Professional Documents
Culture Documents
Hydrological Processes - 2024 - Zheng - Quantifying Intra and Inter Annual Dynamics of River Floodplain Connectivity and
Hydrological Processes - 2024 - Zheng - Quantifying Intra and Inter Annual Dynamics of River Floodplain Connectivity and
DOI: 10.1002/hyp.15137
RESEARCH ARTICLE
1
Ecohydrology and Biogeochemistry, Leibniz-
Institute of Freshwater Ecology and Inland Abstract
Fisheries, Berlin, Germany
We used imagery from remote sensing (FORCE Time Series Analysis submodule
2
Department of Geography, Humboldt-
Universität zu Berlin, Berlin, Germany
(combining Landsat and Sentinel-2 imagery) to derive spatially distributed times
3
Northern Rivers Institute, School of series (8 years) of NDWI data to infer patterns of floodplain inundation and river-
Geosciences, University of Aberdeen,
floodplain connectivity in two contrasting polders in the Lower Oder Valley National
Aberdeen, UK
4
National Park Lower Oder Valley, Schwedt/
Park. The upstream Polder A (14.4 km2) was extensively flooded for prolonged
Oder, Germany periods most winters. Wavelet analysis showed that this strong seasonality was pri-
5
Integrative Research Institute on marily driven by winter water levels in the river Oder that could enter and leave the
Transformations of Human-Environment
Systems (IRI THESys), Humboldt-Universität polder through two opened floodgates. Subsequent drainage was slow and aided by
zu Berlin, Berlin, Germany a pumping station. Inundation of the downstream Polder 10 (17.7km2) was lower and
Correspondence had less marked seasonality. This reflected the impact of flood attenuation by storage
Hanwu Zheng, Leibniz-Institute of Freshwater in Polder A upstream, but also the greater connectivity (via 10 floodgates) to the
Ecology and Inland Fisheries, Berlin, Germany.
Email: hanwu.zheng@igb-berlin.de Oder and a functional network of channels, which facilitated rapid drainage after
flood peaks. In Polder A, secondary periods of transient inundation could also occur
Funding information
Leibnitz Association Germany; China in response to local intense summer rainfall. Wavelet analysis also showed that
Scholarship Council; Einstein Stiftung Berlin; groundwater recharge in and around Polder A is primarily induced by floodwater,
Berlin University Alliance
whilst Polder 10 also reflects the influence of local rainfall-driven recharge. The flood
regimes of the two polders showed marked inter-annual variation, largely dependent
on flows from the upper Oder catchment. Understanding these patterns and pro-
cesses of inundation is important for both managing flows and sustaining valuable
wetland habitats within the National Park. Given projected climate change in eastern
Europe and possible management alterations to the flow regime of the Oder, the
potential implications for these habitats needs urgent attention.
KEYWORDS
floodplain, Oder river, remote sensing, river-floodplain connectivity, wavelet analysis, wetlands
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2024 The Authors. Hydrological Processes published by John Wiley & Sons Ltd.
1. What are the spatial patterns of flooding in the two polders in There is limited topographic variation in the LOVNP's flood-
terms of inundation area and flood storage? plain, and a network of raised embankments (dykes) subdivide the
2. What are inter- and intra-annual variations in the patterns and tim- region into polders and regulate the interactions between the flood-
ing of inundation and storage? plain and the Oder River (Figure 2a). The general landscape structure
3. What is the relative importance of the Oder River, local groundwa- in the LOVNP was formed at the end of the last ice age
ter dynamics and/or climatic factors in governing the polders' flood (12 000 year ago), with the floodplain fringed by elevated areas of
regimes? ground moraines (Ministerium für Landwirtschaft, 2023). The valley
floor is dominated by a mixture of sand and clay differentiating for-
mer channels and areas of overbank flows, with peat soils in the
2 | STUDY SITE North and moors formed along the floodplain edge (Ministerium für
Landwirtschaft, 2023).
The Oder originates in the Czech Republic and flows as the sixth larg- Until the LOVNP was established with increased emphasis on the
est river entering the Baltic Sea. The length of 854-km makes the conservation of the floodplain landscape since 1995, land in
Oder the second longest river in Poland, where 89% of its catchment the Lower Oder Valley was used for agricultural production
is located. The Lower Oder Valley National Park (LOVNP) is located (Ministerium für Landwirtschaft, 2023). The river-floodplain system
60 km NE of Berlin, Germany, and comprises a large polder system has been extensively engineered for over 100 years for agricultural
along 50 km of river length (Figure 1). drainage and flood protection, as well as improvement of navigation
The meteorological conditions are relatively constant across the in the Oder channel. As part of this process, embankments have been
2
whole LOVNP, which covers an area of 72 km . Mean annual precipi- built to create seven polders within the floodplain (Figure 1). Several
tation in the National Park is 561 mm/year (since 2005), and summer floodgates can regulate flow of the Oder River into the floodplain.
tends to be wettest with some high-intensity convectional events The annual discharge volume can reach 17.3 km3, while the station
reaching ≥40 mm/day (Deutsher Wetterdienst, 2023). Local average upstream of the LOVNP showed mean daily discharge variations
daily temperature (1990–2021) is 17.3 and 0.1 C in summer (June to between 132 and 1620 m3/s from 2015 to 2021 (Figure S2).
August) and winter (December to February), respectively (Deutsher About 60% and 20% of the area in the LOVNP are grassland and
Wetterdienst, 2023). Mean annual potential evapotranspiration (PET) woodland, respectively, with less than one third of grassland used for
is 643 mm/year (period 1980–2022). grazing. The remaining 20% consists of lakes, canals, and other open
F I G U R E 2 Characteristics of Polders
A and 10: (a) Elevation with locations for
measurements of precipitation, water
levels, groundwater levels and floodgates
(GW7, 8, 20–23 located nearby Polder
10); (b) Landuse; (c) Superficial geology;
(d) Soils.
water bodies (Ministerium für Landwirtschaft, 2023). This creates a (79.8% and 85.8%, in Polder A and 10, respectively) and some forest
rich diversity of wetland habitats reflecting subtle hydraulic gradients, areas (6.4%) are sporadically distributed in Polder A (Figure 2b).
sustaining rich biodiversity, which underpins conservation and eco- Stream and river bank deposits comprise the main surface geology in
tourism in the National Park. both polders, while a bog formed in the North-West of Polder 10, and
In this work, we selected two polders—covering different charac- moors and mires exist nearby the western border (Figure 2c). The soils
teristics of the polder systems (see more details below)—as main study reflect extensively distributed loamy clay in both polders, but they are
areas: Polder A and Polder 10, which cover 14.4 and 17.7 km2, more abundant in Polder 10 (Figure 2d).
respectively. Both are located on the west bank of the Oder River, The most important differences stem from the hydrology in both
that is, in Germany (Figure 1; Table 1). Both polders have a narrow, polders, as Polder A consists of small canals and large disconnected
elongated configuration, where the longer side runs parallel to the lakes, while extensive, well-connected channel systems characterize
Oder River, with 9.2 and 8.5 km in length, respectively. Polder 10. Also, whilst only two floodgates regulate the hydrological
The local topography is flat with elevations of Polder A and connectivity between Polder A and the Oder River, Polder 10 is more
10 varying from 1.6 to 7.8 and from 1.5 to 6.5 m.a.s.l, respectively open and regulated by 10 gates. These floodgates are usually opened
(Figure 2a). Both polders are predominantly covered by pasture in November or December and are closed in April (Table S1),
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZHENG ET AL. 5 of 16
controlling the direct connection between the polders and the Oder Access Hub archives. Preprocessing to Level 2 reflectance data ARD
River in winter but transient pluvial flooding is possible in high rainfall was based on the Framework for Operational Radiometric Correction
events outside winter. Additionally, there is a pumping system in Pol- for Environmental monitoring (FORCE), which is an all-in-one proces-
der A to pump water inside the polder out into the navigation channel sing engine for medium-resolution Earth Observation image archives,
that runs to the East of the floodplain. After pumping, mowing is enabling large area and time series applications and supporting the
conducted. integrated processing of Landsat and Sentinel-2 data (Frantz, 2019).
