The CoralNet carbonate production tool was first introduced in 2021 with updated rates launched in late 2024. In our recent publication “Quantifying coral reef carbonate budgets: a comparison between ReefBudget and CoralNet”, we validated CoralNet carbonate production rates using the initial (v1) and updated (v2) rates against the latest in-water ReefBudget methods.

Data was collected from Puerto Rico, Chagos Archipelago, and Indonesia along fixed transects using photoquadrats for analysis in CoralNet and using both in-water and image-based ReefBudget methodologies (see Figure 1).

Figure 1
Figure 1) Conceptual diagram of ReefBudget and image-based carbonate production surveys and data processing. (a) Illustrates the in-water ReefBudget survey approach, measuring all benthic components along the reef contour under a guide tape, followed by data input into ReefBudget worksheets. (b) Illustrates the two image-based workflows used in this study. Photoquadrats are taken in specified distances along a transect line. For the CoralNet method, the image is uploaded into CoralNet, annotated, and the carbonate production rates are exported directly into a CSV file. For the Coral Colony Rugosity Index (CCRI) method, the image is uploaded into JMicrovision to measure coral colony lengths along virtual transects, which are then converted from planar to contour lengths using the CCRI database (Husband et al. 2022), before being entered into ReefBudget worksheets. HCB: branching hard coral; HCE: encrusting hard coral; HCT: tabular hard coral; HCM: massive hard coral; CCA: crustose coralline algae

Comparing CoralNet with in-water ReefBudget estimates

For each survey area, carbonate production estimates along each transect in each site using both image and ReefBudget methods mapped well along a 1:1 line for both the original (v1) and updated (v2) CoralNet carbonate production rates. Differences in calcifier cover followed the same trends as carbonate production, suggesting that differences in production estimates are driven by the differences in how cover is estimated in each method (i.e. overhead area for images vs linear measurements including underhangs for in-water ReefBudget). At the Indo-Pacific sites, the v2 rates showed improved agreement with observed calcifier cover compared to v1 rates (see Figure 2).

Figure 2
Figure 2) Comparison of gross carbonate production estimates and total calcifier cover between methods. CoralNet versus ReefBudget gross carbonate production estimates (Gross G in kg CaCO3 m−2 yr−1) in the Western Atlantic (a-b) and the Indo-Pacific (d-e), using CoralNet calcification version 1 (a,d) and version 2 (b,e), as well as total calcifier cover in each region (c, f). Data points include 95% confidence intervals for both CoralNet and ReefBudget Coral G estimates. Black lines indicate the identity (1:1) line where values should fall if methods (ReefBudget and CoralNet) would yield exactly the same results. Coloured lines indicate major axis type II linear regressions between method estimates, with shaded areas and dotted lines showing 95% confidence intervals of models. The overlap between regression model uncertainties and the identity line indicate similar estimates between methods for both surveyed regions. Note that Western Atlantic plot axes (a,b) are at 1/10th scale of the Indo-Pacific plots

Comparing CoralNet with image-based ReefBudget approaches

In the Chagos Archipelago, CoralNet and image-based ReefBudget-CCRI estimates closely aligned with the 1:1 line for coral carbonate production, although variance between methods increased with overall production estimates. Greater relative ReefBudget-CCRI estimates came from transects dominated by tabular, massive, and encrusting coral morphologies, while CoralNet calcification v2 had higher relative coral carbonate production values in branching coral-dominated transects (see Figure 3).

Figure 3
Figure 3) Comparison of coral carbonate production estimates and method comparison metrics for the Chagos Archipelago. (a) CoralNet calcification v2 versus ReefBudget-CCRI coral carbonate production estimates (kg CaCO3 m−2 yr−1). Data points include 95% confidence intervals for both CoralNet calcification v2 and ReefBudget-CCRI Coral G estimates. The black line indicates the identity (1:1) line where values should fall if both image-based methods would yield exactly the same results. The coloured line indicates major axis type II linear regression between method estimates, with shaded areas and dotted lines showing 95% confidence intervals of the model. The overlap between regression model uncertainty and the identity line indicate good agreement between methods

Comparing the effect of regional and locally-derived rates of calcification

Implementing local calcification rates resulted in higher carbonate production estimates for Indonesia, as corals in this area had higher growth rates compared to average Indo-Pacific rates. In contrast, CoralNet production estimates using both locally-determined and regional calcification rates were nearly identical in the Chagos Archipelago. These results show that locally-determined carbonate production can be a major factor in generating estimates, particularly on reefs where local rates differ from regional estimates due to post-disturbance differences in coral demographics (see Figure 4).

Figure 4
Figure 4) Comparison of carbonate production estimates using local versus regional coral growth rates. (a) Estimates of gross carbonate production (Gross G in kg CaCO3 m−2 yr−1) in Indonesia, using locally-derived (y-axis) versus average regional growth rates (x-axis) with the ReefBudget method. (b) Estimates of coral carbonate production (Coral G in kg CaCO3 m−2 yr−1) in the Chagos Archipelago, using locally-derived (y-axis) versus average regional growth rates (x-axis) with the CoralNet calcification v2 method. Data points include 95% confidence intervals for both estimates. Black lines indicate the identity (1:1) line where values should fall if both methods would yield exactly the same results. Coloured lines indicate major axis type II linear regressions between estimates, with shaded areas and dotted lines showing 95% confidence intervals of the model. An overlap between regression model uncertainty and the identity line indicates similarity in estimates by each method

Considerations

This study shows that the area normalized calcification rates available in CoralNet can produce carbonate production estimates similar to in-water techniques like ReefBudget. This suggests that both methods can provide comparable carbonate production estimates, although some variation may occur depending on the dominant coral morphologies. CoralNet can allow rapid, reef-scale calcification estimates from photo or video imagery, often covering larger reef areas and be applied to historically collected benthic cover data to track carbonate production trends. We provide a further discussion of the pros and cons of using each method in the open access publication available here:

Sannassy Pilly S, Townsend JE, Alisa CA, Razak TB, Roche RC, Turner JR, Chan S, Kriegman DJ, Andersson AJ, Perry CT, Lange ID, Courtney TA. Quantifying coral reef carbonate budgets: a comparison between ReefBudget and CoralNet. Coral Reefs. 2025 Mar 17:1-5. https://doi.org/10.1007/s00338-025-02620-1