Temporal and Spatial Changes in Crop Patterns, Use of Inputs and Hydrological Alteration in the Case of Fogera Floodplain, Ethiopia

Abstract

More than half of the world’s population consumes rice. Recently, the area sown with modern rice varieties has expanded, and the use of chemical fertilizers and pesticides has increased in various countries. Wetland hydrology is also influenced by chemical and physical characteristics. Hence, this research focused on temporal and spatial changes in crop patterns, input usage, and hydrology in the Ethiopian Fogera floodplain, with the following objectives: (a) What are the spatial and temporal trends in crop production patterns? (b) What input changes have occurred to produce rice and other crops? (c) What hydrological changes have occurred in the area with intensification of production systems? Primary data were gathered through a questionnaire, focus group discussions, interviews, and field observations. Secondary data were obtained from Landsat imageries, the SWAT model, water flow measurements, and normalized difference vegetation index (NDVI). NDVI results indicated that the area cultivated for rice is increasing while the area of other crops is decreasing. Agricultural inputs are used in rice systems but were not used before the introduction of rice. Recession farming activities have also diminished wetland areas. Water flow showed a decrease, whereas Nitrogen and Phosphorus showed an increase with Pearson’s correlation values −0.069 and −0.072, respectively. Flow of water was negatively correlated with N and P water concentration, whereas N and P contents were positively correlated. In conclusion, growth of intensive rice systems has had negative environmental consequences on wetland ecology. Therefore, policies to regulate and manage wetland uses are recommended.

1. Introduction

Rice feeds more than half of the world’s population, with 148 million hectares of land devoted to the crop yielding 519 million tons in 1991 [1] and 503.17 million tons in 2020/21, out of which Ethiopia predicted to produce 91,000 metric tons [2]. For millennia, wetlands, especially floodplains with fertile soils and abundant water, have been used for rice production. Wetlands have undeniably aided the development of many significant cultures around the world, but wetlands being drained and reclaimed for agriculture has become more widespread in the last 20 years. Globally, more than half of the floodplains have been lost, with the agricultural conversion being one of the primary causes of the ongoing wetland losses [3]. In addition, modern rice production has increased the use of fertilizer where rice receives around 10% of the total nitrogen fertilizer applied globally [4]. Furthermore, around half of the Catla (common carp) fish died when the ammonia concentration reached 29.4 mg NH3–N/L [4]. Chemical pesticides used to control rice pests (microbial diseases, weeds, nematodes, snails, insects, and rodents) resulted in significantly higher grain yields, but its accumulation brings environmental hazards [5].

The hydrology of the wetland, which is affected by soil, biota, and build-up materials, influences and determines the wetland’s characteristics and processes such as amount of oxygen in the soil, as well as the supply of nutrients and toxicants [6,7]. Longer periods of inundation can result in longer periods of anaerobic conditions and limit the plants’ survival [8]. A study in Vietnamese floodplain rice production revealed that intensive rice systems have negative long-term effects on the environment [9]. In addition, a study in Ethiopia found that drainage and cultivation of floodplain wetlands induce extreme spatial and temporal variations in the wetland groundwater linked in part to soil structural and chemical changes which affect hydraulic conductivity [10].

A noticeable and important change in streamflow characteristics such as magnitude, timing, frequency, length, and rate of change of flow is referred to as hydrological alteration [11]. When the usual amount of water entering a wetland is decreased or raised, or when the time of saturation and inundation is prolonged, the ecosystem changes to an upland system [11]. Rice necessitates long-term hydrological processes. Rainfed wetland rice in Ethiopia is grown in areas where there is standing water during the growing season. Because of their high biomass production, wetlands in Ethiopia’s Lake Tana region such as the Fogera floodplain have historically provided valuable feedstock for livestock, later being cultivated for rice production [12,13]. Wetlands play an important ecological and hydrological role in environmental protection, as well as providing residents with a range of environmental and socioeconomic benefits [14]. Like the Fogera floodplain wetlands, agriculture faces a major challenge in developing strategies to maintain production while reducing environmental impacts. The trade-off between crop production (such as rice) and water quality ecosystems services is one of the most serious issues facing agriculture, as well as interest in achieving win–win outcomes through ecosystem service management [15].