The preprocessing included corrections for atmospheric, topographic,
adjacency, and directional reflectance effects (Buchner et al., 2020;
3 | DATA AND METHODS Frantz et al., 2016; Roy et al., 2017), cloud and cloud shadow masking
(Frantz et al., 2018; Zhu et al., 2015; Zhu & Woodcock, 2012), and
3.1 | Hydroclimatic and GIS data spatial co-registration of Sentinel-2 to Landsat (Rufin et al., 2020). The
FORCE Time Series Analysis submodule was subsequently used to
Given the increased availability of remote sensing data at 10–30 m derive combined Landsat and Sentinel-2 time series of the NDWI
spatial resolution following the launch of Sentinel 2 and a correspond- (Normalized Difference Water Index) and the NDVI (Normalized Dif-
ing range of basic hydrometric parameters, which were monitored on ference Vegetation Index) with an interpolated equidistance of
the river and floodplain of the Oder during the period, we chose 16 days and a spatial resolution of 30 m. Interpolation was based on a
8 years for our study period (2015–2022). Daily precipitation and air RBF (Radial Basis Function) kernel following Schwieder et al. (2016).
temperature data were obtained from two weather stations of Ger- The NDWI was calculated based on green and the near-infrared
man Weather Service (Deutsher Wetterdienst, 2023), which are bands. This NDWI ranges from 1 to 1, with water pixels generally
located 2 and 16 km NW of Polder A, respectively (Figure 2a). Daily showing higher values than land pixels. The NDVI was calculated
river water level data were recorded by two stations near Polder A based on red and the near-infrared bands, and was used to quantify
and 10 (WL4 and WL8) from 2015 to 2022 (Deutsher the vegetation greenness inside the polders. The NDVI also ranges
Wetterdienst, 2023; IMGW-PIB, 2023). Discharge was only available from 1 and 1, with higher values denoting greater vegetation pro-
from two stations: D4 is at WL 4 (Figure 2a) and UD1 is located ductivity. An overview of data availability is provided in Table 2.
15 km upstream from WL8. Groundwater levels were monitored at
two stations the LOVNP (Figure 2a). The temporal resolutions range
from weekly to monthly (Brandenburg, 2023). PET was calculated 3.3 | Water area classification
using the FAO-56 Penman-Monteith method (Allen et al., 1998), with
meteorological data (i.e. net radiation, air temperature, wind speed, The workflow for classifying water inundated area was summarized in
actual vapour pressure, altitude) acquired from a DWD station, 16 km Figure S1. The total water inundation area of each imagery date was
west of Polder A (Deutsher Wetterdienst, 2023). GIS data (i.e. DEM,
TABLE 2 Overview of data used for each polder.
land use, geology, and soil type) were provided by the Landesamt für
Umwelt Brandenburg (Brandenburg, 2023). Polder A Polder 10
RS imagery 2015–2022/08 2015–2022/08
WL4 2015–2021/10
3.2 | Remotely sensed data WL8 2015–2021/10
GW15 2015–2022/08
We used all available Landsat and Sentinel-2 images acquired
GW22 2015–2022/08
between 01/01/2015 and 16/08/2022 with a cloud coverage of less
P3 2015–2022/08 2015–2022/08
than 75% for our analyses. Landsat and Sentinel-2 images were down-
PET 2015–2022/08 2015–2022/08
loaded as Level 1 products from the USGS and the Copernicus Open
T A B L E 1 Physiographic
Polder A Polder 10
characteristics of Polders A and 10.
Area km2 14.4 17.7
Topographic relief m.a.s.l 1.6 7.8 1.5 6.5
Land use Pasture % 79.80 85.79
Shrub 10.49 7.48
Forest 6.43 1.62
Water bodies 2.68 5.03
Urban settlements 0.57 0.08
Arable land 0.03 0
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 of 16 ZHENG ET AL.
calculated based on the number of classified water pixels multiplied and water pixels. Edge areas between water and land and unclear
by pixel size. The classification of water and land was achieved by the pixels were excluded (which are the mixed feature in a pixel). Overall,
Otsu method (an unsupervised classification method separating pixels accuracy (OA), user accuracy (UA), producer accuracy (PA) and kappa
into two classes and determining the threshold by minimizing coefficient (K) were calculated to evaluate the performance of classifi-
intra-class variance) based on NDWI (Otsu, 1979). A threshold was cation (see equations in supplementary material). OA (SEq1) repre-
calculated using this method, and pixels with values higher than this sents the percentage of total correctly classified pixels (including
threshold were classified as “water”. Different proportions of water water and land). PA (SEq 2) represents the percentage of the correctly
and land will affect the Otsu-based threshold: The water area esti- classified feature (water) in the reference sites. UA (SEq3) is the per-
mates are strongly affected when a large number of pixels are close centage of the correctly classified feature (water) in the classified
the threshold value. An accurate threshold detection usually requires sites. K (SEq 4; SEq 5) shows the classification performance compared
that the number of pixels in each class reaches at least 10% of total with random classification, and ranges from 1 to 1. As closer the
number of pixels (Bazi et al., 2007). values are to 1, the better the performance.
In order to evaluate the uncertainty of water area estimation in
relation to the Otsu-based threshold, we created virtual maps for each
original imagery by reorganizing the proportions of the number of 3.5 | Wavelet analysis
pixels in water and land classes. Each map satisfied the accurate
detection requirements. The water inundated area estimates of the Continuous wavelet transformation (CWT) was used to identify tem-
original imagery determined by the new Otsu-based threshold from poral dynamics in terms of periodic signals of different hydrological
these virtual maps were regarded as the “uncertainty” of the original variables (i.e., water level, water inundation area, groundwater level
imagery. and precipitation). The Morlet wavelet function (Equation 1) was
The detailed steps were as follows: For each original imagery, the applied to calculate the wavelet coefficient (Equation 2) by convolu-
initial water and land classes were determined by the Otsu method. tion between the time series data (xn ) (Labat, 2005):
We then created virtual maps by randomly selecting the pixels from
rffiffiffiffi N
each initial class according to a prefixed proportion ensuring that each δt X 0 δt
W Xn ðsÞ ¼ xn Ψ0 ðn0 nÞ ð1Þ
virtual map had the same number of pixels as the original imagery. We s n0 ¼1 s
created two virtual maps with 20% and 80% of water proportion,
Ψ0 ðηÞ ¼ π 4 eiω0 η e2η ð2Þ
1 1 2
which indicated the upper and lower limits for the final water inun-
dated area uncertainty (20% water proportion means small numbers
of pixels with high values exist in the virtual map, representing a low
Otsu-based threshold, and more pixels in the original imagery were where ω0 and η are the dimensionless parameters frequency and time,
classified as water class based on this threshold). The Otsu method respectively, δt is time step and s is the wavelet scale.
