For decades, Surfline focused on the coast with one goal: helping surfers score better waves. Over time, it became clear that the tools we built to fuel our surf obsession had other applications as well—like protecting coastlines, communities, and even surf spots themselves.

That realization led to the development of Surfline Coastal Intelligence (SCI), a new system that extracts data from decades of coastal imagery to generate insights to guide smarter coastal decisions.

Powered by machine learning and our global camera network, SCI helps track how beaches evolve, how waves are affected, and empowers those we work with in coastal communities to make more informed decisions on how to protect the people that enjoy them.

Protecting Our Beaches

SCI captures how sand moves in response to swell, storms, sea level rise, and development. That data helps cities evaluate beach nourishment projects, test new protections, and better understand which interventions are working, and which aren’t.

Check out our recent analysis of Sessions data and RE:BEACH case study that highlight how SCI is working to protect the beach and surf environment in Oceanside, CA.

Sessions Report RE:BEACH Case Study

ReportAssessing Surfline Sessions^™ Data at Oceanside, CA

A quantitative analysis of surfed waves, built from more than 700,000 per-second GPS-tracked rides recorded through Surfline Sessions.

Prepared by Michael Reinwald, Director of Science & Innovation · Observation window Jan 2019 – May 2026 · 77,394 sessions · 4,335 surfers · Issued May 2026

01Executive summary

This study establishes a quantitative baseline of surfed-wave characteristics at Oceanside, CA, using Surfline session data. The methodology is built on Surfline's own per-second GPS tracks, recorded through the Surfline Sessions app and supported wearables (Apple Watch, Rip Curl SearchGPS). We focus on ride length and surfer speed, reported both as overall distributions and against the environmental conditions on the wave (tide level, significant wave height, peak period, and incoming swell direction).

Sessions queried

77,394

2019–2026

Surfed waves

718,321

Distinct surfers

4,335

Typical ride

56 mmean length

21.3 km/hmean peak speed

02Where the surfing happens

The take-off density across the ten-kilometre stretch is heavily concentrated at two locations: the inside of the harbor's south jetty, and the sandbar either side of the pier.

takeoff heatmap — Figure 2.1Density of wave take-offs — the first GPS sample of each ride — counted on a 5 m × 5 m grid, log-scaled.

track overlay — Figure 2.2Sub-sample of individual ride tracks at the two hotspots: **Oceanside Pier** on the left, the **harbor mouth** (south jetty) on the right. Each segment is coloured by along-track speed (km/hr).

seasonal diff — Figure 2.3Where each season's take-offs are over-represented relative to the other. Red cells are favoured in summer, blue cells in winter; cells with very little activity in either season are masked. Each season is normalised to its own total before the subtraction, so the colour reflects *relative* seasonal preference rather than absolute count.

03When the surfing happens

Logged activity grew rapidly through the late 2010s and has stabilised since. Monthly counts peak in winter and dip in late spring. Hourly counts peak in the early morning, with a smaller secondary peak in the late afternoon.

rides per year — Figure 3.1Rides per day, by year. Each year's total is divided by the number of days that year contributes to the data window — shown in grey under each bar — so partial years (the current year, and any year where the data series starts or ends mid-year) are directly comparable to full ones.

rides per month — Figure 3.2Rides aggregated by calendar month across all years. Months are coloured to mark the summer / winter split used in the rest of the report.

rides per month per year — Figure 3.3Monthly ride counts broken out year-by-year.

rides by hour — Figure 3.4Ride counts by local hour of day.

04What the rides look like

Once we know where and when, the obvious next question is how long the rides are and how fast surfers are travelling. We report ride length and surfer speed as overall distributions, then break them down by tide and wave height, and finally show the joint distributions. A full summary table follows.

speed length hists — Figure 4.1Distributions of average speed, maximum speed, and ride length for all rides (top row), summer rides (middle), and winter rides (bottom). Vertical overlays mark the median (solid), mean (dashed), and 95^th percentile (dotted).

tide / wave height hists — Figure 4.2Ride counts grouped by tide-level quintile (top row) and significant wave height (bottom row, fixed 0.5 m bins), split into all rides, summer, and winter columns. Tide quintiles are computed from the session-level tide distribution (one observation per session, not per wave) — so uneven bars indicate preferred tide windows rather than a definitional artifact.

max speed vs length — Figure 4.3Smoothed 2-D density of maximum surfer speed versus ride length.

duration vs length — Figure 4.4Smoothed 2-D density of ride duration versus ride length.