Achieving a global transition to sustainable farming would necessitate a concerted effort to respond to our population’s rapid growth without jeopardizing the environment’s integrity [16]. Reducing trade-offs between crop production, water quality, and habitats increase the likelihood of win–win outcomes [15]. Different studies in the Fogera floodplain such as impact of fertilizer on rainfed rice [17], profitability and market chain of rice [18], evaluation of hand-weeding on rice yield [19], and response of rice on different levels of fertilizer in determining optimal application [20] have overlooked environmental issues and trade-offs. More research is required to determine how trade-offs can be mitigated and whether the likelihood of win–win outcomes can be increased in agricultural landscape management to encourage win–win strategies [15].

Therefore, this research aims to identify temporal and spatial shifts in rice-cropping patterns, input levels, and linkages with hydrological alteration in the Fogera floodplain. The study first identified the spatial and temporal trends in crop production patterns in the area, then estimated inputs used in the past and present to produce rice and other crops, finally trying to link it to the hydrological alteration in both quantity and quality in the area.

2. Methods and Materials

2.1. Description of Study Area

Fogera Woreda is situated between latitudes 11°46′ and 11°59′ north and longitudes 37°33′ and 37°52′ east, at an altitude of 1774 to 2410 m above mean sea level and a mean annual rainfall of 1216 mm, with long rains from June to September. Lake Tana borders the district on the west (Figure 1). Fogera Woreda is one of the surplus crop-producing areas, with a high potential for rice production [21]. The overflow from Lake Tana and the two surrounding major rivers, Ribb and Gumara, are channeled through the Fogera floodplain. The Ribb River has a dam under construction since 2010 and although not operational it is commissioned to store water in 2020. In the study area, rice is planted on lower slopes where the water table rises to the surface for a long time during the cropping season. Rice is also irrigated with water diverted from streams in a drainage system’s upper reaches [13]. The Fogera floodplain has permanent wetlands which are assigned as biosphere reserve areas having roosting sites for migratory birds and spawning grounds for migratory fish species from Lake Tana [22].

2.2. Sampling and Data Collection

Questionnaires, interviews, focus group discussions (FGDs), and field observations were used to gather data. Sample Kebeles were chosen with purposive sampling among 27 Kebeles in Fogera Woreda which are rice producers and close to the lake and floodplain wetlands. Interviewees were chosen with random sampling together with development agents going into villages. The rice-producing 5 Kebeles closest to the Lake Tana Fogera floodplain were first identified. The analysis included all five Kebeles, with sample households drawn proportionally from each kebele (385 in total). The total sample household number for interview was determined by the following formula [23]:

$$\mathrm{n}=\frac{\mathrm{N}}{1+\mathrm{N}{(\mathrm{e})}^2}$$

(1)

where, n = Sample; N = Population; e = Error term (5%)

On the other hand, focus groups were chosen based on their kebele’s representatives.

Quantitative data were gathered from both primary and secondary sources such as Landsat images for land use land cover mapping, and ground truth data were collected for land use classification signatures (

Table 1. Data type and sources.

S.N.	Data Type	Spatial Resolution	Source
1	Landsat 8 image	30 m	USGS Landsat images
2	DEM SRTM, 2000	30 m	NASA/USGS/JPL-Caltech
3	NDVI	250 m	MODIS from: https://doi.org/10.5067/ASTER/ASTGTM.003 (accessed on 16 September 2020)
4	Rib River flow, 1981–2005	discharge in m3s−1	MoWIE, Ethiopia
5	Precipitation, 1981-now	0.25°	CHIRPS 2.0 Africa
6	Temperature, Minimum/Maximum, 1981-now	0.25°	ERA5 Africa
7	Soil	1:250,000	MoWIE, BCEOM. [24]