was then used again to determine the new thresholds of the two vir- The wavelet power value was given by the square of the wavelet
2
tual maps. These two new thresholds were used in the original imag- coefficient (W Xn ðsÞ ), which is an indicator of how the energy of a sig-
ery to determine the higher and lower limits of water inundated area nal is distributed in a time-frequency domain (Labat, 2005), and was
estimates indicating the uncertainty of these estimates. shown as 2D time-frequency plot in this study. Importantly, the CWT
Water storage of each Polder was calculated according to classi- will be influenced by missing information at the start or end of limited
fied water pixels and the DEM map. The inundated water level was time series; therefore, the cone of influence (COI) zone, which is the
assumed as the level, of which the elevation of 80% of classified influenced region, was discarded to avoid such edge effects
water pixels are below. The water storage of each classified water (Torrence & Compo, 1998).
pixels was calculated by multiplying pixel area (30 30) and level dif- Wavelet coherency (WTC) was calculated via CWT to investigate
ference between inundated water level and pixel elevation. the relationship between timeseries of water inundated area esti-
mates and other hydrological variables, that is, water level in Oder,
groundwater level and precipitation:
3.4 | Validation and assessment
1 XY 2
So s W ðsÞ
R2n ðsÞ ¼ X n Y ð3Þ
In order to validate the estimated water inundation area, visual inter- So s W n ðsÞ So s1 W n ðsÞ
1
statistically significant. This level was determined as 95% confidence variability and high intra-annual oscillations. The peaks and troughs in
levels assuming the time series followed a lag-1 autoregressive pro- groundwater and water levels in the Oder were in general similar, with
cess and a power value with a chi-square distribution (Torrence & a slight time lag, while being inversely out of phase with PET in terms
Compo, 1998). The Monte Carlo method was used to evaluate the of timing of troughs and peaks.
95% confidence level for WTC (Grinsted et al., 2004). Wavelet analy- Overall, the hydrological variables from stations nearby Polder A
sis was realized with the Python package “PyCWT” (Krieger (i.e., WL8 and GW15), exhibited strong connections throughout the
et al., 2023). The hydrological variables used in WTC were resampled period, as their peaks and troughs were very close in time. Peaks of
to the same temporal resolution as processed imagery. Precipitation GW15 appeared prior to WL8 in dry years (e.g., 2016, 2019, and
was resampled by accumulating the previous 16 days daily values, 2020), while this time lag decreased in wet years (e.g., 2017, 2018).
water levels were averaged by a 16 days neighbourhood range, and Connections between GW22 and WL4 (Polder 10) were less obvious,
groundwater levels were linearly interpolated. though they showed higher intra-annual oscillations.
T A B L E 3 Accuracy assessment of
5 December 2018 1 February 2018 30 September 2017 24 March 2017
water inundation area by NDWI in Polder
PdA Pd10 PdA Pd10 PdA Pd10 PdA Pd10 A. Kappa coefficient evaluates how well
the classification performed compared
OA (%) 85.7 84.6 84.8 91.5 91.1 87.3 86.4 85.2
with randomly sampling. “CA” means the
Kappa 0.75 0.73 0.74 0.84 0.84 0.77 0.76 0.74 result of our classification scheme. “PdA”
PA (%) 71.4 69.5 94.4 83.5 82.1 74.6 85.2 70.4 and “Pd10” represent Polder A and
UA (%) 100 99.6 79.2 99.4 100 100 87.3 100 Polder 10, respectively (The processing
work flow of accuracy assessment is
CA (%) 2.05 2.37 52.2 11.8 2.25 2.34 44.7 3.64
shown in Figure S2).
F I G U R E 6 Water storage (a, c) and percentage of time water storage is exceeded or equaled (b, d) in (a, b) Polder A and (c, d) Polder 10. The
average daily water volume discharged from the Oder to the Baltic sea is about 496 106 m3/day.
were found in WL4 compared to WL8, indicating that WL4 experi- were more pronounced than in Polder A. This corresponded to the
enced more frequent short-term flooding. Considering the timing more short-term fluctuations in inundation of Polder 10.
when such signals appeared, significant periodicities and variabilities For the groundwater stations near the polders, significant
mainly occurred in winter for both WL stations (2 to 64 days). 1-year cycles persisted in GW15, with only local high power values in
For the estimated water inundation area of Polder A, annual GW22 (Figure 7e,f). In other words, clear seasonality was apparent
(1-year) cycle signals were present during most of the study, reflecting in GW15, but not in GW22 (see also Figure 3). However, in the high
the strong seasonality (Figure 7c,d). In the half-year frequency frequency domain, higher periodicities occurred in GW22. These
domain, the power values were greater than the higher monthly fluctuations were observed both in summer and winter in
frequency domain (i.e., monthly or weekly), as sometimes two peaks GW22, while they were mostly concentrated in summer for GW15.
in summer or winter occurred. Particularly in 2015–2017, obvious Overall, the power values were more evenly distributed along the
periodicities from 1 to 4 months (i.e., high frequent fluctuations) entire frequency spectrum in GW22 and the oscillations in the time
occurred. series were more evenly distributed at the entire frequency scale,
In Polder 10, noticeable periodicities for all frequencies rather than showing a distinct seasonality as in GW15.
(i.e., weekly, monthly, or yearly fluctuations) only occurred in winter For the climate variables, significant 1-year cycles were persistent
during high inundation and the higher frequencies (weekly, monthly) in the power spectrum of PET reflecting expected strong seasonality,
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 of 16 ZHENG ET AL.
F I G U R E 7 Wavelet power spectrum (with wavelet basis of Morlet) of hydrological variables. (a, b) Water level in the Oder River at WL8 and
WL4 (daily data); (c, d) Water inundation area in Polder A and Polder 10 (16-days data); (e, f) Groundwater level at GW15 and GW22 (16-days
data); (g, h) Precipitation at P3 and PET (daily data). A higher value in the power spectrum indicates a greater likelihood of the presence of a
periodic signal with a corresponding period and time. The black circles in the heat map show the significant regions (95% confidence level). The
crossed-out shaded areas were not considered because of edge effects.
though this was only evident for precipitation in 2017 (Figure 7g,h). frequency scale, the phase differences between water inundation
These cycles were manifested in seasonal peaks and troughs at sum- area and river water or groundwater level were small (small time
mer and winter, respectively. Both precipitation and PET showed high lags). However, the greater phase difference for groundwater means
frequency cycles in summer, but the periodic signals were more regu- that signals between water level and inundation were transferred
lar in PET. At high frequencies, daily and weekly fluctuations were faster (Figure 8a,b).
observed both in PET and precipitation, with monthly fluctuations Between water inundation area and precipitation for the 1-year
being only common in precipitation. Overall, power values were more period, a large phase difference was apparent (Figure 8c). Precipitation
evenly distributed along the entire frequency scale in precipitation, also had an anti-phase (time lag of half a year) trend.
