Abstract. We present Top-kriging, or topological kriging, as a method for estimating streamflow-related variables in ungauged catchments. It takes both the area and the nested nature of catchments into account. The main appeal of the method is that it is a best linear unbiased estimator (BLUE) adapted for the case of stream networks without any additional assumptions. The concept is built on the work of Sauquet et al. (2000) and extends it in a number of ways. We test the method for the case of the specific 100-year flood for two Austrian regions. The method provides more plausible and, indeed, more accurate estimates than Ordinary Kriging. For the variable of interest, Top-kriging also provides estimates of the uncertainty. On the main stream the estimated uncertainties are smallest and they gradually increase as one moves towards the headwaters. The method as presented here is able to exploit the information contained in short records by accounting for the uncertainty of each gauge. We suggest that Top-kriging can be used for spatially interpolating a range of streamflow-related variables including mean annual discharge, flood characteristics, low flow characteristics, concentrations, turbidity and stream temperature.

No abstract

[1] We analyzed spatial and temporal variograms of precipitation, runoff, and groundwater levels in Austria to examine whether characteristic scales exist and, if so, how big they are. In time, precipitation and runoff are stationary with characteristic scales on the order of a day and a month, respectively, while groundwater levels are nonstationary. In space, precipitation is almost fractal, so no characteristic scales exist. Runoff is nonstationary but not a fractal as it exhibits a break in the variograms. An analysis of the variograms of catchment precipitation indicates that this break is due to aggregation effects imposed by the catchment area. A spatial variogram of hypothetical point runoff back calculated from runoff variograms of three catchment size classes using aggregation statistics (regularization) is almost stationary and exhibits shorter characteristic space scales than catchment runoff. Groundwater levels are nonstationary in space, exhibiting shorter-scale variability than precipitation and runoff, but are also not fractal as there is a break in the variogram. We suggest that the decrease of spatial characteristic scales from catchment precipitation to runoff and to groundwater is the result of a superposition of small-scale variability of catchment and aquifer properties on the rainfall forcing. For comparison, TDR soil moisture data from a comprehensive Australian data set were examined. These data suggest that in time, soil moisture is close to stationary with characteristic scales of the order of 2 weeks while in space soil moisture is nonstationary and close to fractal over the extent sampled. Space-time traces of characteristic scales fit well into a conceptual diagram of Blöschl and Sivapalan [1995]. The scaling exponents z in T $ L z (where T is time and L is space) are of the order of 0.5 for precipitation, 0.8 for runoff from small catchments, 1.2 for runoff from large catchments, 0.8 for groundwater levels, and 0.5 for soil moisture.

[1] This paper proposes a geostatistical method for estimating runoff time series in ungauged catchments. The method conceptualizes catchments as space-time filters and exploits the space-time correlations of runoff along the stream network topology. We hence term the method topological kriging or top kriging. It accounts for hydrodynamic and geomorphologic dispersion as well as routing and estimates runoff as a weighted average of the observed runoff in neighboring catchments. Top kriging is tested by cross validation on 10 years of hourly runoff data from 376 catchments in Austria and separately for a subset of these data, the Innviertel region. The median Nash-Sutcliffe efficiency of hourly runoff in the Innviertel region is 0.87 but decreases to 0.75 for the entire data set. For a subset of 208 catchments, the median efficiency of daily runoff estimated by top kriging is 0.87 as compared to 0.67 for estimates of a deterministic runoff model that uses regionalized model parameters. The much better performance of top kriging is because it avoids rainfall data errors and avoids the parameter identifiability issues of traditional runoff models. The analyses indicate that the kriging variance can be used for identifying catchments with potentially poor estimates. The Innviertel region is used to examine the kriging weights for nested and nonnested catchments and to compare various variants of top kriging. The spatial kriging variant generally performs better than the more complex spatiotemporal kriging and spatiotemporal cokriging variants. It is argued that top kriging may be preferable to deterministic runoff models for estimating runoff time series in ungauged catchments, provided stream gauge density is high and there is no need to account for causal rainfall-runoff processes. Potential applications include the estimation of flow duration curves in a region and near-real time mapping of runoff.