Abstract. In this study we propose a new formulation of subsurface water storage dynamics for use in rainfall–runoff models. Under the assumption of a strong relationship between storage and runoff, the temporal distribution of storage is considered to have the same shape as the distribution of observed recessions (measured as the difference between the log of runoff values). The mean subsurface storage is estimated as the storage at steady-state, where moisture input equals the mean annual runoff. An important contribution of the new formulation is that its parameters are derived directly from observed recession data and the mean annual runoff and hence estimated prior to calibration. Key principles guiding the evaluation of the new subsurface storage routine have been (a) to minimize the number of parameters to be estimated through the, often arbitrary fitting to optimize runoff predictions (calibration) and (b) maximize the range of testing conditions (i.e. large-sample hydrology). The new storage routine has been implemented in the already parameter parsimonious Distance Distribution Dynamics (DDD) model and tested for 73 catchments in Norway of varying size, mean elevations and landscape types. Runoff simulations for the 73 catchments from two model structures; DDD with calibrated subsurface storage and DDD with the new estimated subsurface storage were compared. No loss in precision of runoff simulations was found using the new estimated storage routine. For the 73 catchments, an average of the Nash–Sutcliffe Efficiency criterion of 0.68 was found using the new estimated storage routine compared with 0.66 using calibrated storage routine. The average Kling–Gupta Efficiency criterion was 0.69 and 0.70 for the new and old storage routine, respectively. Runoff recessions are more realistically modelled using the new approach since the root mean square error between the mean of observed and simulated recessions was reduced by almost 50 % using the new storage routine.

Spatial and temporal patterns of riverine woodlands in arid regions of Africa are poorly documented despite their considerable conservation value. We studied 1540 ha of riverine woodland in the lower Turkwel River floodplain, Kenya, between 1990 and 1998. Forty‐one woodland patches were mapped and their soil physical and chemical characteristics, tree species diversity, woody cover, tree density, wood volume and woodland regeneration were determined. The riverine woodland comprised nine vegetation types and a total of 14 woody species. Woodland patch mosaics were associated with microtopographical features and selected soil attributes. The most important woody species were Hyphaene compressa H. Wendl., Acacia tortilis (Forssk.) Hayne and Cadaba rotundifolia Forssk. The exotic Prosopis chilensis (Mol.) St. was invading parts of the riverine woodland. Overall, woody species diversity was low compared to similar riverine woodlands in East Africa. Tree density, wood volume and woody plant regeneration declined over the 8‐year study period, while woody cover was unchanged. Reduced tree density, wood volume and regeneration of woody species might be linked to changes in river flood patterns following the impoundment of the Turkwel Gorge Dam. It is suggested that spatially heterogeneous and temporally stochastic regeneration events, together with occasional tree mortality caused by channel abandonment, create the complex pattern of woodland patches in the lower Turkwel River floodplain. The mapped woodland patches may serve as monitoring units, which in future could reveal the interplay between changes in flooding patterns as a result of dam impoundment, anthropogenic disturbance and the well‐being of the riverine woodlands.

Abstract. In this study, we propose a new formulation of subsurface water storage dynamics for use in rainfallrunoff models. Under the assumption of a strong relationship between storage and runoff, the temporal distribution of catchment-scale storage is considered to have the same shape as the distribution of observed recessions (measured as the difference between the log of runoff values). The mean subsurface storage is estimated as the storage at steady state, where moisture input equals the mean annual runoff. An important contribution of the new formulation is that its parameters are derived directly from observed recession data and the mean annual runoff. The parameters are hence estimated prior to model calibration against runoff. The new storage routine is implemented in the parameter parsimonious distance distribution dynamics (DDD) model and has been tested for 73 catchments in Norway of varying size, mean elevation and landscape type. Runoff simulations for the 73 catchments from two model structures (DDD with calibrated subsurface storage and DDD with the new estimated subsurface storage) were compared. Little loss in precision of runoff simulations was found using the new estimated storage routine. For the 73 catchments, an average of the Nash-Sutcliffe efficiency criterion of 0.73 was obtained using the new estimated storage routine compared with 0.75 using calibrated storage routine. The average Kling-Gupta efficiency criterion was 0.80 and 0.81 for the new and old storage routine, respectively. Runoff recessions are more realistically modelled using the new approach since the root mean square error between the mean of observed and simulated recession characteristics was reduced by almost 50 % using the new storage routine. The parameters of the proposed storage routine are found to be significantly correlated to catchment characteristics, which is potentially useful for predictions in ungauged basins.

Use of long‐term herbivory studies in understanding the effects of livestock grazing on dwarf shrubs of arid zones of Africa is uncommon. Moreover, research has seldom focused on monitoring a 4–5 yr effect of herbivory at the level of individual plants. This study provided information on field‐based experiments and simplified statistical modeling to test compensatory growth responses of individuals of the African dwarf shrub Indigofera spinosa in northwestern Kenya. From August 1986 to January 1990, we simulated livestock grazing during dry and wet seasons and over the full year, whereby plants were defoliated during both wet and dry seasons. Individuals of I. spinosa (n = 480 plants) were subjected to one of five clipping intensities: unclipped control (0%), light (30%), moderate (50%), severe (70%), and very severe (90%) clipping; and defoliated of regrowth (i.e., new biomass). Our findings showed that rainfall, clipping regimes, and seasons of treatment influenced the compensatory growth response. Rainfall more than residual biomass influenced regrowth, while plants with greater residual biomass produced more regrowth than those with less. Optimum residual biomass was achieved under the 30% clipping level, while least was maintained under the 90% level. We separated compensatory growth response into under‐ and overcompensation. We showed that overcompensation occurred under some conditions but not in others. The shrub displayed relative overcompensation with a compensatory ratio (CR) > 1.0 for three continuous years with light clipping regime during the wet season defoliations (WSD). In two of five years there was overcompensation with the dry season defoliations (DSD), but undercompensation (CR < 1.0) with full‐year defoliations (FYD). On average, under the moderate, severe, and very severe clipping regimes the plants had undercompensation. Patterns of change of cumulative regrowth and its derivative, relative growth rates (RGR), provided different compensatory responses. RGR was more positive at lower cumulative regrowth but gradually declined and became negative when cumulative regrowth was maximum. The exception was in WSD where RGR remained positive for three years. DSD by far achieved greater cumulative regrowth than WSD and FYD. However, compared to the controls, FYD (except under light regime) exactly overcompensated for total “biomass budget” while WSD overcompensated under all clipping regimes except very severe. DSD under the light and moderate regimes overcompensated by +343.6% and by +202.7%, respectively. The study showed that I. spinosa combines tolerance with compensatory growth response to cope with a wide array of herbivory and seasons of use. The shrub may be grazed under light regime during the wet and dry seasons as opposed to the full‐year grazing which is unsustainable.