). Vegetation was sampled for biodiversity and evenness analysis using standard sampling every 100 m distance along the longest cross-section of the study area and taking 1 m² plot for identifying types and counting number of species in 3 × 15 plots in three Kebeles. Climate data such as rainfall (CHIRPS) and temperature (ERA5), and digital elevation model (USGS website), and soil map shapefiles [24] were collected as input for hydrological analysis of floodplain rivers including Ribb River using SWAT model. In addition, Ribb observed flow data (MoWIE) for validating simulated flow data, modelled and secondary data of Nitrogen (N) and Phosphorus (P) from Abebe et al. [25] and Negash et al. [26] for water quality analysis and normalized difference vegetation index (NDVI) data (MODIS from https://doi.org/10.5067/ASTER/ASTGTM.003) (accessed on 16 September 2020) for vegetation cover trend analysis were also collected. To verify the NDVI classes (range) with the existing cover of rice field, water body and dense vegetation, ground truth of 30 points for rice field, 30 points for dense vegetation and 30 for water and others (emergent aquatic vegetation) were collected from June to September (wet season) 2018 (Figure 2).

2.3. Methods of Data Analysis

Both qualitative and quantitative data were analyzed using percentage, mean, and correlation in Excel. Pearson’s correlations were used for water flow, Nitrogen (N) and Phosphorous (P) that indicate water pollution status. Indicators of Hydrological Alteration (IHA) software was used to figure out environmental flow components and analyze flow alterations of the Ribb River which floods to the Fogera floodplain [27]. Image processing and analysis were performed using ArcGIS10.4 software. SWAT2012 for flow and nutrient simulation and SWATCUP for calibration and uncertainty analysis were also used. Ribb observed flow between 1981 to 1999 and 2000 to 2004 were used for calibration and validation, respectively.

NDVI data for the years 2000, 2009, and 2018 were generated using Google Earth engine, cloud computing platform, to see rice expansion in the area. The NDVI was used to identify the rice, dense vegetation and water bodies, and other small grass areas within a wetland classified in ArcGIS 10.4 environment based on signatures suitable threshold indices values for rice 0.4–0.71 (Average = 0.66 and STDEV = 0.03), dense vegetation > 0.71 (Average = 0.71 and STDEV = 0.01), water and others < 0.4 (Average = −0.04 and STDEV = 0.08) based on Tucker, C.J. and P. Sellers [28] and the ground truth collected during the study period from June to September (wet season) 2018.

Shannon Diversity Index (H’) was also used to compute the diversity and composition of herbaceous and grass species of the wetlands in the Fogera floodplain [29]. This index takes into consideration species composition and evenness within the given area calculated using Equations (2) and (3) as follows:

$$\mathrm{H}\prime =-{{\displaystyle \sum }}_{\mathrm{i}=1}^{\mathrm{s}}\mathrm{Pi} (\mathrm{In}\left(\mathrm{pi}\right),\to \mathrm{pi}=\frac{\mathrm{ni}}{\mathrm{N}}$$

(2)

$$\mathrm{Evenness} (\mathrm{J}): \mathrm{J}=\frac{\mathrm{H}}{\mathrm{Hmax}}$$

(3)

where ni = number of individuals of species “i”, N = total number of individuals of all species, pi = relative abundance of species “i”, S = total number of species, and H′ = the Shannon Diversity Index.

3. Results and Discussion

3.1. Spatial and Temporal Trends in Crop Production Pattern in the Area

Spatial and temporal patterns in the area’s crop production pattern based on three image classes showed that the rice crop increased at a rate of 1198 NDVI with R² of 0.8 whereas dense vegetation, water and other small vegetation showed decreasing trends at a rate of −251 and −948 NDVI with R² 0.5 in the year 2000, 2009 and 2018, respectively (Figure 3 and Table S4).

Furthermore, the spatial analysis of the NDVI values in 2000 for dense vegetation, rice, water and others indicated 2691 ha, 11,127 ha, and 3328.0 ha area coverages, respectively (Figure 4a), whereas the coverage in 2009 for dense vegetation was 120 ha, for rice 13,469 ha, and water and others 3556 ha (Figure 4b). In the latest analysis in 2018, the area coverage for dense vegetation, rice and water, and others was 796 ha, 13,523 ha, and 2826 ha, respectively (Figure 4c). Hence, based in the spatial analysis, the area coverage for rice showed an increase while dense vegetation and water and others showed a decline. Overall, rice area increased by 2397 ha whereas “water and others” decreased by −502 ha. This is in line with other similar studies in Asia on rice-cropping system changes on floodplains [30,31].

Accordingly, as indicated in

Table 2. Expansion of farmland for rice production.