with no fluctuations of mid-range frequency occurring in PET. In the higher frequency spectrum of Polder A, coherency ridges
only sporadically occurred in periods from 1-month to 8-months for
water level and precipitation, and from 2-months to 4-months
4.4 | Coherence of wavelets in inundation extent for groundwater level. This reflects that the inundation fluctuations
and controlling factors are restricted to these shorter periods showing dependence on river
water level, groundwater level, precipitation at different time and
Statistically significant coherency between water inundation area in scale positions. Precipitation showed comparatively high coherency
Polder A and all other variables in 1-year cycles occurred throughout values in periods from 4-months to 8-months (Figure 8c), which sug-
the study period (Figure 8a–c), reflecting the strong co-varying char- gests a potential contribution for such intermediate periods, though
acter between seasonal variations. Water level showed the highest this was not statistically significant. Further, water level and
coherencies in 1-year periods, with the strongest co-varying proper- groundwater showed unstable phase differences (Figure 8a–c). These
ties with the inundation area in Polder A (Figure 8a). In this instabilities reflect high uncertainty at high frequencies, as significant
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZHENG ET AL. 11 of 16
F I G U R E 8 Wavelet coherency of water inundation area in Polder A (a, b, c) and Polder 10 (d, e, f) with other hydrological variables: (a, d)
water level in Oder river; (b, e) groundwater level; (c, f) precipitation at P3. The black circles in the heat map show the significant regions (95%
confidence level). The crossed-out shaded areas were not considered because of edge effects. The arrows indicate the local phase difference
between two variables, and the phase value was determined by the angle between the arrow and positive x-axis. Significant regions reflect
similarity of the fluctuation patterns between two variables and higher coherency values mean stronger relationships. The negative phase means
the response of the inundation area is behind the other variable and the exact time lag is calculated by (phase difference corresponding
period)/2π.
co-varying parts could be captured, but how long the time lag was 5 | DI SCU SSION
between the variables often remains unclear.
In the 1-year period for Polder 10, only precipitation showed sig- 5.1 | Spatial–temporal dynamics of floodplain
nificant coherency with water inundation area (i.e., a strong inter- inundation
annual co-varying trend, Figure 8f), which was similar to Polder
A. However, almost all other domains of the frequency spectrum In this study, remote sensing products based on NDWI and NDVI
showed low power values (Figure 8c,f). The phase difference for pre- were integrated with empirical data and analysed using wavelet analy-
cipitation was similar to Polder A with an asynchronous appearance, sis to better characterize and understand the spatial–temporal pat-
an anti-phase trend for precipitation (Figure 8c,f). The phase differ- terns of river-floodplain inundation in contrasting polders in the
ence between water level and inundations was also similar in both LOVNP along the river Oder in Germany. Previous studies have also
polders, however, the seasonal variations were not as relevant in Pol- used topography and DEMs in remote sensing-based analysis to esti-
der 10, while groundwater showed a more unstable time lag across mate flood inundation (Liu et al., 2020b; Munasinghe et al., 2018;
the study period. Rosser et al., 2017): These typically show patterns of inundation being
In the higher frequency spectrum (1–8 months), significant more likely distributed adjacent to channels and low-lying areas, with
coherency occurred for water level in Polder 10, indicating stronger connectivity between flooded areas increasing with increasing water
river-floodplain links, while groundwater is less connected to the inun- levels. However, in our work in the Oder floodplain, the very different
dation in this frequency spectrum (Figure 8d,e). The phase differences inundation patterns between the two polders studied were mainly
at higher frequency for water and groundwater levels was unstable related to differences in their floodplain structure and management.
along the time scale for both polders. These results again reflect the The winter inundation and water storage were generally much
uncertain linkages at high frequencies, though Polder 10 showed greater in extent and more prolonged in Polder A than Polder
more stochastic behaviour. 10, which being upstream had a first order influence on flood
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 of 16 ZHENG ET AL.
attenuation of the main annual flow peaks. During periods of high area classification and the use of the informal uncertainty indicators
winter inundation, the flooded areas corresponded well with the provided qualitative insight into potential errors in estimated inun-
topographic depressions in the DEM. However, in Polder 10, the dated states.
upstream flood storage, together with the well-connected channel Our approach based on indices is straightforward and consistent
system and high number of floodgates, contributed to a strong river- through time and overall showed good results. However, to account
floodplain connectivity and thus, the polder was less extensively inun- for observed limitations, such as the inability to detect for small water
dated as it drained more rapidly. Czuba et al. (2019) similarly found proportions in a pixel or water hidden by vegetation, spectral unmix-
that channel orientation inside a floodplain plays an important role in ing techniques or the additional use of Synthetic Aperture Radar
river-floodplain exchange in their modelling results. (SAR) data would be helpful. For example, wetland inundation classifi-
Although summer flooding was occasionally evident in Polder A, cation by SAR imagery were examined by using multi-temporal
it was rarely detected in Polder 10, which has greater coverage of per- RADARSAT-2 C-band SAR intensity and showed good performance
manent open water bodies and well-connected channels. Usually, (Canisius et al., 2019). Similarly, spectral unmixing techniques for eval-
summer flooding is pluvial in origin as the flood gates are closed, limit- uating water bodies in sub-Landsat pixels were also applied in eight
ing river-floodplain connectivity. Thus, in part, these differences imageries from southern Arizona, USA by Sall et al. (2021), and
reflect the more extensive dry areas in Polder A that are prone to showed accurate result compared with digitized surface water area.
flooding. In addition, inundation in the polders likely reflected meth-
odological uncertainties associated with high vegetation surrounding
the channels. Consequently, some inundation was probably unde- 5.2 | Disentangling the influence of river
tected due to mixed pixels of water and land. Such mixed pixels were connectivity, groundwater and climatic variability on
also proportionally more important during times of low inundation. floodplain inundation with wavelets
Similar vegetation effects on classification have also identified in other
inundation estimations, especially in wetland areas (Leblanc Due to the physical structure of floodplains and their interactions with
et al., 2011). local and catchment-scale drivers, river-floodplain systems are com-
The wavelet analysis allowed a more systematic assessment of plex hydrological interfaces (Alaghmand et al., 2014; Li et al., 2019).
the contrasting inundation patterns of the two polders, as well as a Continuous wavelet transformation allowed us to extract both the
basis for comparing intra- and inter-annual variability. The 1-year peri- frequency and temporal position of coherent signals in concurrent
odicity (i.e., seasonality) was strongest in Polder A reflecting the timeseries, allowing more insight than, for example, Fourier transfor-
annual pattern of floodplain inundation in response to high water mation (Cheng et al., 2021). Via wavelet coherency, the co-varying
levels in the Oder. In contrast, in Polder 10 monthly-scaled fluctua- tendency and phase difference of two timeseries could be used to
tions were more pronounced, reflecting the upstream floodplain stor- indicate their relationship in frequency-time scale (Cheng et al., 2021).
age in Polder A and the subsequent pulses of higher connectivity in Therefore, wavelet analysis has been widely used in hydrology for
response to increases in river levels in winter. Similar temporal varia- decomposition, de-noising, complexity description and hydrological
tions in floodplain inundation have been shown to reflect dominant forecasting (Sang, 2013).
