec-

tromagnetic coupling in both directions, either through radiated links or physical connections.

This type of coupling implies that all actors must operate in harmony: they must not emit

signals that interfere with others, nor be vulnerable to external signals. In other words, each

system must perform its function without degrading the performance of the others.

Although additional elements such as ground stations are involved in reality, this

figure allows for a large-scale understanding of how EMC relationships manifest during a



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino27

space mission.

Examples of relationships that arise:

• Rocket ↔ Spacecraft: the rocket may generate emissions that affect the spacecraft’s

electronic systems, while this one, if emitting at unintended times, could interfere with the

launch vehicle’s sensors or communication systems.

• Rocket ↔ Launchsite: during ground operations, conducted emissions may occur through

umbilicals, as well as radiated emissions that interfere with control infrastructure.

• Spacecraft ↔ Environment: the environment may amplify, reflect, or attenuate the spa-

cecraft emissions, unintentionally altering its performance.

• Launchsite ↔ Rocket/Spacecraft: command and communication systems at the launch

site could interfere with the spacecraft receivers if not properly filtered or shielded.

3) EMC in the mission design process: As space missions become more frequent, smaller,

and subjected to tighter integration constraints, the likelihood of interference increases due

to the close proximity of components. Solving an EMC problem once the system is fully inte-

grated may be difficult, costly, or even impossible. Therefore, electromagnetic compatibility

must be addressed from the conceptual design phase of the mission. The process should follow

an iterative approach, as represented in Figure 4, which includes stages from conception to

production, considering EMC at each step.

Fig. 4. EMC process vs. mission design process. Source: [3]

Meeting EMC requirements not only ensures that the spacecraft functions properly

but also protects the rocket and ground systems, ensuring mission safety and success.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino28

C. Applicable EMC Standards and Technical Regulations

In the development and validation of space missions, electromagnetic compatibility

(EMC) compliance is not arbitrary. Depending on the role within the project—whether as

launch operator, payload manufacturer, or testing authority—reference documents are used

to guide design and validation according to the corresponding phase. These documents can

be classified into:

1) Regulatory Standard: is a document established by a competent authority that defines

criteria, requirements, and minimum guidelines to ensure that products, processes, or services

perform specific functions in a safe, reliable, and compatible manner with other systems. Its

purpose is to standardize practices, facilitate interoperability, improve quality, and reduce

technical or safety risks.

Standards are characterized by:

1. Being mandatory when required by law, contract, or national/international regulation.

2. Being issued by recognized standardization bodies, such as:

• ISO (International Organization for Standardization),

• IEC (International Electrotechnical Commission),

• ASTM, IEEE, among others.

3. Having a formal structure, which includes:

• Scope and field of application,

• Normative references,

• Terms and definitions,

• Specific technical requirements,

• Verification or testing methods.

A standard must be clear, consistent, and verifiable. It allows for auditing or certifi-

cation. However, in the space domain, the term ”standard̈ıs often used to refer to technical

documents that acquire normative value when required by contract or included in documents



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino29

such as Payload User’s Guides. In these cases, their application becomes mandatory even if

they are not legal standards in the strict sense. For example:

• MIL-STD-461G is a military standard that becomes mandatory if required by contract.

• ECSS-E-ST-20-07C is a European standard, but ESA may mandate its application as a

condition to participate in a joint mission.

2) Technical Standards: Is a formal document approved by a competent organization (such

as a space agency or a standardization body) that establishes technical criteria and procedures

to ensure the quality, safety, compatibility, and interoperability of systems, products, or

processes. A standard should contain:

1. Clear definitions of technical parameters,

2. Test and verification methods,

3. Quantitative limits,

4. And, in many cases, the technical rationale behind each requirement (annexes or ap-

pendices).

Standards provide common references for engineers, evaluators, and contractors. In the

space sector, they are fundamental for the design, testing, and integration of critical systems.

Although not always mandatory, they can become enforceable when incorporated into a

contract or technical specification. Additionally, they may allow some degree of adaptation or

customization (tailoring), provided there is technical justification and proper documentation.

In highly demanding contexts such as aerospace, they are key tools for ensuring reliability

and compatibility from the early stages of development.

3) Tailoring: in space engineering, the term tailoring refers to the process of adapting,

selecting, or modifying a standard or technical specification to fit the specific needs of a

mission, project, or operational context.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino30

Unlike applying a standard rigidly and in full, tailoring allows requirements to be

adjusted based on payload type, budget constraints, available infrastructure, or the project’s

technical goals. This is especially common in small-scale projects, such as CubeSats, where

not all requirements in a standard can or should be applied without adaptation. The tailoring

process may involve:

1. Omitting irrelevant requirements.

2. Simplifying procedures that would be too costly or complex.

3. Incorporating internal criteria or combining them with organizational practices.

4. Justifying and documenting which parts of the standard are followed, which are not,

and why.

This must be done with technical rigor, and is typically formalized in a verification

plan that explicitly states which parts of the standard are implemented and which are waived,

along with the rationale.

In low-budget missions or commercial environments, tailoring helps maintain compa-

tibility without imposing disproportionate requirements. Even agencies such as NASA and

ESA formally recognize this process. For example, ECSS-MS-ST-10-02 is entirely devoted to

proper tailoring practices.