Sample Kebeles	No. of Sample Respondents	Their Farmland				Size of Their Farmland in timad
							(1 timad = 0.25 ha)
		Yes	No	Before Rice	%	After Rice	%	Increased	%
Kidest Hana	69	68	1	124	19.5	221	19.4	97	19.3
Shina	81	81	0	130	20.5	234	20.5	104	20.6
Shaga	57	57	0	95	15	175	15.4	80	15.9
Wagetera	91	91	0	146	23	261	22.9	115	22.8
Nabega	87	87	0	140	22	248	21.8	108	21.4
Gross cropped area	385	384	1	635	100	1139	100	504	100

, the sampled respondent mentioned that the size of their farmland before the introduction of rice was 160 ha but after the introduction of rice this number flipped to 285 ha, or an increase of 125 ha (Table 2). A study of rice commercialization in the Fogera floodplain revealed similar conditions [32].

Respondents were asked about their previous production style before the introduction of rice. They indicated that they produced different crops in the study areas, including small amounts of teff (Eragrostis tef), maize (Zea mays L.), noug (Guizotia abyssinica), finger millet (Eleusine coracana), chickpea (Cicerarietinum), lentil (Lens culinaris), grass pea (Lathyrus sativus), green pepper (Capsicum spp.) and barely (Hordeum Vulgare), and devoted a larger amount of land for animals. They also stated these minor crops were produced far away from the wetlands in the study site.

Most respondents reported being engaged in rice production during the summer season while in the winter season, they focused more on horticultural crops such as onion (Allium cepa), garlic (Allium sativum), and tomatoes (Solanum Lycopersicum). These were grown using small-scale irrigation.

Regarding the area cultivated for the wetlands of the Fogera floodplain, the two-year data showed that the cultivated area coverage of the rice crop was 245 ha and 285 ha in the years 2014 and 2015, respectively (

Table 3. The rice cultivated area for the year 2014 and 2015.

Name of Sample Kebele	No. of Sample Respondent	2014	2015	Increment in ha
		Total Cultivated Land in ha	Total Cultivated Land in ha
Kidest Hana	69	46	55.25	9.25
Shaga	81	52	58.5	6.5
Shina	57	36	43.75	7.75
Wagetera	91	56.25	65.25	9
Nabega	87	55	62	7

and Figure 5). In Figure 5 the land-use coverage for rice and other land use for the years 2013, 2015, and 2016 is also mapped, and the result revealed the coverage for rice has shown an increment.

About 89% of the sampled household respondents reported that the types of cropping pattern practiced are mono cropping whereas about 11% of sampled respondents said that they practiced crop rotation (

Table 4. Types of cropping pattern.

No. of Kebeles	No Sample Respondents	Mono Cropping	Crop Rotation	Intercropping	Other
Kidest Hana	69	62	7	0	0
Shaga	81	69	12	0	0
Shina	57	51	6	0	0
Wagetera	91	81	9	0	0
Nabega	87	79	8	0	0

). Those who practiced crop rotation explained they used rotation because their farmland is far away from the floodplain which is located on the uplands.

3.2. Agricultural Inputs Application

The entire respondent group mentioned that the methods to maximize yield are applying agricultural inputs such as chemical fertilizer, improved seed, insecticide, and herbicide, resulting in expansion of rice farming through encroaching wetlands and communal grazing lands (

Table 5. Inputs used for the production of crops.

If There Is an Increase in Production, What Are the Causes
No. of Kebeles	No. of Respondents	Fertilizers	Manure	Improved Seed	Fallowing	Crop Rotation	Using Insecticides	Using Herbicides
Kidest Hana	69	69	0	69	0	0	69	69
Shaga	81	81	0	81	0	0	81	81
Shina	57	57	0	57	0	0	57	57
Wagetera	91	91	0	91	0	0	91	91
Nabega	87	87	0	87	0	0	87	87

). Insecticide has been used for most crops of chickpea (Cicer arietinum) and grass pea (Lathyrus sativus) which are relayed after rice (Oryza sativa). Not all respondents used agro-chemical inputs, but rather they used manure for the production of maize (Zea mays L.) and green pepper (Capsicum spp.), whereas manuring, fallowing and crop rotation were not applicable for rice in the area (Table 5).