hydrological processes in river-floodplain systems and differ in con- Even so, overlapping signals meant that it was difficult to fully dis-
trasting geographical regions (Scott et al., 2019). For example, strong entangle the contributions of floodplain structure, river flow regime
inter-annual variability of maximum inundation extent has also been and other drivers on inundation patterns. In general, the analysis
identified in floodplains in South American (Hamilton et al., 2002) showed that the dominant influences of river levels in the Oder and
where the duration of the inundation period was linearly related to local rainfall on inundation was, respectively, pronounced in winter
the maximum extend of the inundation area. and summer, and stronger in Polder A. River-floodplain connectivity
These inferences were derived from analysis for the RS imagery varied predominantly at an inter-annual scale in both polders, but Pol-
and highlight some of the potential as well as limitations of such der 10 also showed more pronounced connectivity at the monthly
approaches. Detection of inundation was extracted from RS optical scale.
imagery; a technique which has been widely applied and validated for In the wavelet coherency spectrum, the limited monthly co-
larger scale assessments of soil moisture and vegetation dynamics varying links during winter between the Oder and Polder A suggest
(Huang et al., 2018). RS provides global long-term and spatially exten- that the rapid fluctuations of water levels in the Oder do not corre-
sive distributed earth surface observations, which is particularly useful spond to varying inundation in Polder A. In contrast, the numerous
in places where ground observations are difficult (Schmugge monthly co-varying links between the Oder and Polder 10 highlight
et al., 2002). Due to the variations in clear sky observations resulting the importance of the polders structure with the more limited inunda-
from cloud coverage, we combined observations from multiple satel- tion in the well-connected and freely-opened Polder 10 responding
lite sensors and employed an interpolation algorithm to achieve tem- rapidly to water level changes in the river and channels.
poral continuity of the data series as in Frantz (2019). From direct Local climatic variables, that is, snowmelt, rainfall intensity, have
comparison with time series of aerial photographs, we found accept- been shown to play an important role in the temporal variation of
able accuracies in our validation of the Otsu-based water inundated floodplain inundation (Scott et al., 2019). Effects due to wetter climate
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZHENG ET AL. 13 of 16
conditions and disruption by flow regulation on river-floodplain con- Further limitations in the wavelet analysis stem from the quality
nectivity are amongst such influences identified by others (e.g., Karim of the estimated inundation timeseries derived from RS. Errors here
et al., 2015; Ward & Stanford, 1995). For Polder A, the apparent can affect the assessment of where significant regions in the wavelet
inter-annual linkages between rainfall and inundation are suggested spectrum appear, and the inconsistent duration and gaps of the time-
by the high coherency values, but the large time lag between these series confined the analysis to a limited time-frequency region
two time series are physically implausible (i.e., local summer rainfall (Torrence & Compo, 1998). The high but insignificant values in the
cannot affect winter inundation directly). Hence, this apparent linkage wavelet spectrum may suggest some potential periodicities, but they
is just an artefact of the mathematical similarity between rainfall and may also be the result of intermodulation or subharmonics and may
inundation inter-annually, with no physical connection. The monthly- not have a physical meaning (Cheng et al., 2021).
scale co-varying links in Polder A in summer highlight the short-term
nature of pluvial flooding there. Although the linkages between rain-
fall and inundation are not frequent in our analysis, undoubtedly ante- 5.3 | Wider implications for floodplain
cedent rainfall reduced floodplain's storage capacity, affecting the management
water residence time inside polders (Tull et al., 2022), and subse-
quently reducing the time lag between water level in the Oder river Better understanding the controls of the spatial extent and longevity
and inundation in the polders (Tull et al., 2022). Overall, the limited of the two Polders has important process-based implications for man-
contribution of local rainfall to inundation in both polders during win- agement of land and water in the National Park. This is particularly
ter is reflected by its low coherency values in the high-frequency the case for the future as climate change in eastern Europe and
domain. changing management of the Oder upstream in Poland may affect
Subsurface water storage in floodplain soils and groundwater can future patterns of inundation (Anders et al., 2014). Whilst the main
act as buffers to precipitation or input signals (Liu, Yu, et al., 2020), driver of annual floodplain inundation is the flow regime of the Oder
and could be recharged by different sources. For example, groundwa- river, operation of the floodgates in the polders contributed to quite
ter levels showed less periodic signals than precipitation or water different flood responses in Polder A and Polder 10. Moreover, the
levels in the Oder river, and the lagged response of groundwater to pumping in Polder A had an important influence on the recession of
inundation in Polder A suggests that groundwater was mainly flooding, whereas Polder 10 drained along pre-existing channels. Con-
recharged by flood water. In contrast, weaker linkages in Polder sequently, there are many possibilities for floodplain management that
10 may be evidence that the groundwater was also recharged by local could be used to develop resilient strategies in the face of future cli-
rainfall as well as flood water. mate and environmental change.
Polder A showed a greater low-pass filtering effect for signals In this regard, the analysis presented here provides a framework
transferring from precipitation or river water to inundation, compared for better conceptual understanding of the characteristics and causes
with Polder 10, with high-frequency signals rarely apparent in the of inundation and flood storage that have two obvious uses. Firstly,
inundation of Polder A. The differences of the water levels between such spatial–temporal information has potential utility in providing
WL8 (upstream) and WL4 (downstream) also show the polder's buff- readily communicable information for various stakeholders
ering effects of flood storage as the mid-frequency signals (1– (e.g., regulatory agencies, land owners, farmers, conservationists, etc.)
4 months) in the water level of the Oder at WL8 were not detected at as an evidence base that could be used for discussing potential man-
WL4, although high frequency signals (<1 month) were strengthened agement changes in the face of climate and other threats (e.g., Smith
in winter. It is likely that the confluence of the navigation canals in the et al., 2021; Smith et al., 2022). Secondly, and linked to the first point,
west of the Oder river help transmit some high frequency distur- this conceptual understanding is an essential prerequisite for numeri-
bances downstream. cal model development that could characterize the water balance of
Again, whilst the wavelet analysis proved useful; capturing and river-floodplain dynamics in a quantitative manner (Tull et al., 2022).
localizing multiple periodicities as well as illustrating the oscillations Again, such quantification, with acknowledgement of associated
in various temporal scale, limitations were also apparent. For exam- uncertainty is important evidence in option appraisal.
ple, the identification of relationships between two variables is From a conservation perspective the variability of inundation
highly mathematical and sometimes can produce artefacts that lack a across the floodplain creates a rich mosaic of wetland habitats that
physical basis (Sang, 2013). This can lead to misleading results such sustains a high degree of biodiversity. And, although the relationship
as the high coherency values depicting the inter-annual link between between flooding duration and fish production has been globally
precipitation and inundation in both polders. In addition, the wavelet documented (Arthington & Balcombe, 2011), for many wetland spe-
analysis was also limited by data availability and some of the locally cies and communities the habitat requirements in terms of timing,
high—but statistically insignificant—relationships may have become extent and longevity of saturation are not well-known (Drayer &
clearer with finer temporal resolutions in the time series. Despite Richter, 2016). Polder A and 10 have different land use distributions
this, the wavelet analysis allowed us to detect the multi-temporal and contrasting flora or fauna assemblages due to their distinct inun-
scale and non-stationary characteristics of many hydrological dation variability. For example, the high hydrological heterogeneity
variables. resulting from the short-term inundation in Polder 10 may offer
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 of 16 ZHENG ET AL.