4) Handbook: is a non-binding technical document that provides recommendations, best

practices, technical guidance, and application examples on a specialized topic. It does not im-

pose requirements, but helps interpret and apply them correctly. Handbooks are not subject

to compliance verification, audits, or certification, and they may include flexible or adaptable

content suited to various mission scenarios. They are mainly used to:

1. Clarify how to execute a specific technical procedure.

2. Present suggested methodologies, diagrams, checklists, or flowcharts.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino31

3. Support the design, testing, integration, or validation of systems.

4. Complement existing standards and regulations without replacing them.

In the space sector, handbooks are widely used, especially those issued by agencies

such as NASA or ESA, to accompany technical standards or to assist development teams

facing limitations in resources or infrastructure. For instance, NASA-HDBK-4002A offers

guidance on implementing electromagnetic compatibility (EMC) strategies and serves as a

companion to the MIL-STD-461G standard.

5) Payload User’s Guide (PUG): each launch provider typically specifies in this docu-

ment the type of technical references they use, whether enforcing strict compliance with a

standard, allowing flexible interpretation, or accepting equivalent internal testing. Outlines

the requirements, constraints, and operational conditions that payloads must meet to be

integrated and launched with a specific rocket. While PUGs are essential for planning the

design, integration, and validation of satellites or space devices, they cover multiple techni-

cal aspects beyond electromagnetic compatibility (EMC) and is always given by the launch

service provider.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino32

VII. METHODOLOGY

Before conducting any formal analysis, a comprehensive familiarization phase was

carried out to understand the scope and structure of the available information related to

electromagnetic compatibility (EMC) in spacecraft launches. This step was essential to de-

fine a realistic and structured methodological path, grounded on the actual accessibility of

technical and regulatory data.

The process began by identifying which companies currently offer commercial launch

services for spacecrafts and, more specifically, which of them provide access to Payload User’s

Guides (PUGs). A thorough review was conducted to assess the availability, completeness,

and technical depth of these documents. Particular attention was paid to those PUGs that

explicitly describe emission limits, frequency constraints, regulatory references, or verification

processes.

This early exploration revealed two parallel and complementary research opportuni-

ties. On one hand, the PUGs provided sufficient numerical and categorical data to analyze

the behavior of emissions generated or tolerated by the launch vehicle, spacecraft, and launch

site. On the other hand, many PUGs mentioned technical standards or constraints in a vague,

incomplete, or inconsistent manner, raising questions about the clarity and comprehensive-

ness of the requirements. This insight motivated the division of the study into two parts: (1)

statistical analysis of the electromagnetic environment, and (2) an analysis of electromagnetic

compatibility requirements and referenced standards.

This familiarization stage also allowed for the definition of a variable framework sui-

table for statistical analysis, identification of missing or ambiguous parameters, and stan-

dardization of data formats and units. EMC in space missions does not only concern field

strength and emission frequency, it also encompasses the processes of verification, validation,

and compliance negotiation between the CubeSat developer and the launch provider. As a

result, the methodology was structured to address both the technical emissions context and

the regulatory implications of EMC at launch.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino33

A. Part I: Statistical Analysis of the Electromagnetic Environment

In order to evaluate the electromagnetic environment to which a spacecraft may be

exposed during its launch, information was collected and analyzed regarding the electromag-

netic radiated emissions generated by the launch vehicle, the launch site, and the emission

limits allowed for spacecrafts. This information enables the characterization of the environ-

ment and the establishment of electromagnetic compatibility (EMC) criteria for future testing

or missions.

Since a spacecraft must comply with emission limits while also withstanding the elec-

tromagnetic radiation generated during launch, data were gathered from the publicly available

Payload User’s Guides (PUG) of various launch providers. These guides specify the environ-

mental conditions to which the payload is subjected, including maximum emission limits per

frequency and electric field intensity.

A structured database was built to compile detailed information on electromagnetic

emission limits and the operating parameters of launch vehicle communication systems. This

data organization allows comparative analyses across different launch actors and operational

contexts. However, not all launch companies provide free access to their PUGs. Some impose

confidentiality clauses, restricting access to those with signed contracts or pre-approved mis-

sion feasibility. Thus, the database was built using only publicly available PUGs, prioritizing

official, updated documents and selecting the most recent or detailed version when multiple

versions were available.

Despite these limitations, a significant amount of data was collected for variable analy-

sis. Table I provides a summary of the launch vehicles and deployment modules for which

information was gathered. It includes the rocket name, its owner and operator, manufac-

turer’s country, country of launch, and the PUG version used 1. This data offers a general

overview of the evaluated launch vehicles and their operational characteristics, helping iden-

tify the different infrastructures used for launches and the diversity of operators that share

technical information. The complete version is available for download in Appendix A.

1Note: Based on publicly available Payload User’s Guides [9, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32]

https://drive.google.com/drive/folders/12Q51cHNYELBOaeqv_j5Z4vz7nK5rk7Z1?usp=sharing


ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino34

Table I Summary of Launch Modules and Key Information

LV / Module Owner Operator Country

(Mfr.)

PUG

Version

Launch Site Country

(Launch)

Falcon 9 SpaceX SpaceX USA 2024 CCSFS SLC-40,

KSC

USA

Falcon Heavy SpaceX SpaceX USA 2021 VSFB SLC-4E USA

Electron Rocket Lab Rocket Lab New Zea-

land

2022 LC-1, LC-2 USA

New Glenn Blue Origin Blue Origin USA 2018 CCSFS LC-36 USA

Ariane 6 ArianeGroup Arianespace France,

Germany

2021 CSG French

Guiana

Soyuz RKTs Progress Roscosmos,

Arianespace

Russia 2018 CSG French

Guiana

Rocket-4 Astra Space Astra Space USA 2022 PSC LC-46,

CCSFS SLC-46

USA

H-IIA JAXA JAXA, MHI Japan 2015 TNSC Japan

Minotaur IV Northrop

Grumman

Northrop

Grumman

USA 2020 VSFB SLC-8 USA

Angara 1.2 Khrunichev International

Launch Services

Russia 2002 PC Russia

Proton RKTs Progress ILS, Roscosmos Russia 2009 BC LC-45 Kazakhstan

Long March 3A CASC CNSA, CASC China 2011 XSLC China

NRCSD NanoRacks NanoRacks USA 2018 ISS ISS

J-SSOD JAXA JAXA Japan 2023 ISS ISS

1) Structure and parameters of the database To ensure a comprehensive analysis, the fo-

llowing key parameters were defined and systematized in the database:

1. Rocket owner: company that owns the launch vehicle.

2. Operator: may differ from the rocket owner in cases where the launch is managed by

third parties.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino35

3. Launch vehicle name: specific identification of the launch vehicle.

4. Mission type, categorized as:

• Rideshare: shared with multiple payloads.

