Functions of traditional ponds in altering sediment budgets in the hilly area of the Three Gorges Reservoir, China

Highlights

• UAV was used to evaluate the spatial distribution of ponds.

• Deposited sediments in the pond indicate SSY of a pond drainage catchment.

• Cumulative effects of ponds on watershed sediment balance were evaluated by the WaTEM/SEDEM model.

• Ubiquitous ponds reduce watershed sediment export.

• Dry farmland is main sediment source, while paddy field is main sediment sink.

Abstract

The landscape pattern will affect the sediment transport process. The cluster of ponds is a common landscape, which has traditionally been used for irrigation in the hilly area of the Three Gorges Reservoir (TGR). However, little is known about how the landscape elements temporally changed over the past decades and if the ponds can be applied to function in balancing watershed sediments against soil erosion. The Jinglingxi watershed, covering 20.5 km², was selected as the study area. The changes in pond number, surface area, and drainage catchment were analyzed with aid of high-resolution typographical map and unmanned aerial vehicles imagery. The spatial WaTEM/SEDEM model was developed to simulate watershed soil erosion and sediment deposition under the absence and presence of water bodies scenarios. Results from different simulation scenarios were compared and revealed the trapping effects of the multi-pond system. From 1983 to 2016, the number and total area of ponds roughly doubled. The density reached 30 ponds/km². From 1983 to 2016, the total drainage area of ponds increased from 13.22% to 35.4% of the whole watershed. The sediments deposited at the bottom of ponds can indicate the past specific sediment yield (SSY) in drainage catchments. Our results suggest that the multi-pond system not only reduce watershed sediment export but also alter the sediment deposition in different land uses. The reduced sediments export is expected to prolong the service life of downstream reservoirs at the expectancy of ponds' storage capacities. The ecological compensation from downstream reservoirs' revenues to upstream regions should be established to drive dredging actions for the upstream ponds.

Introduction

Small water bodies like ponds, small reservoirs and lakes are important human-made or natural landscapes, performing hydrological, biogeochemical and biological functions. These landscapes dominate the number of global water bodies. Ponds are small lentic waterbodies less than about 2–5 ha in area and may be permanent or seasonal, man-made or naturally created (Biggs et al., 2007, Biggs et al., 2017; Cereghino et al., 2008; Hill et al., 2018). Estimated by the global size distribution of global water bodies, ca. 5.47 × 10⁸ to 3.2 × 10⁹ ponds (smaller than 0.1 ha) are distributed throughout the world (Downing, 2010; Verpoorter et al., 2014; Holgerson and Raymond, 2016). Due to their small size (mostly smaller than 0.1 ha), these ponds are not included in national or provincial water bodies database. Even including the ponds, they are more likely to be underrepresented. In the United States, the National Wetlands Inventory (NWI) exclude the majority of ponds, since the minimum mapping size of NWI (0.4 to 1.2 ha) is too large (Burne and Lathrop, 2008). In Canada, the minimum mapping unit of wetland inventory program is approximately 1.0 ha (Milton et al., 2003). Yang and Lu (2014) firstly gave the distribution of water bodies > 0.36 ha in the mainland China. Similarly, the majority of ponds were excluded in the database. Mapping the number and size of ponds are the first step to understand and quantify their ecological functions.

Ponds retain sediment delivered to the river, thereby mitigating siltation of downstream reservoirs and rivers. Due to the dominant number of ponds, the trapping function for sediment might play a major role in global sediment budget. However, much attention has been focused on effects of large water bodies on sediment transporting processes (Vörösmarty et al., 2003; Walling, 2006; Beusen et al., 2005; Syvitski et al., 2005). Ponds are easily overlooked due to the small size and inconsequential influence conceived by the public and scientists. Renwick et al. (2005) examined the role of ponds on sediment budget in the United States and found that the ponds receiving about 21% of the total drainage area of the conterminous U.S. captured 25% of total sheet and rill erosion. Since their study was based on the database of underestimated ponds, the greater proportion of soil erosion was supposed to be retained by the ponds. Brainard and Fairchild (2012) analyzed sediment accumulation rates among ponds in southeastern Pennsylvania and northern Delaware. Berg et al. (2016) studied the changes in downstream reservoirs sedimentation to analyze the roles of upstream ponds on sediment dynamic in central Texas. However, these studies results were obtained by analyzing individual sediment cores and tend to be the individual impacts. Cumulative impacts of the ponds are not the simple addition of individual impacts and include interaction effects. The extension from individual impacts to cumulative impacts can be reached by the aid of the models.

The sediment yield is the integrated results of soil erosion, sediment transporting and sediment deposition. Numerous models are developed to simulate sediment yield at the outlet of the watershed, such as RUSLE, AnnAGNPS, SWAT, WaTEM/SDEM, WEEP and SWAT (De Vente et al., 2013). WaTEM/SDEM, a spatially distributed model, can simulate sediment deposition process. This model has been used at various spatial scales from several hectares (Quijano et al., 2016) to several million square kilometers basins (Borrelli et al., 2018a, Borrelli et al., 2018b). However, calibrating above-mentioned models is one of the hardest issues, which is labor-intensive, time-consuming and costly to acquire sediment yield dataset. In China, no authorities are assigned to carry out sediment observations especially for small watersheds. Sediments deposited in impoundments can be transformed to soil erosion rates for contributing drainage catchment. Recently, many studies retrieved changes in soil erosion process and calibrated the sediment yield modeling by sediments stored in reservoirs and check- dams (Quiñonero-Rubio et al., 2016; Fang, 2017; Zhao et al., 2017b; Pal et al., 2018).

The Three Gorges Reservoir area (TGRA) in China is mountainous and hilly dominated landscape with high population density. Ponds are ubiquitous landscape elements in the TGRA. Initially, ponds were constructed to store irrigation water for resistance to drought. These ponds have >100 years of history. Over the past decades, the number of ponds was increasingly kept to provide fishing and recreation services for local people. Though these ponds are intended for irrigation or aquaculture industry, they inevitably accumulate sediment to change sediment budget for the whole watershed. Little has been done in cataloging the number, size and drainage catchment of ponds, which already impede further understanding of their retaining functions in catching sediments.

In this study, the high-resolution imageries were acquired by the aid of unmanned aerial vehicles (UAV) platform. The ponds can be extracted on the basis of the imageries. Based on the data obtained, objectives of the study are: (1) revealing spatial distribution characteristics of ponds; (2) analyzing cumulative effects of the multi-pond system on watershed scale sediment balance.