_
greater refuge opportunities for species at risk of flooding (Zmihorski Weaker linkages in Polder 10 (between groundwater and inundation)
et al., 2016). In summer, Polder 10, with its widely connected chan- are consistent with a greater influence of local rainfall-driven recharge.
nels, provides better aquatic habitat compared to the more isolated The well-connected Polder 10 (with 10 flood gates) responds more rap-
ponds in the drier Polder A, and overall, such high physical habitat idly to water input signals (i.e., water level in Oder river, precipitation),
diversity and hydrological connectivity contributes to the survival of as evidenced by greater coherence between inundation patterns and
aquatic organisms (Arthington & Balcombe, 2011). In addition, it is other hydrological variables.
important to note that the closed floodgates in summer disconnect RS techniques provided direct evidence of how wetland habitat
the river-floodplain system, and subsequently, water quality gradually varied over extensive areas and wavelet analysis proved useful in dis-
becomes the dominant factor contributing to the habitat quality of entangling the linkages between inundation dynamics and potential
aquatic organisms (Arthington et al., 2010). Thus, assessing the water driving variables in this complex river-floodplain system. Using these
quality regimes of the floodplain wetlands of the Oder is a major pri- techniques in an integrated way, this study provided conceptual
ority for future work. understanding of how wetland habitat was influenced by hydrological
As most of the polders are managed for some sort of farming, processes that are currently highly controlled in managed polder sys-
other management practices include patterns of mowing and grazing tems. As well as providing a framework for future quantitative model-
on grasslands, which can affect wetland habitats. For example, differ- ling, the study provides an evidence base for relevant stakeholders to
ent wetland bird species have different vegetation requirements, with assess whether future changes in management are needed to main-
short-term flooding in Polder 10 and unmowed areas appearing to tain habitats in the face of projected climate change in eastern Europe
benefit birds nesting and foraging (Åhlén et al., 2023). Similarly, areas and flow regulation upstream in the Oder catchment.
of forestry are likely to have higher evapotranspiration and may create
more buffering against summer floods by contributing to soil moisture AC KNOW LEDG EME NT S
deficits. Likewise, pumping operations can have a strong influence on Hanwu Zheng is funded by the Chinese Scholarship Council (CSC).
inundation variability, thus playing an important role for manipulating Tetzlaff's and Okujeni's contribution was partly funded through the
habitats to the benefit of particular species in the floodplains. Whilst Einstein Research Unit “Climate and Water under Change” from
these anthropogenic management practices, (i.e., floodgate operation, the Einstein Foundation Berlin and Berlin University Alliance (grant
pumping system operation and land use) can be adjusted to maintain no. ERU-2020- 609). Contributions from Soulsby are supported by
certain habitat conditions (Gumiero et al., 2013), conceptual and the Leibnitz Association Germany in the project Wetland restoration
quantitative understanding of inundation dynamics and their controls in peatlands.
as provided in this study will be essential to evidence-based decision
making. CONFLIC T OF INTER E ST STATEMENT
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
6 | C O N CL U S I O N S ence the work reported in this paper.
Our study investigated the inundation area in two polders in the Lower DATA AVAILABILITY STAT EMEN T
Oder Valley National Park by classification of NDWI imagery from The data that support the findings of this study are available from the
FORCE Time Series Analysis submodule (combined Landsat and corresponding author upon reasonable request.
Sentinel-2 images) and employed wavelet analysis to investigate the
spatial–temporal variation of wetland inundation and associated con- OR CID
trolling factors. We further analysed spatial differences of inundation Hanwu Zheng https://orcid.org/0009-0007-8931-5226
patterns using a DEM in relation to management practices and their Chris Soulsby https://orcid.org/0000-0001-6910-2118
effects on wetland habitat. Overall, seasonality was well-captured in
the wavelet power spectrum of most variables, showing marked RE FE RE NCE S
inter-annual variation of wetland inundation, though the influence was Åhlén, I., Jarsjö, J., & Hambäck, P. A. (2023). Connecting wetland flooding
weaker for inundation of Polder 10 due to its shorter duration and the patterns to insect abundance using high-resolution inundation fre-
quency data. Wetlands, 43(6), 74.
effects of floodplain storage upstream in Polder A. Water levels in
Alaghmand, S., Beecham, S., Jolly, I. D., Holland, K. L., Woods, J. A., &
the Oder river showed the strongest influence on winter inundation in Hassanli, A. (2014). Modelling the impacts of river stage manipulation
the polders, with short-term variations more pronounced in Polder on a complex river-floodplain system in a semi-arid region. Environ-
10, compared with Polder A. Inter-annual variations in extent and lon- mental Modelling & Software, 59, 109–126.
Allen, R. G., Pereira, L. S., Raes, D., & Smith, M. (1998). Crop
gevity of inundation could be marked. Linkages between inundation
evapotranspiration-guidelines for computing crop water requirements-
and local precipitation were only detected in Polder A during summer. FAO irrigation and drainage paper 56. J Fao, Rome, 300(9), D05109.
These were transient and spatially localized compared with winter inun- Alsdorf, D., Han, S.-C., Bates, P., & Melack, J. (2010). Seasonal water stor-
dation from the Oder. Groundwater showed delayed response to inun- age on the Amazon floodplain measured from satellites. Remote Sens-
ing of Environment, 114(11), 2448–2456.
dation in polder A, likely reflecting induced recharge during flowing.
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZHENG ET AL. 15 of 16
Anders, I., Stagl, J., Auer, I., & Pavlik, D. (2014). Climate change in central movement and space use in a dynamic floodplain landscape. Science of
and eastern Europe. In Managing Protected Areas in Central and Eastern the Total Environment, 758, 144000.
Europe under Climate Change (pp. 17–30), Springer Nature. Grinsted, A., Moore, J. C., & Jevrejeva, S. (2004). Application of the cross
Arthington, A. H., & Balcombe, S. R. (2011). Extreme flow variability and wavelet transform and wavelet coherence to geophysical time series.
the ‘boom and bust'ecology of fish in arid-zone floodplain rivers: A Nonlinear Processes in Geophysics, 11(5/6), 561–566.
case history with implications for environmental flows, conservation Gumiero, B., Mant, J., Hein, T., Elso, J., & Boz, B. (2013). Linking the resto-
and management. Ecohydrology, 4(5), 708–720. ration of rivers and riparian zones/wetlands in Europe: Sharing knowl-
Arthington, A. H., Olden, J. D., Balcombe, S. R., & Thoms, M. C. (2010). edge through case studies. Ecological Engineering, 56, 36–50.
Multi-scale environmental factors explain fish losses and refuge quality Hamilton, S. K., Sippel, S. J., & Melack, J. M. (2002). Comparison of inunda-
in drying waterholes of Cooper Creek, an Australian arid-zone river. tion patterns among major south American floodplains. Journal of Geo-
Marine and Freshwater Research, 61(8), 842–856. physical Research: Atmospheres, 107(D20), LBA 5-1–LBA 5-14.