• Dedicated launch: exclusive mission for a single customer.

• Hybrid: applicable to both cases depending on the information provided in the PUG.

5. Deployment method, two modalities were considered:

• Rocket: traditional launch using an orbital vehicle.

• ISS (International Space Station): deployment from the station, included initially

due to its growing relevance in recent years.

6. Information category, divided into two groups:

• External emissions: related to the electromagnetic signals that affect or are generated

by the launch system.

• Operational characteristics: technical parameters of the launch vehicle’s radiofre-

quency systems.

7. Emission source, depending on the deployment method:

• For rocket-based launches:

• Maximum emissions generated by the launch vehicle.

• Maximum emissions allowed for the spacecraft.

• Maximum emissions generated by the launch site.

• For ISS deployment:

• Internal electromagnetic environment of the ISS.

• Internal electromagnetic susceptibility.

• Maximum emissions allowed for the spacecraft.

8. Emission type / Part description: defines the nature of the emission, indicating whether

it corresponds to a specific communication band (e.g., GPS, telemetry).

9. Subclassification: additional details on the emission, if applicable.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino36

10. Transmitter / Receiver: functional identification.

11. Frequency (MHz) and Electric field (dBuV/m): the two main variables used to quantify

electromagnetic emissions.

12. Safety margins, this information was recorded when provided by the analyzed docu-

ments, considering:

• Safety Margin (dB).

• EMI Safety Margin by Test (dB).

• EMI Safety Margin by Analysis (dB).

13. Adjusted Electric Field (dBuV/m) and Electric Field (V/m): adjusted values were

included when documentation presented different levels of attenuation or applied safety

margins.

14. 99% Bandwidth (MHz): operating frequency range of the communication system.

15. Modulation: type of modulation used in the signal.

16. Power (W) and Sensitivity (dBW): transmission and reception characteristics or system

capabilities.

17. Polarization: signal polarization type.

18. Antenna position: location of the transmitting/receiving system.

2) Analysis of the electromagnetic environment by actor type This analysis represents a

fundamental initial step in characterizing the electromagnetic environment to which space-

crafts are exposed during launch. By studying the emissions generated by rockets, launch

sites, and the limits imposed on payloads, the aim is to identify patterns and relationships

that help better understand the severity of the electromagnetic environment in commercial

space missions.

The objective is not to analyze the entire electromagnetic phenomenon nor to fully

model its physical behavior, but rather to study representative variables of the operational



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino37

conditions described by operators in their PUGs. It is essential to understand the behavior

of emissions and the factors that influence them, so that the requirements imposed by launch

providers gain meaning and justification for any stakeholder.

The collection and organization of the data serve multiple purposes:

• To assess the severity of the electromagnetic environment in actual launches based on

published specifications.

• To identify common trends or divergences among different launch vehicles and launch sites.

• To explore correlations between vehicle characteristics and the emissions they generate

or allow, and how those characteristics influence these values—helping to establish rela-

tionships between launch vehicle design/operation and its electromagnetic impact.

• To support the future definition of more representative test scenarios or acceptance criteria

based on observed evidence.

• To optimize launch vehicle selection—this emission analysis provides a foundation enabling

any stakeholder to evaluate which options are most suitable depending on the electromag-

netic requirements of each mission, thus facilitating planning and reducing risks.

The data are classified into three major groups according to the actor that generates

or regulates the emissions:

• Maximum radiated emissions generated by the launch vehicle: intentional or spurious sig-

nals produced during ignition, flight, and separation. These emissions can affect spacecraft

operation and must be considered when assessing the spacecraft’s electromagnetic suscep-

tibility.

• Maximum radiated emissions allowed for the spacecraft: emission limits imposed on spa-

cecrafts to ensure they do not interfere with the rocket or with other onboard payload.

These values are set by launch providers and must be met for mission approval.

• Maximum radiated emissions from the launch site: electromagnetic noise generated by

radars, communication systems, and other sources present in the spaceport infrastructure.

Since radiated emissions are expressed in terms of frequency (MHz) and electric field

strength (dBµV/m), the study focuses on evaluating two key parameters:



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino38

1. Frequency range amplitude: calculated as the difference between the maximum and

minimum frequencies at which electromagnetic emissions are expected, regardless of

the electric field strength within that range.

2. Electric field intensity range: defined as the difference between the maximum and mi-

nimum recorded field strength values, regardless of the frequency at which they occur.

These indicators quantify the extent of electromagnetic environments in terms of se-

verity and spectral coverage, providing a solid foundation for the characterization of the

electromagnetic conditions associated with launches.

Figure 5 presents a preliminary example of the methodological approach applied to

each of the collected electromagnetic emission categories. This graph visualizes the maximum

radiated emissions generated by rockets through two key parameters: frequency range and

electric field intensity range. These parameters are simultaneously compared with the rocket’s

length-to-diameter ratio (L/D) and its propulsion type.