About 45.5% of respondents applied fertilizer, and 23% used improved seed. In addition, about 95% of respondents used pesticides, and herbicides were used by about 64% of the respondents (Figure 6). This spraying of pesticides kills insects and those insects are eaten by birds (Figure 7). Therefore, the birds could be affected. This is in line with other studies that confirmed ecosystem functioning effects of pesticides application on rice [33,34].

About 74% of respondents recognized negative environmental effects of using agricultural inputs, including reduced numbers of birds (73%), bee colonies (96%), fish (73%) and other animals such as macroinvertebrates, reptiles, and amphibians (73%) in the wetlands of the study area (

Table 6. Negative impacts of application of agricultural inputs.

Kebele	No. of Respondents	Water Pollution	Reduction of Birds	Reduction of Bees	Reduction of Fish	Reduction of Other Animals
Kidest Hana	69	53	47	55	50	48
Shaga	81	64	63	80	63	63
Shina	57	41	41	57	41	41
Wagetera	91	66	66	91	66	66
Nabega	87	62	62	87	62	62

and Figure 7).

About 22% of the respondent farmers said that they practice recession farming, i.e., they plant along the shore as water recedes (Figure 8 and

Table 7. Availability of recession farming practices.

No. of Sample Kebeles		Recession Farming
	No. of Sample Respondents	Yes	No
Kidest Hana	69	11	58
Shaga	81	16	65
Shina	57	8	49
Wagetera	91	25	66
Nabega	87	23	64

). After rice is harvested, they tend to plant maize during the dry season. The majority who did not practice recession farming did so because their farms were too far away from receding bodies of water.

All respondents recognized effects of the practice on ecosystem services such as papyrus and grass production, bird populations, and the amount and quality of water, including increased siltation and turbidity. However, some farmers were trying to maximize their productivity through the expansion of recession farming at the expense of wetlands and Lake Tana in the study area. The interviewed farmers and FGD participants also recognized changes they had observed in wetland shrinkage and water resource depletion in their kebele. In summary, they observed a reduction in water levels and fish populations, lower papyrus and grass production, fewer birds and hippopotami, and a change in insect populations. Respondents were asked to rank changes in wetland resources. The respondents’ problem ranking results indicate that papyrus production was most affected, followed by fish, grass, amount of water, birds, hippopotamus, and sand. Reductions are presented in

Table 8. Rank of wetland and Lake Tana resources.

Description of Wetland Resources	Resource Rank	Papyrus	Fish	Sand	Birds	Grass	Amount of Water	Hippopotamus
Wetland resources highly reduced is ranked from the smallest number 1 to the large number 7, e.g., if 1 is selected, the resource is highly reduced.	First = 1	385	0	0	0	0	0	0
	Second = 2	0	135	5	7	191	47	0
	Third = 3	0	43	63	10	95	171	3
	Fourth = 4	0	178	4	117	14	60	12
	Fifth = 5	0	27	30	189	15	71	53
	Sixth = 6	0	1	38	32	28	13	273
	Seventh = 7	0	1	245	30	42	23	44
In this table, the data are the weighted values for each resource responded by sampled respondents, e.g., 385 sampled respondents give rank for each resource.	1	385	0	0	0	0	0	0
	2	0	270	10	14	382	94	0
	3	0	129	189	30	285	513	9
	4	0	712	16	468	56	240	48
	5	0	135	150	945	75	355	265
Rank value ∗ each value given by respondents, e.g., 1 ∗ 385 = 385 continuous like this.	6	0	6	228	192	168	78	1638
	7	0	7	1715	210	294	161	308
Weighted Sum Total		385	1259	2308	1859	1260	1441	2268
	Rank	1st	2nd	7th	5th	3rd	4th	6th

, and ranked 2nd, 3rd, 4th, 5th, 6th and 7th, respectively.

Respondents were unanimous in their strong desire to conserve the wetlands of Fogera and Lake Tana. Perceived benefits included preservation of papyrus, fish, birds, grass, clean water, reduced flooding, increased soil fertility, wildlife habitat, and vegetation, and reduced land and water degradation. Recession farming is considered as a disadvantage for certain users because respondents linked it to increasing number of hippopotamus and birds that damage their crops, and there would be no free fish harvest in the wetlands and the lake (Figure 9). Dejen, Anteneh [35] mentioned related effects of recession agriculture in the Fogera floodplain.

