**5. Conclusions**

Urban development has significant impacts on watershed sediment production in a developing country context. Management activities, especially the practice of filling gullies with poorly-consolidated materials, represent a persistent source of sediment in the watershed. Simulated total runoff correlated well with the observed data whereas simulated peak discharge was best predicted for medium-sized events. Simulated gully erosion contributed significantly to the total sediment load at the watershed scale (20% to 26%, annual average), even though most (40%–50%) of the total load was from sheet and rill erosion. Hotspots of erosion cover 23% of the total catchment area but generate 50% of the total sediment yield, and occur on steep slopes on highly erodible soils. This investigation highlights the necessity to implement management activities to mitigate soil erosion such as asphalt or other stabilization measures on unpaved roads, as well as other management activities (i.e., revegetation, sediment basins, channelization, etc). The scenario analysis showed that paving roads in the hotspots reduced total sediment production by 30%, but may increase peak discharge moderately (2%–21%) at the outlet. Any mitigation activity in the watershed that includes road paving needs to consider the potential impacts on downstream communities and channel erosion. Future studies for improving model calibration, and evaluating more mitigation scenarios are critical for proper sediment management in the LLCW and potentially in other urbanizing watersheds, particularly those in developing countries. Our maps of the spatial distribution of sediment yield are uncertain due to the coarse resolution of land use and soil properties for small sub-watersheds (AnnAGNPS cells), and possible overestimation of sheet and rill erosion on steep slopes. Future research should include more detailed spatial information on soil properties to improve parameters' estimates, which would improve the accuracy of model simulations under current conditions and in various management scenarios.

**Author Contributions:** Conceptualization, T.W.B., R.L.B., Y.Y., and N.G.-E. Methodology, T.W.B., R.L.B., E.J.L. Y.Y., E.V.T., T.K, and N.G.-E. Software, R.L.B. Validation, T.W.B. and N.G.-E. Formal analysis, T.W.B., R.L.B., and N.G.-E. Investigation, T.W.B. and N.G.-E. Resources, T.W.B., D.L., T.K., and Y.Y. Data curation, T.W.B., R.L.B., K.T.T.-Q., and N.G.-E. Writing—original draft preparation, N.G.-E.; Writing—review and editing, T.W.B., R.L.B., E.V.T., E.J.L., T.K., K.T.T.-Q., and Y.Y. Supervision, T.W.B., T.K., and Y.Y.; Project administration, T.W.B., D.L., and Y.Y. Funding acquisition, D.L. and Y.Y.

**Funding:** The Consejo Nacional de Ciencia y Tecnología (CONACyT, México; Grant/Award Number: 210925) and the US Environmental Protection Agency (EPA) (Interagency Agreement ID # DW-12-92390601-0) in collaboration with the US Department of Agriculture (USDA, Agreement # 58-6408-4-015), San Diego State University (USA), University of Córdoba (Spain), and the Centro de Investigación Científica y de Educación Superior de Ensenada (CICESE, Mexico) funded this study.

**Acknowledgments:** We thank Kraemer Stephen and Babendreier Justin from USEPA, and two other anonymous reviewers from *Water* for their review, technical editing, and valuable comments. Thanks to Chris Peregrin and Bronti Patterson for compiling data on sediment removal from the traps in the United States, and to Oscar Romo for initial field visits and allowing installation of a rain gauge at a field station. Special thanks to residents of Los Laureles Canyon, who provided valuable help for data collection.

**Conflicts of Interest:** The authors declare no conflict of interest. The views expressed in this paper are those of the authors and do not necessarily represent the views or policies of the U.S. Environmental Protection Agency (EPA). Mention of trade names or products does not convey, and should not be interpreted as conveying, official EPA approval, endorsement, or recommendation.