Fig. 5. Example of correlation methodology



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino39

The color gradient of the bars represents the emission frequency range, with lighter

shades indicating lower frequencies and darker shades indicating higher frequencies. The

colors also distinguish propulsion type: green and yellow tones indicate liquid propulsion,

and blue indicates hybrid propulsion. The height of the bars represents the L/D ratio of each

rocket, allowing for the identification of whether vehicle dimensions may be associated with

emission magnitude. In addition, a red dotted line indicates the electric field intensity range

recorded for each rocket, helping to detect patterns related to physical design or propulsion

type.

The rockets are arranged along the horizontal axis in order of increasing frequency

range, which enables visualizing radiated emission behavior trends based on the vehicle.

This organization facilitates the identification of regularities and visual comparison among

different cases.

A relationship can be observed, for instance, between the propulsion type and the

frequency range: rockets with hybrid propulsion tend to show wider frequency ranges than

those with liquid propulsion, suggesting that a combination of technologies might influence

frequency dispersion. However, when comparing the rocket L/D ratio and the electric field

range, no clear patterns are observed, indicating that other factors may influence the mag-

nitude of emissions. These preliminary observations justify the need for a deeper statistical

analysis to better understand such variations.

The figure aims to simply illustrate the logic behind the upcoming analysis, which

will not be limited to the study of the L/D ratio but will also integrate other technical

characteristics of the launch vehicles along with the various categories of electromagnetic

emissions. The goal of this analysis is to identify relationships or patterns that help clarify

the EMC constraints imposed, as well as to build a more comprehensive understanding of

the factors that influence emissions during launch. This information will be key for designing

mitigation strategies and ensuring EMC compliance in future space missions.

The analysis will be carried out in four stages:

• Analysis of rocket characteristics in relation to the radiated emissions they generate.

• Analysis of rocket characteristics in relation to the radiated emission limits allowed for



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino40

spacecraft.

• Comparison of the severity of radiated emissions from launch sites versus those from rockets

and the limits imposed on spacecraft.

• Consideration of environmental conditions that affect the launch frequency.

It is important to distinguish between the launch site and the surrounding environ-

ment in which the process occurs. The launch site refers to the infrastructure used for liftoff,

including communication antennas, radars, and other equipment that contribute to elec-

tromagnetic emissions in the immediate area, while the environment encompasses external

factors that can influence the propagation and behavior of these emissions, as well as the

frequency of launches.

This analysis will allow the grouping of studied rockets and the establishment of a solid

foundation for comparing their radiated emissions under different criteria. In doing so, it will

be possible to identify which technical conditions are more strongly associated with severe

electromagnetic environments and which features should be prioritized when characterizing

such environments. This characterization is essential for guiding mitigation strategies and

informing future decisions regarding design, launch vehicle selection, or mission validation

under strict electromagnetic constraints.

It was not possible to carry out a conclusive analysis regarding the operational charac-

teristics, as these are highly specific to each launch vehicle. Therefore, no general relationship

could be established. The only consistent requirement is to ensure that there is no interfe-

rence with the rocket’s RF. However, the corresponding parameters were still included in the

database for the reader’s reference, even though no analysis was conducted on them.

Similarly, no analysis could be performed on the launch modules deployed from the

ISS, as data was only available from two companies offering this service. In both cases, the

information was unclear and incomplete. The sample size was not sufficient to support any

meaningful analysis, so this launch modality was only acknowledged in the study without

further exploration.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino41

3) Launch Vehicle maximum radiated emissions under worst-case scenario To analyze the

relationship between electromagnetic emissions and the characteristics of launch vehicles,

up to 17 parameters were collected for each rocket in the sample. These variables include

dimensions, mission type, propulsion type, number of stages, total number of engines, total

mass, payload capacity, among others. The selection of these parameters responds to the

need for a descriptive baseline of the rocket before diving into more specific analyses.

One of the main constraints of the study was the focus on launches to Low Earth

Orbit (LEO). Since several rockets in the sample are capable of missions to different orbits,

their characteristics may vary depending on the destination. Additionally, factors such as the

launch site and orbital inclinations largely depend on the type of mission. In cases where

a rocket could launch both to LEO and to GEO, significant differences were identified in

parameters such as the flight trajectory and launch vehicle configuration.

To avoid inconsistencies and ensure a homogeneous analytical framework, only data

corresponding to missions to LEO and launch sites compatible with this orbit were included.

This ensured that the emissions collected were within the same operational context and were

not influenced by factors associated with higher-orbit missions.

Since the first analysis focused on the maximum radiated emissions generated by

each rocket, priority was given to collecting those characteristics that could preliminarily

influence these values. Without delving into detailed technical aspects at the outset, this

initial dataset allowed for the exploration of possible relationships between the physical and

operational parameters of the launch vehicles and the frequency and electric field strength

values of their emissions.

As the analysis progressed, other factors were integrated to refine the assessment of

emissions and their impact. Nevertheless, this initial set of characteristics served as a starting

point to identify general trends and facilitate the grouping of rockets based on common

patterns. Regarding the emissions, the frequency spectrum width and electric field intensity

range were used to quantify the extent and variability of the electromagnetic signals generated

by each launch vehicle.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino42

1. Correlation analysis and variable importance: the first step consisted of calculating

a correlation matrix to identify the relationship between rocket characteristics and

their electromagnetic emissions. This matrix was built using the Pearson correlation

coefficient, which measures the linear association between two variables and ranges from

-1 to 1. Values close to 1 indicate a strong positive correlation (as one variable increases,

the other also increases), values close to -1 suggest a strong negative correlation (as

one variable increases, the other decreases), and values near 0 indicate no significant

correlation.