3.3. Biodiversity in the Fogera Floodplain

The diversity of herbaceous vegetations in the three Kebeles of the Fogera floodplain was found to be 2.15, 2.27 and 2.44 for Nabega, Shaga and Wagetera Kebeles respectively (

Table 9. Biodiversity of herbaceous vegetations in three Kebeles of Fogera floodplain.

	Shannan Wiener Index (H′)				Evenness
Kebele	Sample A	Sample B	Sample C	Average
Nabega	1.91	2.31	2.23	2.15	0.80
Shaga	2.28	2.27	2.26	2.27	0.86
Wagetera	2.46	2.43	2.44	2.44	0.93

). The overall average for the floodplain was 2.30. The evenness for the area was 0.80, 0.86, and 0.93 for Nabega, Shaga and Wagetera Kebeles, respectively. Most of the species identified in the area were found to be weeds which are encroaching the area because of cultivation (see Supplementary Materials). Hence, the evenness exhibited close to one, having many of those diverse weeds, whereas the maximum diversity index for the area was 2.71 and yet not reached by either of the Kebeles.

The increasing trend in rice cultivated land is supported by the fact that wetlands are characterized by fertile soil and sufficient water, which resulted in more land for rice production [36]. Farmers in the study area tend to grow rice because of its high yield and high market price relative to other food crops [37]. The introduction of rice has changed the production system, especially in the wetlands of the Fogera floodplain, from livestock-dominated to rice-dominated. Mono cropping is the most common cropping pattern recorded in the studied wetland areas, in line with results of Tilahun-Tadesse and Nigussie-Dechassa [17]. This has an effect on the biodiversity in the Fogera floodplain as studies indicated [25].

Before the introduction of rice, manure was used instead of chemical fertilizer to grow maize (Zea mays L.) and green pepper (Capsicum spp.), being that the soil was extremely fertile [38]. According to FGD participants, soil fertility depletion and addition of sand and other unfertile soil particles from upland erosion to the study area’s wetland decreased soil fertility and the amount of water in the wetland. This resulted in the use of agricultural inputs (chemical fertilizer, improved crop, insecticide, and herbicide) [38]. These have contributed to the biodiversity loss of the Fogera floodplain wetlands, as previous studies confirmed [26,39]. For example, rice cultivation allowed for a landscape rich in diversity, particularly in macroinvertebrates, fish, and waterfowl until the early twentieth century in rice-producing Asian countries, but the intensification of agricultural practices has changed their ecological character significantly [40,41].

After the start of chemical input applications, according to respondents, there is recognized water contamination, extinction of wetland vegetations such as papyrus, and a decrease in the number of birds, bee colonies, fish and other animals such as macroinvertebrates, reptiles and amphibians in the Fogera floodplain wetlands. This is in line with recent quantitative studies [25,26].

This had a significant negative impact on the attractiveness of the local tourism-based economy [42]. Earlier studies revealed that traditional rice systems were sufficiently multifaceted ecosystems that did not involve the use of chemical fertilizers but had optimum production for longer periods of time [43].

3.4. Hydrological Alteration in the Fogera Floodplain Wetlands

The performance of the SWAT2012 model was found to be R² 0.61 and NSE 0.60 for calibration and R² 0.51 and NSE of 0.50 for validation.

3.4.1. Monthly Flow Trends

The monthly flow analysis indicated that lower flows occur from December to June, the lowest being in April, with a median flow of 0.01 m³ s⁻¹ and higher flows occurring from July to November, the highest being in August, with a median flow of 94.5 m³ s⁻¹ (Figure 10). A high coefficient of variation (COV) of an average of 1.0 was found from November to May during the dry season, and a low COV of an average of 0.5 was observed from June to October during the wet season. Dry monthly flows have an increasing trend in the beginning months, and later months showed a decreasing trend, but rainy season months showed an overall decrease (Figure 11). Flows in the dry season in March and April increased in time likely due to the flow releases from the upstream Ribb dam during construction time, and May to September (rainy season) decreased likely due to that the dam holds water. This is contrary to the Gumara River, with no dam, which also floods the Fogera floodplain [44].

