Quantifying the impacts of land use change on flooding in data-poor watersheds in El Salvador with community-based model calibration

Abstract

Urbanization can decrease the flood mitigation capacity of a catchment, and these impacts can be measured with hydrologic modeling. Models are typically calibrated against observed discharge and satellite data, but in a developing country context like El Salvador, these data are often unavailable. Even if a model is well calibrated and tested, its ability to influence land use plans requires additional stakeholder engagement. This study uses a participatory modeling approach to calibrate a watershed model and estimate flood impacts of land use scenarios in two urbanizing catchments in El Salvador with a linked land use-catchment hydrology-hydraulic model calibrated on flood height observed by community members. This paper explores both the value of household flood observation in model calibration and differences of flood extent estimates for land use scenarios with an uncalibrated versus community-calibrated model. We find that calibration using household surveys improves model performance. Results of scenario modeling suggest that while past urbanization has significantly increased household flood exposure in one catchment, future land use scenarios that further urbanize or reforest large areas of either catchment have little effect on the number of houses at risk for flooding. The success of the participatory methodology to increase model accuracy and link results to local land use planning makes clear the contribution of social science to traditional hydrological methods to understand land useflood links in data-poor catchments.

A systematic review of dynamics in climate risk and vulnerability assessments

“…For instance, there are numerous methodologies to predict demographic and land-use changes (Brown et al 2013), and these land-use maps can then be used in assessing the spatial configuration of exposure and vulnerability. There are also other options for obtaining data; for example, many have downscaled global or macro-regional socioeconomic scenarios (van Ruijven et al 2014, Viguié et al 2014), and data can also be gathered using participatory methods, surveys and interviews (Tellman et al 2015, Ordóñez and Duinker 2015, Djoudi et al 2013, Corobov et al 2013. Additionally, we suggest that further involvement of stakeholders, particularly local administrations, may increase the availability of socioeconomic and spatial data through co-production, benefiting both the accuracy of assessment results, and their usability by the local administration and decision-making.…”

Sediment load variability in response to climate and land use changes in a Carpathian catchment (Raba River, Poland)

“…Such changes may have contradictory impacts on the formation of runoff, and therefore for the sediment delivery from the cat chment. Concurrently, the forest expansion attenuates runoff (e.g., due to retention increase), while urbanization increases it (e.g., due to the increase of impermeable areas) (Tellman et al 2015). However, given the extent of the predicted changes (decrease of areas used for agriculture by 16%), the general outcome of the adopted land use scenario (LU) will result in a decrease of the sediment loads by approximately 0.16 Gg y −1 on average.…”

Assessment of the flood hazard arising from land use change in a forested catchment in northern Iran

“…This process combines both the severity and the probability of all relevant hazard loss scenarios” (Li, , p. 239). One of the challenges is to be able to assess the probabilities and magnitudes of future floods in places where relevant data (including community observations: Tellman, Saiers, & Cruz, ), resources, and expertise may be very limited (e.g., Bajracharya, Shrestha, & Shrestha, ). Several types of flood hazards may arise in different contexts, including coastal flooding, flooding due to unusually high groundwater, surface water flooding (rainwater accumulates faster than it can dissipate), channel capacity exceedance (including downstream floodplain inundation), and infrastructure failure (Morris et al, ), and also “flash floods” in tributary catchments.…”

“…HEC‐HMS contains a much wider range of modelling options than SWAT and many similar models, including an option to use the CN method for determining the direct storm runoff from a catchment. Its data requirements are much greater than SWAT, requiring careful optimisation and calibration with the possible risk of structural non‐identifiability (Shin et al, ), but it has been successfully used in a wide range of global contexts (e.g.,northwestern USA—McColl & Aggett, ; eastern China—Chen, Xu, & Yin, ; Kenya—Olang & Fürst, ; eastern India—Sanyal, Densmore, & Carbonneau, ; Greece—Papathanasiou, Makropoulos, & Mimikou, ; El Salvador—Tellman et al, ; northern Iran—Hajian et al, ).…”