In cases where non-linear correlations were suspected, the Spearman correlation coef-

ficient was used. This coefficient measures the monotonic association between two va-

riables and also ranges from -1 to 1. Unlike Pearson, Spearman considers the rank of

the data instead of their absolute values, allowing the identification of relationships

that are not necessarily linear but follow an increasing or decreasing pattern. The p-

value associated with the coefficient indicates the statistical significance of the observed

correlation.

• If the p-value is less than 0.05, the observed correlation is considered statistically

significant and unlikely to have occurred by chance.

• If the p-value is greater than 0.05, the correlation may have occurred by chance and

is not considered reliable.

Due to the limited amount of available data and considering that the analysis is ex-

ploratory in nature, aimed at understanding general trends in the electromagnetic re-

quirements of launchers, a minimum significance threshold of ±0.35 was established

to consider a correlation minimally useful. Although this threshold may seem low for

purely statistical studies, it is considered reasonable in this context, since the goal is

not to generate definitive conclusions but rather to identify patterns that can guide

future studies[33].

This approach allowed the identification of relevant relationships without losing valua-

ble information at this initial stage. However, it is acknowledged that the significance



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino43

threshold may vary depending on the level of rigor required in future analyses.

In addition to correlation, feature importance was evaluated using the Random Forest

method[34], in order to determine the impact of each rocket characteristic on the fre-

quency range and electric field amplitude of its emissions. This approach allowed for

quantifying the relative influence of each feature in emission generation, providing a

more comprehensive view of the determining factors in the electromagnetic behavior of

rockets.

It is important to note that the features importance is interpreted as a relative measure

within the analyzed dataset. It is not compared across different studies nor expressed

in absolute values, but rather used to rank variables as having high, medium, or low

relevance according to their relationship with the evaluated response variable in each

case [35]

For example, in an analysis of frequency range, the most important feature might have

a value of 0.14, whereas in another analysis focused on electric field amplitude, the most

influential variable could reach a value of 0.58. This does not mean that one variable

is universally more relevant than another, but rather that within the context of that

specific analysis, its relative impact is greater.

To classify the importance of each feature in a specific analysis, the following ranges

were used:

• High importance: variables with a value of at least 70% of the most important

variable’s value.

• Medium importance: variables with a value between 40% and 70% of the most

important variable’s value.

• Low importance: variables with a value below 40% of the most important variable’s

value.

For example, if the most important variable has a value of 0.50, those with values ≥ 0,35

are considered of high importance; between 0.20 and 0.35, of medium importance; and

below 0.20, of low importance. This classification facilitates the interpretation of results



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino44

and helps identify the factors with the greatest influence on emissions for each specific

context.

2. Error assessment and iterative variable refinement: to ensure that the selected variables

provided relevant and useful information for the analysis, an error-based refinement

approach was implemented. The Mean Squared Error (MSE) was used as an indicator

of prediction accuracy, as it quantifies the average squared difference between actual

and predicted values.

To explore whether variable reduction improved model performance, the relative error

reduction was also calculated after removing variables. Although MSE does not in-

dicate statistical significance, noticeable reductions in error were interpreted as signs

that uninformative or redundant variables had been eliminated. This approach made

it possible to assess the contribution of individual variables to the overall prediction

accuracy.

In the first iteration, both the MSE and the relative error reduction values were not

satisfactory, suggesting the presence of noise in the dataset. Since the sample of roc-

kets was not extensive and technical characteristics had been collected without prior

knowledge of which ones were most relevant, an iterative variable refinement process

was necessary to improve the interpretability and robustness of the analysis.

To determine whether a variable should be discarded, three criteria were applied simul-

taneously:

• Low correlation with the frequency range.

• Low correlation with the electric field range.

• Low importance in both analyses (frequency and electric field) according to the

Random Forest model.

Only those variables that met all three criteria were removed from the analysis, ensuring

that potentially useful data were not prematurely discarded.

After this iterative process, the model showed improved accuracy, reflected in a lower



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino45

MSE and greater relative error reduction. At this point, a refined dataset was conso-

lidated and used as the basis for further analysis, ensuring that only variables with

significant influence on electromagnetic emissions were included.

3. Collinearity evaluation using the Variance Inflation Factor (VIF): before continuing

with the analyses, a final variable refinement step was performed using the Variance

Inflation Factor (VIF). This indicator quantifies collinearity between variables, i.e., how

redundant a variable is in relation to others. [36]

If a variable presents a high VIF, it means its information is already represented in other

variables, which can lead to overfitting and hinder the interpretation of the analysis. To

avoid this, a second iterative adjustment process was implemented, in which variables

with the highest VIF values were removed until an acceptable level of collinearity was

achieved. [37]

The reference values for interpreting VIF are as follows:

• VIF < 5 → Low collinearity (acceptable, no need for elimination).

• 5 ≤ VIF < 10 → Moderate collinearity (it is recommended to evaluate whether the

variable is necessary or can be removed).

• VIF ≥ 10 → High collinearity (the variable should be removed or adjusted, as it

indicates significant redundancy with other variables).

In this analysis, a general acceptance threshold of VIF < 10 was established, removing

variables that exceeded that limit. However, before opting for direct elimination, efforts

were made to adjust certain variables to reduce redundancy without losing relevant in-

formation. [38] An example was the features for rocket length and diameter: since it is

reasonable to assume that one is partially represented by the other, they were merged

into a single variable, resulting in the new variable Rocket L/D ratio. This transforma-

tion effectively reduced multicollinearity, and an improvement in model performance

was also observed, as evidenced by a lower MSE and greater relative error reduction.

This procedure ensured that the final variables were more independent from each other,

avoiding redundancies and ensuring that each contributed unique information to the



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino46

analysis. Once acceptable VIF values were achieved, the dataset was considered suffi-

ciently refined to proceed to the next stage of the study.