3.4.2. Water Chemistry

The summary statistics for water flow, Nitrogen and Phosphorous revealed that the Pearson’s correlation of water flow with N and P was −0.069 and −0.072, respectively. These have negative relationships, whereas the correlation between N and P has a positive correlation with the value of 0.242 (

Table 10. Summary statistics for correlation of flow, N and P.

Variable	Observations	Obs. with Missing Data	Obs. without Missing Data	Minimum	Maximum	Mean	Std. Deviation
Flow m3/s	35	0	35	18.670	42.280	31.276	6.134
N m3/s	35	0	35	0.040	2.830	0.911	0.651
P m3/s	35	0	35	0.170	0.360	0.261	0.047

). Hence, the water quality deteriorates with flow decrease.

Seasonal and annual variations in the hydrological functions of the study wetlands were observed. In line with this, floodplain wetland communities have also recognized the decline in water level in the wetlands, where there is loss of fish, depletion of papyrus and other wetland grass species, and disappearance of the hippopotamus population as well as migratory birds. In aquatic ecosystems, hydrological regimes create biotic diversity, and hydrological variation is recognized as a primary driving force [45]. When humans alter the flow regime, the normal cycle of hydrologic variation and ecosystem dynamics is disturbed [39,46,47].

In addition, the effects of recession farming on wetland and lake resources have been demonstrated with change in the color of water, increased siltation and turbidity of the water, and decreased flow quality (

Table 11. Water quality status of the Welala and Shesher wetlands in the Fogera floodplain from different studies. Note: Shesher 01 and 02 are two sampling points in the wetland where a river comes in and out.

Parameters	Units	Negash et al. [26]		Abebe et al. [25]		Simulated
		Shesher	Welala	Shesher 01	Shesher 02	Wetland Areas
		(2009 to 2010)	(2009 to 2010)	2020	2020	2009	2010	2011	2012	2020
Nitrate	mg/L	1.18	1.14	0.349	0.176	2.67	1.23	0.36	1.61	0.32
Phosphate	mg/L	0.54	2	0.443	1.371	0.36	0.2	0.24	0.34	0.05

). Recession farming has had a detrimental effect on the fauna, flora, and hydrology of the study site. New lands were developed at the land/swamp interface as a result of the recession and the draining of wetlands, resulting in major destruction of the wetland environment. Several studies have found a gradual shift in farming system enterprises from terrestrial to wetland-based development by households close to wetlands [48].

Aside from better land-use practices, livestock overgrazing and macrophyte overharvesting in the Fogera floodplain play a synergistic role in disrupting the area’s hydrological cycle and microclimate regulation. As wetland macrophyte vegetation disintegrates due to over-exploitation and loss of wetland vegetation, pollutants and nutrients are transported directly into the lake [49]. The euphotic zone in littoral area lakes gets shallow due to increased turbidity caused by the destruction of the buffer zone vegetation by cultivation, suggesting a drastic drop in water quality.

4. Conclusions

There is spatial increase in rice-cropping and temporal changes in crop production in the Fogera floodplain. The introduction and expansion of rice farming in the study area were at the expense of draining the wetlands and grazing land losses. In addition, the agricultural input applications on rice fields have also increased. Dry monthly flows have an increasing trend in the beginning months of the year, and later months showed a decreasing trend, but rainy season months showed an overall decrease. The diversity of herbaceous vegetation in the three Kebeles of the Fogera floodplain was found to be lower than the overall average index. Changes in wetland hydrology can affect soil chemistry as well as the plant and animal population in wetlands. Altering the natural amount of water entering a wetland or the cycle of saturation and inundation will cause the deterioration of biodiversity and loss of ecosystem services. The application of inputs caused water pollution and soil degradation. These have an effect on wetlands that play an important role in water purification, nutrient preservation, flood prevention, and erosion control. Generally, intensification of farming practices has an impact on the biodiversity of fauna and flora of the floodplain area. Therefore, the Fogera floodplain needs to be regulated based on a wetland management plan and the already established proclamations for biodiversity conservation areas in this region.