Bazi, Y., Bruzzone, L., & Melgani, F. (2007). Image thresholding based on Holgerson, M. A., Duarte, A., Hayes, M. P., Adams, M. J., Tyson, J. A.,
the EM algorithm and the generalized gaussian distribution. Pattern Douville, K. A., & Strecker, A. L. (2019). Floodplains provide important
Recognition, 40(2), 619–634. amphibian habitat despite multiple ecological threats. Ecosphere, 10(9),
Buchner, J., Yin, H., Frantz, D., Kuemmerle, T., Askerov, E., Bakuradze, T., e02853.
Bleyhl, B., Elizbarashvili, N., Komarova, A., & Lewińska, K. E. (2020). Huang, C., Chen, Y., Zhang, S., & Wu, J. (2018). Detecting, extracting, and
Land-cover change in the Caucasus Mountains since 1987 based on monitoring surface water from space using optical sensors: A review.
the topographic correction of multi-temporal Landsat composites. Reviews of Geophysics, 56(2), 333–360.
Remote Sensing of Environment, 248, 111967. IMGW-PIB. (2023). Dane publiczne. https://danepubliczne.imgw.pl/
Canisius, F., Brisco, B., Murnaghan, K., Van Der Kooij, M., & Keizer, E. apiinfo
(2019). SAR backscatter and InSAR coherence for monitoring wetland Jakubínský, J., Prokopova, M., Raška, P., Salvati, L., Bezak, N., Cudlín, O.,
extent, flood pulse and vegetation: A study of the Amazon lowland. Cudlín, P., Purkyt, J., Vezza, P., & Camporeale, C. (2021). Managing
Remote Sensing, 11(6), 720. floodplains using nature-based solutions to support multiple ecosys-
Cheng, V. Y., Saber, A., Arnillas, C. A., Javed, A., Richards, A., & tem functions and services. Wiley Interdisciplinary Reviews: Water, 8(5),
Arhonditsis, G. B. (2021). Effects of hydrological forcing on short-and e1545.
long-term water level fluctuations in Lake Huron-Michigan: A continu- Karim, F., Dutta, D., Marvanek, S., Petheram, C., Ticehurst, C., Lerat, J.,
ous wavelet analysis. Journal of Hydrology, 603, 127164. Kim, S., & Yang, A. (2015). Assessing the impacts of climate change
Cui, X., Guo, X., Wang, Y., Wang, X., Zhu, W., Shi, J., Lin, C., & Gao, X. and dams on floodplain inundation and wetland connectivity in the
(2019). Application of remote sensing to water environmental pro- wet–dry tropics of northern Australia. Journal of Hydrology, 522,
cesses under a changing climate. Journal of Hydrology, 574, 892–902. 80–94.
Czuba, J. A., David, S. R., Edmonds, D. A., & Ward, A. S. (2019). Dynamics Karim, F., Petheram, C., Marvanek, S., Ticehurst, C., Wallace, J., &
of surface-water connectivity in a low-gradient meandering river Hasan, M. (2016). Impact of climate change on floodplain inundation
floodplain. Water Resources Research, 55(3), 1849–1870. and hydrological connectivity between wetlands and rivers in a tropi-
Deutsher Wetterdienst (DWD). (2023). Daily station observations (tem- cal river catchment. Hydrological Processes, 30(10), 1574–1593.
perature, pressure, precipitation, sunshine duration, etc.) for Germany Krieger, S., Freij, N., Brazhe, A., Torrence, C., & Compo, G. P. (2023).
25 May 2023. https://cdc.dwd.de/portal/ PyCWT: Spectral analysis using wavelets in Python. https://pycwt.
Drayer, A. N., & Richter, S. C. (2016). Physical wetland characteristics influ- readthedocs.io/en/latest/getting_started.html#authors
ence amphibian community composition differently in constructed Labat, D. (2005). Recent advances in wavelet analyses: Part 1. A review of
wetlands and natural wetlands. Ecological Engineering, 93, 166–174. concepts. Journal of Hydrology, 314(1–4), 275–288.
Felipe-Lucia, M. R., Comín, F. A., & Bennett, E. M. (2014). Interactions Landesamt für Umwelt (LfU) Brandenburg. (2023). Auskunftsplattform
among ecosystem services across land uses in a floodplain agroecosys- Wasser. https://apw.brandenburg.de/
tem. Ecology and Society, 19(1): 20. Landesregierung Brandenburg. (2023). Geoportal Brandenburg.Accessed
Frantz, D. (2019). FORCE—Landsat+Sentinel-2 analysis ready data and date: 25 May 2023. https://geoportal.brandenburg.de/de/cms/portal/
beyond. Remote Sensing, 11(9), 1124. geodaten/themenkarten
Frantz, D., Haß, E., Uhl, A., Stoffels, J., & Hill, J. (2018). Improvement of Leblanc, M., Lemoalle, J., Bader, J.-C., Tweed, S., & Mofor, L. (2011). Ther-
the Fmask algorithm for Sentinel-2 images: Separating clouds from mal remote sensing of water under flooded vegetation: New observa-
bright surfaces based on parallax effects. Remote Sensing of Environ- tions of inundation patterns for the ‘Small'Lake Chad. Journal of
ment, 215, 471–481. Hydrology, 404(1–2), 87–98.
Frantz, D., Röder, A., Stellmes, M., & Hill, J. (2016). An operational radio- Li, Y., Zhang, Q., Lu, J., Yao, J., & Tan, Z. (2019). Assessing surface water–
metric landsat preprocessing framework for large-area time series groundwater interactions in a complex river-floodplain wetland-
applications. IEEE Transactions on Geoscience and Remote Sensing, isolated lake system. River Research and Applications, 35(1), 25–36.
54(7), 3928–3943. Liu, H., Yu, Y., Zhao, W., Guo, L., Liu, J., & Yang, Q. (2020). Inferring subsur-
Funk, A., Tschikof, M., Grüner, B., Böck, K., Hein, T., & Bondar-Kunze, E. face preferential flow features from a wavelet analysis of hydrological
(2021). Analysing the potential to restore the multi-functionality of signals in the Shale Hills catchment. Water Resources Research, 56(11),
floodplain systems by considering ecosystem service quality, quantity e2019WR026668.
and trade-offs. River Research and Applications, 37(2), 221–232. Liu, X., Zhang, Q., Li, Y., Tan, Z., & Werner, A. D. (2020). Satellite
Gaglio, M., Aschonitis, V., Pieretti, L., Santos, L., Gissi, E., Castaldelli, G., & image-based investigation of the seasonal variations in the hydrologi-
Fano, E. (2019). Modelling past, present and future ecosystem services cal connectivity of a large floodplain (Poyang Lake, China). Journal of
supply in a protected floodplain under land use and climate changes. Hydrology, 585, 124810.
Ecological Modelling, 403, 23–34. Ministerium für Landwirtschaft, Umwelt und Klimaschutz. (2023). Natio-
Globevnik, L., Januschke, K., Kail, J., Snoj, L., Manfrin, A., Azlak, M., nalpark Unteres Odertal—Offizielle Website der Nationalparkverwal-
Christiansen, T., & Birk, S. (2020). Preliminary assessment of river tung Unteres Odertal? https://www.nationalpark-unteres-odertal.eu/
floodplain condition in Europe. ETC/ICM Technical Report, 5:2020. Mohanty, M. P., & Simonovic, S. P. (2021). Changes in floodplain regimes
Goswami, V. R., Vasudev, D., Joshi, B., Hait, P., & Sharma, P. (2021). over Canada due to climate change impacts: Observations from
Coupled effects of climatic forcing and the human footprint on wildlife CMIP6 models. Science of the Total Environment, 792, 148323.