4) Spacecraft maximum radiated emissions This analysis was conducted by cross-referencing

the permitted emissions with the characteristics of the associated rockets, following an ap-

proach analogous to the study of emissions generated by the launch vehicles. However, the

characterization followed a different methodology, since the variables relevant to determining

the maximum limits allowed for spacecraft are not necessarily the same as those related to

the generation of emissions by the launch vehicle.

For this purpose, technical characteristics of the rockets that could be associated

with the restrictions imposed on spacecraft were collected, such as flight duration, payload

capacity, and other operational parameters that could influence the definition of emission

limits. Some of the variables used in the rocket emissions analysis were retained, while others

were discarded for not being relevant in this context. It is important to note that these

restrictions do not depend on the spacecraft’s structure itself, but rather on the criteria

defined by the rocket operator to prevent interference with their own systems or with co-

passenger satellites, depending on the mission type.

It was not possible to collect these values for all previously analyzed rockets, as not all

PUGs specified these limits. However, information was obtained for 10 of the 12 considered

launch vehicles, which allowed the construction of a representative sample. As in the rocket

emissions analysis, the same focus criteria on missions to Low Earth Orbit (LEO) was applied,

thereby ensuring data consistency and avoiding biases resulting from differences in mission

profiles.

Regarding the quantitative analysis, the same procedure used in the study of rocket

emissions was followed, again using the frequency range and the amplitude of electric field

intensity:

• Correlation matrix: correlations were calculated to evaluate the relationship between

rocket characteristics and the maximum radiated emissions allowed for spacecrafts.

• Variable importance: the Random Forest model was used to determine which launch



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino47

vehicle features have the most significant influence in defining these limits.

• Categorical variable analysis: to evaluate relationships within categorical data, box-

plots were used, since the correlation coefficient is not sufficient for this type of data.

• Variable refinement: an iterative process was applied to refine the variable set. Initially,

collinearity between some parameters was detected and unsatisfactory Mean Squared Error

(MSE) values were observed. Characteristics that could be redundant were adjusted, ge-

nerating an optimized dataset and retaining only those variables with a significant impact

on the prediction of the maximum permissible limits.

5) Comparison of launch site emissions with launch vehicle and spacecraft limits

a) Launch site radiated emissions relative to spacecraft limits To assess the relative

severity of the electromagnetic environment at launch sites, the initial plan was to apply

the same statistical approach used in the analysis of launch vehicles emissions and the limits

allowed for spacecrafts. However, it was identified that data was available for only five specific

launch sites regarding their maximum radiated emissions. This limitation prevented the use

of statistical tools such as correlations or direct impact estimates, due to the insufficient

sample size.

Although some rockets share the same launch site, not all PUGs include information

about the emissions generated at these sites. It is not accurate to assume that a launch site

maintains constant emissions between missions. Each launch may involve unique configura-

tions depending on the type of vehicle, payload, antenna arrangements, used frequencies, and

other technical variables that influence the emission profile. Therefore, assuming fixed values

between missions could introduce significant bias.

Given this context, an alternative approach was adopted: a comparative analysis bet-

ween the maximum emissions generated by the launch site and the emission limits allowed

for spacecraft. This strategy allows for assessing whether the electromagnetic environment

generated by the ground infrastructure represents a more demanding condition than what

the spacecraft is authorized to tolerate. Therefore, it may indicate the need for additional



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino48

immunity margins to operate safely.

It is important to note that spacecraft maximum emissions are not measured values,

but limits specified by the launch provider to ensure electromagnetic compatibility during

liftoff and in orbit. This analysis does not evaluate the spacecraft’s performance, but rather

how its emission limits compare to the electromagnetic environment generated by the planned

launch site.

Since the emission scales between spacecraft and launch sites are different, a norma-

lization process was applied. This made possible to express the values on a relative scale,

suitable for proportional comparison. In this way, it was possible to assess how demanding

the launch site environment is in relation to the emission limits set for the spacecraft, without

magnitude differences distorting the results.

Two normalized metrics were defined: Electric Field Relative Emission (EFRE) and

Frequency Relative Coverage (FRC):

• EFRE assesses whether the emissions at the launch site exceed the maximum limit allowed

for a spacecraft, which would indicate that the latter must be prepared to operate in a

more severe environment.

• FRC analyzes whether the electromagnetic spectrum covered by the launch site is broader

than that of the spacecraft, which could pose interference risks outside the spacecraft’s

operational range.

The equations used were:

EFRE =
E- Field range of LS emissions (dBµV/m)

E- Field range of allowed emissions for S/C (dBµV/m)
(2)

FRC =
Frequency range of LS emissions (MHz)

Frequency range of allowed emissions for S/C (MHz)
(3)

Both metrics were calculated for the available rocket–launch site pairs. EFRE allows

assessing whether the launch site environment is more or less demanding than the spacecraft’s

allowable limit. This is relevant because a higher electric field may increase the likelihood of

interference.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino49

• If EFRE > 1, the site is more severe than the permitted limit → additional immunity may

be required.

• If EFRE < 1, the site poses more benign conditions.

• If EFRE ≈ 1, both are aligned, indicating favorable compatibility.

Similarly, FRC indicates whether the site covers a broader frequency range:

• If FRC > 1, there may be a risk of interference outside the spacecraft’s operational range.

• If FRC < 1, the spacecraft operates in a broader range than the site, which may require

additional testing in those bands.

• If FRC ≈ 1, both operate in similar ranges, favoring spectral compatibility.

