⚛️ phenomenology

GGI Lectures on Large-Scale Structure Perturbation Theory (Effective Field Theory)

This paper provides a pedagogical introduction to non-linear perturbation theory for cosmological large-scale structure, specifically developing the Effective Field Theory framework from scratch to model galaxy clustering, address the limitations of Standard Perturbation Theory, and incorporate galaxy bias and redshift-space distortions for undergraduate and beginning graduate students.

Original authors: Mikhail M. Ivanov

Published 2026-07-31

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mikhail M. Ivanov

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, invisible ocean made of dark matter. Long ago, this ocean was almost perfectly smooth, like a calm lake. But tiny ripples, caused by the Big Bang, started to grow. Over billions of years, gravity pulled these ripples together, turning them into massive waves, then crashing surf, and finally, into the towering islands and deep valleys we see today: the galaxies and the vast empty spaces between them. Astronomers are like divers trying to map this underwater world, but they can't see the water directly; they only see the islands (galaxies) that float on top. The challenge is that the water gets very turbulent and messy near the islands, making it hard to predict exactly where the next wave will crash. To understand this cosmic dance, scientists use a set of mathematical tools called "perturbation theory," which is essentially a way of guessing the shape of the waves by starting with a simple, smooth guess and adding corrections for the messy parts.

This paper, written by Mikhail M. Ivanov, is a guidebook for a new, more powerful way to make those guesses. It introduces a method called the "Effective Field Theory" (EFT) of Large-Scale Structure. Think of EFT as a clever trick used by physicists when they don't need to know every single detail of a complex system to understand the big picture. Just as you can describe the flow of a river using simple rules about water pressure and speed without tracking every single water molecule, EFT allows cosmologists to describe the clustering of galaxies without needing to solve the impossible math of every tiny, chaotic interaction happening inside a galaxy. The paper argues that the old methods were failing because they tried to ignore the messy small-scale details, leading to wrong answers. Instead, this new approach admits that the small scales are messy, but it captures their effect on the large scales using a few simple "knobs" or parameters that can be tuned to match reality.

The core of the paper is a detailed construction of this new theory. The author starts by showing why the old "Standard Perturbation Theory" (SPT) breaks down. Imagine trying to predict the weather by only looking at the average temperature; it works for a day, but eventually, the local storms (the small-scale chaos) mess up your prediction. In the universe, these "storms" are the violent collisions and clumping of dark matter. The paper demonstrates that when you try to calculate how galaxies cluster using the old method, the math blows up, giving infinite or nonsensical results because it tries to account for these tiny, chaotic interactions using rules that only work for smooth, calm water.

To fix this, the paper develops a new set of rules based on symmetry and the "Effective Stress Tensor." This is a fancy way of saying that the messy small-scale physics acts like an invisible pressure or friction on the large-scale flow. The author shows that by adding a few extra terms to the equations—terms that represent this "friction" and random noise—the theory becomes stable and accurate. These extra terms act like "counterterms," which are mathematical patches that cancel out the infinite errors caused by the small-scale chaos. The paper proves that these patches are not just random guesses; they are forced by the fundamental laws of physics, such as the conservation of mass and momentum.

One of the most exciting findings in the paper is how it handles the "Baryon Acoustic Oscillations" (BAO). These are faint, ring-like patterns in the distribution of galaxies, essentially a "fossil" sound wave from the early universe. The old methods struggled to predict how these rings would look after billions of years of cosmic turbulence, often smearing them out incorrectly. The paper shows that by using a technique called "IR resummation" (which is like re-summing the effects of the long, gentle waves that push the galaxies around), the theory can perfectly preserve the shape of these rings. This is crucial because these rings act as a "standard ruler" for measuring the expansion of the universe.

The paper also tackles "Galaxy Bias," which is the fact that galaxies don't form randomly; they prefer to form in the densest parts of the dark matter ocean. The author explains that while the process of galaxy formation is incredibly complex and involves stars, gas, and black holes, the large-scale pattern of galaxies can still be described by a simple expansion, provided we include a few extra "bias" parameters. These parameters act as a bridge, translating the smooth dark matter map into the specific, clumpy map of galaxies we observe.

Furthermore, the paper addresses "Redshift-Space Distortions." When we look at galaxies, we measure their speed based on how much their light is shifted (redshifted). But galaxies are also moving due to their own gravity, not just the expansion of the universe. This makes the universe look squashed or stretched in our maps, like a funhouse mirror. The paper provides a rigorous way to untangle this distortion, showing that the messy "fingers of God" (elongated streaks of galaxies caused by fast-moving clusters) can be modeled as random noise and friction, allowing us to see the true underlying structure.