10991085, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hyp.15137 by UNIVERSIDADE ESTADUAL PAULISTA, Wiley Online Library on [01/04/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
16 of 16 ZHENG ET AL.
Munasinghe, D., Cohen, S., Huang, Y. F., Tsang, Y. P., Zhang, J., & Fang, Z. Smith, A., Tetzlaff, D., Kleine, L., Maneta, M., & Soulsby, C. (2021). Quanti-
(2018). Intercomparison of satellite remote sensing-based flood inun- fying the effects of land use and model scale on water partitioning
dation mapping techniques. JAWRA Journal of the American Water and water ages using tracer-aided ecohydrological models. Hydrology
Resources Association, 54(4), 834–846. and Earth System Sciences, 25(4), 2239–2259.
Nakamura, K., Tockner, K., & Amano, K. (2006). River and wetland restora- Smith, A., Tetzlaff, D., Maneta, M. P., & Soulsby, C. (2022). Visualizing
tion: Lessons from Japan. Bioscience, 56(5), 419–429. catchment-scale spatio-temporal dynamics of storage-flux-age interac-
Otsu, N. (1979). A threshold selection method from gray-level histograms. tions using a tracer-aided ecohydrological model.
IEEE Transactions on Systems, Man, Cybernetics, 9(1), 62–66. Tan, Z., Li, Y., Xu, X., Yao, J., & Zhang, Q. (2019). Mapping inundation dynam-
Posthumus, H., Rouquette, J., Morris, J., Gowing, D., & Hess, T. (2010). ics in a heterogeneous floodplain: Insights from integrating observations
A framework for the assessment of ecosystem goods and services; a and modeling approach. Journal of Hydrology, 572, 148–159.
case study on lowland floodplains in England. Ecological Economics, Tockner, K., Pusch, M., Borchardt, D., & Lorang, M. S. (2010). Multiple
69(7), 1510–1523. stressors in coupled river–floodplain ecosystems. Freshwater Biology,
Regier, P., Ward, N. D., Indivero, J., Wiese Moore, C., Norwood, M., & 55, 135–151.
Myers-Pigg, A. (2021). Biogeochemical control points of connectivity Torrence, C., & Compo, G. P. (1998). A practical guide to wavelet analysis.
between a tidal creek and its floodplain. Limnology and Oceanography Bulletin of the American Meteorological Society, 79(1), 61–78.
Letters, 6(3), 134–142. Townsend, P. A., & Walsh, S. J. (1998). Modeling floodplain inundation
Rosser, J. F., Leibovici, D. G., & Jackson, M. J. (2017). Rapid flood inunda- using an integrated GIS with radar and optical remote sensing. Geo-
tion mapping using social media, remote sensing and topographic data. morphology, 21(3–4), 295–312.
Natural Hazards, 87, 103–120. Tull, N., Passalacqua, P., Hassenruck-Gudipati, H. J., Rahman, S.,
Roy, D. P., Li, Z., & Zhang, H. K. (2017). Adjustment of Sentinel-2 multi- Wright, K., Hariharan, J., & Mohrig, D. (2022). Bidirectional river-
spectral instrument (MSI) red-edge band reflectance to nadir BRDF floodplain connectivity during combined pluvial-fluvial events. Water
adjusted reflectance (NBAR) and quantification of red-edge band Resources Research, 58(3), e2021WR030492.
BRDF effects. Remote Sensing, 9(12), 1325. Ward, J. V., & Stanford, J. (1995). Ecological connectivity in alluvial river
Rufin, P., Frantz, D., Yan, L., & Hostert, P. (2020). Operational coregistra- ecosystems and its disruption by flow regulation. Regulated Rivers:
tion of the sentinel-2A/B image archive using multitemporal landsat Research & Management, 11(1), 105–119.
spectral averages. IEEE Geoscience and Remote Sensing Letters, 18(4), Wohl, E. (2021). An integrative conceptualization of floodplain storage.
712–716. Reviews of Geophysics, 59(2), e2020RG000724.
Sajjad, A., Lu, J., Chen, X., Chisenga, C., Saleem, N., & Hassan, H. (2020). Zhu, Z., Wang, S., & Woodcock, C. E. (2015). Improvement and expansion
Operational monitoring and damage assessment of riverine flood- of the Fmask algorithm: Cloud, cloud shadow, and snow detection for
2014 in the lower Chenab plain, Punjab, Pakistan, using remote sens- Landsats 4–7, 8, and sentinel 2 images. Remote Sensing of Environment,
ing and GIS techniques. Remote Sensing, 12(4), 714. 159, 269–277.
Sall, I., Jarchow, C. J., Sigafus, B. H., Eby, L. A., Forzley, M. J., & Zhu, Z., & Woodcock, C. E. (2012). Object-based cloud and cloud shadow
Hossack, B. R. (2021). Estimating inundation of small waterbodies with detection in Landsat imagery. Remote Sensing of Environment, 118,
sub-pixel analysis of Landsat imagery: Long-term trends in surface 83–94.
water area and evaluation of common drought indices. Remote Sensing _
Zmihorski, M., Pärt, T., Gustafson, T., & Berg, Å. (2016). Effects of water
in Ecology and Conservation, 7(1), 109–124. level and grassland management on alpha and beta diversity of birds in
Sang, Y.-F. (2013). A review on the applications of wavelet transform in restored wetlands. Journal of Applied Ecology, 53(2), 587–595.
hydrology time series analysis. Atmospheric Research, 122, 8–15.
Schmugge, T. J., Kustas, W. P., Ritchie, J. C., Jackson, T. J., & Rango, A.
(2002). Remote sensing in hydrology. Advances in Water Resources, SUPPORTING INF ORMATION
25(8–12), 1367–1385. Additional supporting information can be found online in the Support-
Schultz, G. A., & Engman, E. T. (2012). Remote sensing in hydrology and
ing Information section at the end of this article.
water management. Springer Science & Business Media.
Schumann, G. J.-P., Muhlhausen, J., & Andreadis, K. (2019). Rapid mapping
of small-scale river-floodplain environments using UAV SfM supports
classical theory. Remote Sensing, 11(8), 982. How to cite this article: Zheng, H., Tetzlaff, D., Freymüller, J.,
Schwieder, M., Leitão, P. J., da Cunha Bustamante, M. M., Ferreira, L. G., Chmieleski, J., Okujeni, A., & Soulsby, C. (2024). Quantifying
Rabe, A., & Hostert, P. (2016). Mapping Brazilian savanna vegetation intra- and inter-annual dynamics of river-floodplain
gradients with Landsat time series. International Journal of Applied
connectivity and wetland inundation with remote sensing and
Earth Observation and Geoinformation, 52, 361–370.
Scott, D. T., Gomez-Velez, J. D., Jones, C. N., & Harvey, J. W. (2019). wavelet analysis. Hydrological Processes, 38(4), e15137.
Floodplain inundation spectrum across the United States. Nature Com- https://doi.org/10.1002/hyp.15137
munications, 10(1), 5194.