The results were represented by a scatter plot, where each point shows a rocket–launch

site pair. The X-axis represents the FRC and the Y-axis the EFRE. This visualization allowed

the identification of which sites present more demanding environments compared to spacecraft

limits, and which, on the contrary, offer more benign conditions.

b) Launch site radiated emissions relative to launch vehicle emissions Continuing with

the evaluation of the relative severity of the electromagnetic environment, a comparative

analysis was conducted between the emissions generated by the launch vehicle and those

generated by the launch site. The objective was to identify whether the electromagnetic con-

ditions associated with the launch site can exceed, complement, or mitigate those generated

by the rocket, which is critical when interpreting the environmental conditions during liftoff.

The equations used were:

EFRE =
E- Field range of LS emissions (dBµV/m)

E- Field range of LV emissions (dBµV/m)
(4)

CRF =
Frequency range of LS emissions (MHz)

Frequency range of LV emissions (MHz)
(5)

Both metrics were calculated only for cases where complete information was available

on the maximum emissions of both the rocket and its associated launch site.

To visually represent the results of this comparative analysis, a scatter plot was created

in which each point represents a rocket–launch site pair. This visualization helps identify



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino50

which launch sites may present more demanding or more benign electromagnetic conditions

than those generated by the launch vehicle itself. It also facilitates comparison between sites

based on the relative severity of their environment, helping distinguish the most critical ones

from those with more favorable profiles.

This analysis does not aim to establish a causal relationship between launch vehicle

emissions and launch site emissions. Both sources are considered independent emitters that

coexist during the liftoff phase. The goal is to compare their relative severity to understand

which represents a greater challenge from the electromagnetic perspective, especially when

defining immunity margins or testing conditions for space systems.

6) Environmental factors influence on launches The last factor considered in this study

corresponds to the environmental conditions associated with launch sites. Although the main

objective is not to evaluate how the environment directly affects the electromagnetic emissions

generated by the launch sites, it is essential to acknowledge that the climate plays a relevant

role from an operational perspective.

Environmental conditions can significantly influence aspects such as the planning of

launch windows, the frequency of use of certain infrastructures, as well as the definition of

additional requirements by the providers. Factors such as temperature, humidity, atmospheric

pressure, and precipitation affect not only launch logistics but also the necessary precautions

for transportation, storage, and operation of spacecraft, especially in extreme climates where

additional protections or specific electromagnetic compatibility (EMC) tests may be required.

This analysis does not aim to establish direct causal relationships between the en-

vironment and electromagnetic emissions, but rather to highlight how climate conditions

constitute an operational factor that may affect the preparation, validation, and execution

of missions. For example, in environments with high humidity or constant precipitation, the

risks associated with exposing sensitive equipment increase, which may lead to additional

protection requirements or pre-launch testing.

To contextualize this environmental influence, relevant information was collected on

the geographic location, launch activity, and typical climate conditions of the launch sites



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino51

associated with the rockets analyzed in previous chapters. Table X presents a detailed sum-

mary that includes the total number of launches recorded between 2018 and 2025, as well as

the typical ranges of temperature, humidity, pressure, and precipitation for each site.

Additionally, the Köppen climate classification was included, which allows standardi-

zed characterization of the climate types present in each region. This classification helps to

better understand the environmental particularities of each site, providing a useful reference

framework to identify potential operational implications related to climate.

Launch Sites and Their Typical Environmental Conditions

Launch

Site

Location Launches

(2018–2025)

Temperature

(°C)

Humidity

(%)

Pressure

(mbar)

Precipitation

(mm/year)

Köppen

SLC-39A Merritt Island (USA) 94 22 – 24 70 – 80 1012 – 1019 1300 – 1550 cfa

SLC-36 Cape Canaveral

(USA)

1 22 – 24 70 – 80 1012 – 1019 1300 – 1550 cfa

SLC-40 Cape Canaveral

(USA)

208 22 – 24 70 – 80 1012 – 1019 1300 – 1550 cfa

LC-46 Cape Canaveral

(USA)

6 22 – 24 70 – 80 1012 – 1019 1300 – 1550 cfa

LC-2 Wallops Island (USA) 28 11 – 17 65 – 80 1000 – 1100 1000 – 1100 cfa

MARS Wallops Island (USA) 17 11 – 17 65 – 80 1000 – 1100 1000 – 1100 cfa

SLC-4E Vandenberg (USA) 114 12 – 18 65 – 75 1012 – 1018 300 – 400 Csb

LC-1 Mahia Peninsula (NZ) 26 12 – 16 75 – 85 1009 – 1018 1150 –1250 cfb

GSC Kourou (French

Guiana)

12 26 – 27 80 – 90 1010 – 1014 2800 – 3000 af

TNSC Tanegashima (Japan) 10 19 – 24 75 – 85 1012 – 1022 2300 – 2550 cfa

XSLC Xichang (China) 1 16 – 18 70 – 80 1005 – 1016 1250 – 1290 cwb

JSLC Ejin (China) 2 7 – 10 30 – 45 1009 – 1018 38 – 42 BWk

BK Baikonur

(Kazakhstan)

52 9 – 11 30 – 50 1009 – 1018 140 – 160 BWk

PC Mirny (Russia) 15 0 – 4 60 – 80 1010 – 1018 590 – 630 dfc

VTNY Tsiolkovsky (Russia) 18 1 – 5 55 – 75 1010 – 1020 490 – 530 Dwb

Table II Environmental Conditions and Launch Activity of the Main Launch Sites (2018–2025).

The information presented in Table X was gathered from various sources, including



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino52

historical weather reports [14][15] and space launch databases available on the official web-

sites of service providers, as well as recognized platforms such as Gunter’s Space Page and

Space Launch Report [16][17]. This table provides a contextual summary of the typical en-

vironmental conditions and recorded activity at each launch site between 2018 and 2025.