Finally, the author validates this entire theory by comparing it to massive computer simulations of the universe. The results are striking: the new EFT method matches the simulation data with incredible precision, far better than the old methods, and it works up to much smaller scales (higher "k" values) than previously thought possible. The paper concludes that while the small-scale universe is chaotic and unpredictable, the large-scale structure follows a predictable, elegant pattern that can be unlocked by acknowledging the chaos rather than ignoring it. It's a triumph of understanding the big picture by respecting the messy details.

Technical Summary: GGI Lectures on Large-Scale Structure Perturbation Theory (Effective Field Theory)

Problem Statement

The formation of cosmic structure involves a transition from linear, Gaussian initial conditions to highly non-linear, virialized states (halos and galaxies). While linear perturbation theory accurately describes large scales, it fails on small scales where gravitational collapse occurs. Historically, Standard Perturbation Theory (SPT) was extended to higher orders to model these non-linearities. However, SPT exhibits fundamental inconsistencies:

Infrared (IR) Divergences: Loop corrections in SPT contain terms enhanced by the displacement variance of large-scale modes. While leading-order IR terms cancel due to the equivalence principle, sub-leading terms cause the perturbative expansion to break down for specific features like Baryon Acoustic Oscillations (BAO), even on scales where the broadband power spectrum remains perturbative.
Ultraviolet (UV) Sensitivity: Loop integrals in SPT receive contributions from small-scale modes ( $k \gtrsim k_{NL}$ ) where the fluid approximation of collisionless dark matter is invalid. SPT predicts unphysical behavior (e.g., divergence or incorrect scaling) in these regimes because it lacks the necessary counterterms to absorb UV sensitivity.
Galaxy Bias Complexity: Observations measure galaxies, not dark matter. The relationship between galaxies and the underlying matter field (bias) is non-linear and stochastic. Phenomenological models often lack a systematic expansion or fail to respect fundamental symmetries.

The core problem is to develop a systematic, predictive framework for the non-linear clustering of matter and galaxies that respects the separation of scales, incorporates the symmetries of the underlying physics, and remains valid on quasi-linear scales relevant for modern cosmological surveys.

Methodology

The paper adopts the Effective Field Theory (EFT) framework, treating Large-Scale Structure (LSS) as a low-energy effective theory where short-scale physics is integrated out and represented by local operators and stochastic noise. The methodology proceeds through the following steps:

1. EFT Ideology and Symmetry Principles

The lectures establish that LSS can be modeled by isolating long-wavelength physics from short-scale complications. The effective description is constrained by:

Separation of Scales: $L_{long} \gg L_{short}$ .
Symmetries: Galilean invariance, rotational invariance, and the equivalence principle.
Derivative Expansion: An expansion in powers of $k L_{short}$ and field amplitudes.

The equations of motion for the dark matter fluid are derived by including an effective stress tensor ( $\tau_{ij}$ ) to account for the non-linear dynamics of small-scale structures (halos, shell-crossing). This tensor is not zero, contrary to the perfect fluid assumption in SPT.

2. Renormalization of Loop Diagrams

The paper demonstrates that loop corrections in SPT are UV-sensitive. To cure this:

Counterterms: The effective stress tensor generates counterterms (e.g., effective sound speed $c_s^2$ ) that absorb the UV dependence of loop integrals.
Stochasticity: Short-scale modes that have decoupled from long-wavelength initial conditions contribute a stochastic noise term. Mass and momentum conservation dictate that this noise scales as $k^4$ for dark matter, distinguishing it from the $k^2$ scaling of the sound speed counterterm.
Matching: The coefficients of these counterterms (e.g., $\gamma$ ) are not predicted by the EFT itself but must be matched to data or N-body simulations.

3. IR Resummation and BAO

The paper addresses the breakdown of perturbation theory for Baryon Acoustic Oscillations (BAO) caused by large-scale bulk flows.

Equivalence Principle: It is shown that the power spectrum is invariant under homogeneous shifts of the matter distribution. This symmetry forces the cancellation of leading IR-enhanced terms in loop diagrams.
Resummation: The remaining sub-leading IR effects, which damp the BAO peak, are resummed to all orders. This is achieved either via Eulerian resummation or, more naturally, via Lagrangian Perturbation Theory (LPT), where the displacement field is kept non-perturbative.

4. Galaxy Bias and Redshift-Space Distortions (RSD)

The framework is extended to biased tracers (galaxies):

Bias Expansion: The galaxy density is expanded in terms of the matter density, tidal fields, and stochastic noise, respecting the equivalence principle. This introduces a hierarchy of bias parameters ( $b_1, b_2, b_{G_2}$ , etc.).
RSD: The mapping from real space to redshift space is treated non-perturbatively. The resulting kernels are renormalized, introducing new counterterms specific to redshift space (e.g., $k^2 \mu^2$ and $k^2 \mu^4$ terms) to account for the "Fingers of God" effect and velocity dispersion.