Summary statistics

Subset	n	Ride length (m)				Avg speed (km/hr)		Max speed (km/hr)
Subset	n	Mean	Med	P95	Max	Mean	Med	Mean	Med	P95	Max
All waves	718,321	56	48	125	300	15.5	15.3	21.3	19.8	36.0	50.0
By tide level
≤ 0.49 m	141,297	55	47	122	300	15.5	15.3	21.3	19.8	36.5	50.0
0.49 – 0.83 m	137,788	55	47	123	300	15.4	15.2	21.2	19.6	36.0	50.0
0.83 – 1.07 m	146,431	55	47	125	300	15.4	15.2	21.1	19.6	35.8	50.0
1.07 – 1.37 m	139,694	56	48	127	300	15.5	15.3	21.2	19.8	35.9	50.0
> 1.37 m	142,922	58	50	129	300	15.7	15.5	21.5	20.1	35.9	50.0
By significant wave height
0 – 0.5 m	1,683	53	46	113	277	14.8	14.7	20.1	18.7	32.3	49.2
0.5 – 1 m	257,301	58	51	125	300	15.9	15.7	21.9	20.5	36.4	50.0
1 – 1.5 m	157,704	60	52	133	300	16.2	16.1	22.9	21.4	38.6	50.0
1.5 – 2 m	21,036	64	56	141	296	16.5	16.4	23.3	21.7	39.4	50.0
>2 m	5,411	69	58	164	298	16.8	16.8	24.0	22.4	40.7	50.0

05The wave climate behind the patterns

This section provides the environmental context for the rides logged in the dataset: what swells arrived during those sessions, and how ride performance varies with tide level, wave height, period, and direction.

wave rose by season — Figure 5.1Multi-modal swell roses by season. Top row: partition wave height (m); bottom row: partition peak period (s). Each session contributes one entry per active swell partition, typically 2–5 per session depending on the number of distinct components LOTUS resolves at the nearshore grid point. Plotting the partitions individually — rather than just the single peak direction (which is the direction of whichever partition happens to be the dominant one) — keeps the multi-modal directional distribution visible in both seasons.

scatter all — Figure 5.2Smoothed 2-D ride density of average speed, maximum speed, and ride length (columns) against tide level, significant wave height, peak period, and peak swell direction (rows). All rides combined; cells below the noise floor are masked. *Peak direction* here is the direction of whichever swell partition is dominant at the spot at the time — see Figure 5.1 for the multi-modal picture across all partitions.

scatter seasonal diff — Figure 5.3Same grid as Figure 5.2, but plotting the *seasonal preference* per cell rather than absolute count. Within each panel: summer rides and winter rides are first normalised to their own season totals, then subtracted. Red cells are over-represented in summer (April – September); blue cells are over-represented in winter (October – March). Cells with very little activity in either season are masked. Reading these panels directly answers the question "which kinds of waves are more characteristic of each season?".

Case Study · Oceanside, CA, USA

Not just protecting the beach, they're protecting the surf

[data-wf-bgvideo-fallback-img] { display: none; } @media (prefers-reduced-motion: reduce) { [data-wf-bgvideo-fallback-img] { position: absolute; z-index: -100; display: inline-block; height: 100%; width: 100%; object-fit: cover; } }

Oceanside has long been a staple Southern California surf hub—a place where generations have grown up chasing peaks off the pier and down the beach, trading stories in the lot, and shaping boards that end up under feet around the world.

Preserving Wave Quality

Surfline Coastal Intelligence helps communities and coastal engineers actively protect and improve wave quality, using data to inform decisions around dredging, reef construction, and shoreline projects.

These efforts aren’t just about limiting harm—they’re about making surf spots better and more sustainable for the long haul.

Case Study · Gold Coast, Australia

Protecting the Gold Coast

[data-wf-bgvideo-fallback-img] { display: none; } @media (prefers-reduced-motion: reduce) { [data-wf-bgvideo-fallback-img] { position: absolute; z-index: -100; display: inline-block; height: 100%; width: 100%; object-fit: cover; } }

The Gold Coast has long been a sacred ground for surfers. With iconic pointbreaks like Snapper Rocks, Kirra, and Burleigh Heads, this stretch of coast forged world champs, birthed core surf brands, and became a must-visit destination.

Supporting Coastal Safety

Beyond environmental monitoring, SCI also provides useful insights for coastal safety teams. By analyzing things like crowd density, wave activity, and coastal hazards, it helps agencies spot emerging risks and plan ahead.

The result is smarter resource deployment and safer, more accessible beaches for everyone. 

How it works

SCI’s superpower is its ability to turn cameras into sensors. Trained on hundreds of thousands of hours of coastal imagery, the system can detect small changes (like shifting shorelines or growing crowds) with remarkable accuracy, from a distance.

In the aggregate, these observations allow us to recognize patterns and create models to help forecast changes before they happen, giving communities a crucial head start.

Because it’s built to spot tiny objects across broad areas, a single Surfline camera can produce reliable coastal data that would otherwise require aircraft, LIDAR, or labor-intensive fieldwork.

Want to learn more?

Curious about how Surfline Coastal Intelligence can help protect your coastline and community? Explore our website, follow us on LinkedIn, or reach out to our team anytime at [email protected]

Our Commitments

Protecting Coastlines

and Surf Breaks

Protecting Our Beaches

ReportAssessing Surfline Sessions^™ Data at Oceanside, CA

01Executive summary

02Where the surfing happens

03When the surfing happens