# Published porous-clay-pipe field-study summary

**Primary source:** A. A. Siyal and T. H. Skaggs, “Measured and simulated soil wetting patterns under porous clay pipe sub-surface irrigation,” *Agricultural Water Management* 96 (2009), 893–904. DOI: 10.1016/j.agwat.2008.11.013. [USDA-hosted author manuscript](https://www.ars.usda.gov/arsuserfiles/20360500/pdf_pubs/P2266.pdf).

## Source locator and what is plotted

The accompanying CSV transcribes all four Table 1 columns: pressure head, mean error (ME), root mean square error (RMSE), and coefficient of determination (R²). The table is on printed page 900 (page 8 of the 12-page USDA-hosted PDF). The paper defines ME in Equation 5 on that PDF page as the mean of predicted minus observed water content; therefore negative ME means underprediction. RMSE and ME are reported in cm³/cm³; R² is dimensionless. The 2009 paper notes on printed page 901 (PDF page 9) that ME was negative for all four treatments. The original plots are new house-style renderings of these table values, not digitizations of the paper's moisture-profile figures.

The source statistics are transcribed, not recalculated: raw paired observations and the authors' HYDRUS output arrays were not used to recompute Table 1. The separately supplied wall-flow and daily-water-ledger models are hypothetical teaching calculations and are not validated by this field study.

The experiment used baked-clay porous pipes in loam at one field site in Tandojam, Pakistan. Soil samples were taken after five days, and the study compared the measurements with HYDRUS simulations. The test conditions and field are specific; the results do not show crop yield, water saved relative to a control, performance in other soils, or performance of an unspecified project pipe. R² is a comparison statistic, not an efficiency percentage; ME describes average signed bias and RMSE describes error magnitude under the authors' definitions.

## Relationship to the site's interactive examples

This published field comparison is separate from `wall-flow.csv` and the `bucket-days.csv` / `bucket-summary.csv` teaching calculations. Those use assigned hypothetical equations and bookkeeping rules; neither is calibrated to the field experiment. The diagram and chart do not connect their outputs.

## Exact coefficient definition

Equation 7 defines R² = 1 − Σ(Pᵢ − Oᵢ)² / Σ(Oᵢ − mean(O))², where P is predicted and O is observed soil-water content. This is not the squared Pearson correlation. We transcribe the reported values; we do not recompute them without the paired observations.

The concept considered on this page specifies porous mortar and a pressure chosen for soil and plant species. The cited experiment concerns baked clay. The teaching controls do not supply or test that soil/species pressure-selection law, and the experiment does not validate the mortar concept.