Although the table summarizes the main data, the information collected was far more

extensive, including the total number of launches per year for each site during the period of

interest. Before deciding whether to perform the analysis using annual or monthly data, the

distribution of launches per month and per year was calculated to identify possible seasonal

concentrations or recurring operational windows.

To do this, the monthly launch concentration was calculated using the following ex-

pression:

Monthly Concentration =
Number of launches in the month

Total launches in the year
× 100 (6)

As a criterion, a monthly concentration was considered significant if at least 33% of

the annual launches occurred in the same month. However, in most of the analyzed sites, no

representative concentrations were identified. In some cases, a low number of annual launches

led to false positives (e.g., when only two launches were conducted in the year, distributed

in different months). Furthermore, when calculating average monthly values independently

of the year, no clear seasonality was observed, with results showing high dispersion or low

launch frequency.

It is worth noting that some sites, such as Plesetsk Cosmodrome, showed high activity

in years prior to 2015, but during the analyzed period their use was sporadic. The 2018–2025

range was selected because it aligns with the availability of most PUGs and the rise of

commercial launches as services, making it more representative for this study.

Once it was determined that the analysis would be conducted on an annual basis to

avoid introducing noise into the data, launches were filtered to exclude those targeting orbits

other than LEO, as well as specific missions to the International Space Station (ISS) or to

SSO trajectories. This was done because the focus of this work is exclusively on spacecraft

launches to LEO.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino53

Additionally, the success or failure rate of the launches was not considered, since

the interest lies solely in the operational activity of the launch site, without evaluating the

performance of the rockets. Likewise, suborbital launches were excluded, including both test

missions and space tourism activities.

This decision is based on the fact that the electromagnetic emissions of interest are

generated during the launch process itself, regardless of the mission outcome. The objective

is to evaluate the launch frequency and environmental conditions associated with the launch

event, as this is the moment that generates the electromagnetic activity relevant to the

present study.

Regarding environmental conditions, since each site’s typical climate is not conti-

nuous, it was necessary to obtain annual average values of temperature, humidity, pressure,

and precipitation for each year with launch activity. As this information was not consolidated

annually in the sources consulted, historical monthly averages were used [14]. Then, the an-

nual average of each variable was calculated from the monthly values, except for precipitation,

where the monthly values were summed to obtain the total annual precipitation.

Finally, the Köppen climate classification was included as an additional qualitative

reference. This classification allowed for a standardized description of the environment at

each launch site, facilitating the interpretation of general operational conditions and serving

as a foundation for subsequent analyses on the possible implications of the climate on mission

planning, spacecraft protection, and EMC test validation.

To analyze the influence of environmental conditions on launch site activity, a similar

approach was applied as previously used in the study of electromagnetic emissions. First, it

was evaluated whether there was any relationship between environmental conditions and the

number of annual launches recorded at each site.

The correlation matrices were calculated using two complementary methods:

• Pearson correlation, used to identify linear relationships between numerical variables.

• Spearman correlation, suitable for detecting monotonic relationships, especially in cases

where linearity cannot be assumed or when outliers are present.



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino54

This approach allowed for the exploration of potential associations between variables

such as temperature, humidity, pressure, and precipitation with the frequency of launches.

Regarding the categorical variable corresponding to the Köppen climate classification,

a descriptive analysis was conducted. The data were grouped by each climate category, and

the total number of launches associated with each climate type was calculated. This procedure

helped identify which climate types are more common in space operations, complementing

the analysis with a qualitative interpretation of the results.

It was not considered necessary to cross-reference the quantitative environmental va-

riables with the Köppen classification, as the latter is, by definition, a synthesis of those

environmental conditions. Analyzing those relationships would be redundant and would not

contribute value to the objective of the study.

In addition to the quantitative analysis through correlations and groupings by climate

type, further visualizations were created to support the interpretation of the results. Trend

charts were generated for the average environmental variables (temperature, humidity, pres-

sure, and precipitation) considering all launch sites, in order to observe the overall evolution

of the operational environment between 2018 and 2025.

Analyzing these average environmental trends also makes it possible to identify

whether significant changes occurred in the general conditions of the launch sites during

the study period, such as unusual temperature increases, decreases in atmospheric pressure,

or notable variations in humidity and precipitation. This is essential to detect the presence of

anomalous years which, although not necessarily reflected immediately in launch frequency,

could have important implications from an operational and spacecraft preparation stand-

point. This perspective complements the Köppen classification analysis: while the climate

grouping describes the characteristic behavior of different environmental types, global trends

help understand the overall evolution of the environment over time.

Additionally, a bar chart was developed showing the total distribution of launches

grouped by Köppen climate classification, allowing for the identification of whether certain

climate types exhibit a higher concentration of space activity. Furthermore, trend graphs of

the number of annual launches per Köppen category were created to analyze the temporal



ANALYSIS OF THE EMC PHENOMENON USING STATISTICAL CORRELATIONS Rufino55

evolution of launch activity based on predominant climate conditions.

These graphical representations complement the statistical analysis by providing a

visual perspective that facilitates the identification of relevant patterns and trends in the

relationship between the climate environment and the frequency of space operations.

7) Pattern analysis and justification for emission data segmentation After analyzing the

emissions of the rockets, it was observed that, although there were general relationships

between their characteristics and the values of frequency range and electric field strength,

the launch vehicles did not behave uniformly. When visualizing the emission plots, certain

natural groupings were identified, where some rockets exhibited relatively similar emission

patterns in terms of frequency range and electric field amplitude.

As part of the analysis of rocket characteristics, it was important not only to identify

which variables influenced the emissions, but also to assess how similar certain rockets were to

each other based on these fact