5. Lagrangian Perturbation Theory (LPT)

LPT is presented as a complementary approach where the displacement field $\Psi$ is the fundamental variable. The exact power spectrum is expressed via a cumulant expansion of the displacement. LPT naturally implements IR resummation and provides a clear physical picture of why bulk flows damp BAO but do not affect the broadband power spectrum at leading order.

Key Contributions and Results

1. Systematic EFT of Dark Matter Clustering

The paper derives the one-loop matter power spectrum in EFT, showing it takes the form:
$P_{EFT}(k) = P_{SPT}(k) - 2\gamma k^2 P_{lin}(k) + P_{stoch}(k)$
where $P_{SPT}$ is the standard loop result, $-2\gamma k^2 P_{lin}$ is the counterterm absorbing UV sensitivity (effective sound speed), and $P_{stoch} \propto k^4$ is the stochastic contribution.

Result: This renormalized theory agrees with N-body simulations up to $k \approx 0.15 \, h \text{Mpc}^{-1}$ at one-loop order and extends to $k \approx 0.25 \, h \text{Mpc}^{-1}$ at two-loop order, significantly outperforming unrenormalized SPT.

2. IR Resummation of BAO

The paper provides a rigorous derivation of the IR resummation formula, showing that the BAO peak is damped by a factor $e^{-k^2 \Sigma^2}$ .

Result: This resummation restores the accuracy of perturbation theory for the BAO feature, which otherwise suffers from large $O(1)$ corrections in fixed-order SPT. The damping factor $\Sigma^2$ is explicitly calculated and shown to be consistent with the equivalence principle.

3. General EFT Bias Expansion

A complete basis of bias operators is constructed up to third order, including tidal terms ( $G_2, \Gamma_3$ ) and higher-derivative terms ( $\nabla^2 \delta$ ).

Result: The paper identifies which operators are redundant at one-loop order (e.g., $\delta^3$ ) and which are necessary. It establishes that the stochastic power spectrum of galaxies scales as $k^0$ (shot noise) rather than $k^4$ , due to the lack of mass conservation for galaxy number density.

4. Redshift-Space Distortions (RSD) in EFT

The lectures derive the full one-loop RSD power spectrum, including the renormalization of contact operators involving velocity fields.

Result: The theory predicts a specific angular dependence for counterterms (up to $\mu^4$ ) and stochastic terms. The inclusion of these terms allows for a precise fit to redshift-space data up to $k_{max} \approx 0.2 \, h \text{Mpc}^{-1}$ . The paper highlights that the "Fingers of God" effect is captured by the $k^2 \mu^2$ and $k^2 \mu^4$ counterterms and stochastic noise.

5. Field-Level Validation

The paper emphasizes that EFT is not merely a curve-fitting exercise.

Result: Comparisons with simulations at the field level (matching phases, not just amplitudes) demonstrate that the EFT model, with a finite number of parameters, successfully reconstructs the galaxy density field. This validates the structural correctness of the operator basis and the displacement structure predicted by the theory.

Significance and Claims

The paper claims that the EFT of LSS provides the only known systematic, first-principles approach to modeling non-linear structure formation that is consistent with the symmetries of General Relativity and Newtonian gravity.

Resolution of SPT Failures: The paper asserts that the failures of SPT (divergence, incorrect scaling, inability to model BAO) are not due to the need for "resummation" of the wrong theory, but rather the absence of the correct physical ingredients (counterterms and stochasticity) required by the EFT framework.
Predictive Power: By separating calculable long-wavelength physics from non-perturbative short-wavelength physics (encoded in a few parameters), the EFT allows for precision cosmology. The parameters (bias, sound speed, stochasticity) can be fitted to data or simulations, but the functional form of the power spectrum is derived from symmetry.
Universality: The framework is claimed to be universal; the same EFT logic applies to dark matter, biased tracers (galaxies), and redshift-space distortions, with the specific physics of galaxy formation absorbed into the bias coefficients.
Robustness: The paper argues that the success of EFT in fitting simulation data and real survey data (e.g., DESI, BOSS) up to quasi-linear scales is a robust confirmation of the theory, distinct from over-fitting, as evidenced by field-level tests.

In summary, the lectures present the EFT of LSS as a mature, rigorous framework that has superseded Standard Perturbation Theory for precision cosmology, enabling the extraction of cosmological parameters from the non-linear regime of the universe.