Signatures of human impact: size distributions and spatial organization of wetlands in the Prairie Pothole landscape

Abstract

More than 50% of global wetland area has been lost over the last 200 years, resulting in losses of habitat and species diversity as well as decreased hydrologic and biogeochemical functionality. Recognition of the magnitude of wetland loss as well as the wide variety of ecosystem services provided by wetlands has in recent decades led to an increased focus on wetland restoration. Restoration activities, however, often proceed in an ad hoc manner, with a focus on maximizing the total restored area rather than on other spatial attributes of the wetland network, which are less well understood. In this study, we have addressed the question of how human activities have altered the size distribution and spatial organization of wetlands over the Prairie Pothole Region of the Des Moines Lobe using high- resolution LIDAR data. Our results show that as well as the generally accepted 90% loss of depressional wetland area, there has been a preferential loss of smaller wetlands, with a marked alteration of the historical power-law relationship observed between wetland size and frequency and a resulting homogenization of the wetland size distribution. In addition, our results show significant decreases in perimeter-to-area ratios, increased mean distances between wetlands, particularly between smaller wetlands, and a reduced likelihood that current wetlands will, be located in upland areas. Such patterns of loss can lead to disproportionate losses of ecosystem services, as smaller wetlands with larger perimeter-to- area ratios have been found to provide higher rates of biogeochemical processing and groundwater recharge, while increased mean distances between wetlands hinder species migration and thus negatively impact biodiversity. These results suggest the need to gear restoration efforts toward understanding and recreating the size distribution and spatial organization of historical wetlands, rather than focusing primarily on an increase in overall area.

Biogeochemical hotspots: Role of small water bodies in landscape nutrient processing

“…In a previous study, Van Meter and Basu (2015) quantified the historical and current size-frequency functions of depressional wetlands in the southernmost lobe of the PPR (Iowa), and found that, in addition to an overall loss of wetlands across the size classes, there has been a preferential loss of smaller wetlands in upland locations and may allude to a preferential loss of biogeochemical processes in the landscape Indeed, small, geographically isolated wetlands (GIWs) are considered to be at particular risk of drainage due to both a lack of legislative protections and general patterns of land development, i.e. the smallest wetlands are the easiest to drain, and their importance in landscape functionality (whether hydrologically, biogeochemically or ecologically) is easily underestimated [Van Meter and Basu, 2015]. GIWs are defined as wetland systems that do not have an apparent surface connection to a nearby water body (such as a river or lake) and thus are completely surrounded by uplands [Leibowitz, 2015]; however, it should be noted that many GIWs are connected through subsurface pathways or are seasonally connected for a portion of the year and form wetland complexes and thus are not 'visibly' connected or deemed to be useful [Leibowitz and Vining, 2003;Johnson et al, 2010].…”

“…Global datasets such as the HydroLAKES database which characterizes the area, volume, and residence times of lakes are truncated at waterbodies less than 10 ha in area [Messager et al, 2016] -thus would not be useful when attempting to quantify the truly smaller systems. The Des Moines Lobe landscape is part of the Prairie Pothole Region, which has numerous depressional features with surface areas ranging from 100 m 2 to 5x10 4 m 2 [Van Meter and Basu, 2015]. Analyses by Van Meter and showed that there is a power law relationship between the number of depressional wetlands in the lobe and their surface areas ( Figure 5; N = 2x10 10 x SA -1.67 ; p <0.001; r 2 = 0.99).…”

“…Mitsch et al [2005], for example, used a simple empirical model to provide an estimate of the extent of new wetland creation necessary in the Mississippi River Basin to remove 40% of nitrogen loading to the Gulf of Mexico. But as pointed out by many authors [Semlitsch and Bodie, 1998;Downing, 2010;Ghermandi et al, 2010]: the functionality of a wetland is not uniform across systems, and thus wetland restoration must focus not only on goals related to total wetland area, but also to the type, landscape position, and morphometry of the wetlands being restored [Van Meter and Basu, 2015].…”

“…For example, the Prairie Pothole Region (PPR), an approximately 700,000 km 2 area across the central US and Canada, has lost an estimated 65% of wetlands primarily due to drainage of wetlands for cropland from the 1800s to the mid-1980s; southwestern Ontario has similarly lost 72% of wetlands since pre-settlement to urban and agricultural expansion [Ducks Unlimited Canada, 2010]. In a previous study, Van Meter and Basu (2015) quantified the historical and current size-frequency functions of depressional wetlands in the southernmost lobe of the PPR (Iowa), and found that, in addition to an overall loss of wetlands across the size classes, there has been a preferential loss of smaller wetlands in upland locations and may allude to a preferential loss of biogeochemical processes in the landscape Indeed, small, geographically isolated wetlands (GIWs) are considered to be at particular risk of drainage due to both a lack of legislative protections and general patterns of land development, i.e. the smallest wetlands are the easiest to drain, and their importance in landscape functionality (whether hydrologically, biogeochemically or ecologically) is easily underestimated [Van Meter and Basu, 2015].…”

“…The N(SA) relationship was based on a distribution of wetlands found in the prairie pothole region in Iowa. A previous study had used high-resolution (1 m resolution) LIDAR data to estimate the number of water bodies as a function of surface area in the Des Moines Lobe in Iowa [Van Meter and Basu, 2015]. This size-frequency distribution was chosen as there have been few studies that have quantified the distributions of small wetlands at such a fine spatial resolution within landscape scales.…”